model_averaging_Staianet.ipynb

{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[{"file_id":"1zgofglyd5yRtRWBi7mwtw5E3cC37946s","timestamp":1670775803356},{"file_id":"1KOehnwr3yqWAKwtGlWFqyxV_JRSRto2j","timestamp":1670764770790}]},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"},"accelerator":"GPU","gpuClass":"standard"},"cells":[{"cell_type":"code","execution_count":null,"metadata":{"id":"va-QmMMoqrL7","executionInfo":{"status":"ok","timestamp":1671208364512,"user_tz":-60,"elapsed":935,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"colab":{"base_uri":"https://localhost:8080/"},"outputId":"23382486-d4be-43ba-ae85-7fa5f2567cd8"},"outputs":[{"output_type":"stream","name":"stdout","text":["2.9.2\n"]}],"source":["from google.colab import drive\n","import os \n","import numpy as np \n","import pandas as pd \n","import seaborn as sns \n","import matplotlib.pyplot as plt\n","import tensorflow as tf \n","import tensorflow.keras as tfk \n","import tensorflow.keras.layers as tfkl \n","plt.rc('font', size=16) \n","import random\n","from sklearn.metrics import accuracy_score, f1_score, precision_score, recall_score\n","from sklearn import model_selection\n","from sklearn.metrics import confusion_matrix\n","from sklearn.utils import class_weight\n","from sklearn.preprocessing import MinMaxScaler\n","import warnings\n","import logging\n","from sklearn import preprocessing \n","\n","\n","tfk = tf.keras\n","tfkl = tf.keras.layers\n","print(tf.__version__)"]},{"cell_type":"code","source":["# Random seed for reproducibility\n","seed = 89\n","random.seed(seed)\n","os.environ['PYTHONHASHSEED'] = str(seed)\n","np.random.seed(seed)\n","tf.random.set_seed(seed)\n","tf.compat.v1.set_random_seed(seed)"],"metadata":{"id":"AtNSmv7Yu6j_"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["from google.colab import drive\n","drive.mount('/content/drive')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"yA8mp8WzX5sj","executionInfo":{"status":"ok","timestamp":1671208367152,"user_tz":-60,"elapsed":2653,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"7b211edc-91aa-4013-b5e7-9cc9ec43da5c"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"]}]},{"cell_type":"code","source":["%cd /content/drive/MyDrive/ann_challenge2/annHomework2"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"lD_mjbsGX72H","executionInfo":{"status":"ok","timestamp":1671208367153,"user_tz":-60,"elapsed":35,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"4fdeedf8-85e6-4be1-cc3f-22093be33f5e"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["/content/drive/.shortcut-targets-by-id/1MSnaIk1WNF6pWaDixVGVWqHvx61VG3_y/annHomework2\n"]}]},{"cell_type":"code","source":["x_train = np.load(r'x_train.npy')\n","y_train = np.load(r'y_train.npy')"],"metadata":{"id":"_7sucj6TrC-v"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["x_train.shape, y_train.shape "],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"0qxzUpvArntW","executionInfo":{"status":"ok","timestamp":1671208367158,"user_tz":-60,"elapsed":32,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"76dc13b2-64d3-4715-bb70-81f21e47e71f"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["((2429, 36, 6), (2429,))"]},"metadata":{},"execution_count":34}]},{"cell_type":"code","metadata":{"id":"rAprzP2bkm4S","colab":{"base_uri":"https://localhost:8080/","height":291},"executionInfo":{"status":"ok","timestamp":1671208367161,"user_tz":-60,"elapsed":29,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"f83f7741-e0ee-40b0-a6cf-277322d40bb7"},"source":["# plot labels distribution \n","sns.countplot( x = y_train) \n","plt.title('Dataset Labels Distribution')\n","plt.show()"],"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 432x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"code","source":["labels = np.unique(np.fromiter([i for i in y_train], np.int32) )\n","class_weights = class_weight.compute_class_weight(class_weight='balanced', classes=labels, y=y_train) "],"metadata":{"id":"QKDkjdJ7754A"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["# interpolation \n","from scipy.interpolate import interp1d\n","x_new = [] \n","for el in x_train: \n","    interp = interp1d(x=np.arange(36), y=el, axis=0, kind='cubic')\n","    el = interp(np.arange(0, 35, .3))\n","    x_new.append(el)\n","x_new = np.array(x_new) \n","print(x_new.shape) \n","x_train = x_new\n","y_train = tfk.utils.to_categorical(y_train) "],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"ulb_XXpY-qDz","executionInfo":{"status":"ok","timestamp":1671208368209,"user_tz":-60,"elapsed":1072,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"97c7f28c-d321-4819-fe26-fde16ab73f00"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["(2429, 117, 6)\n"]}]},{"cell_type":"code","source":["train_data = np.concatenate([x for x in x_train]) \n","scaler = preprocessing.StandardScaler().fit(train_data)\n","\n","x_train_trans = [] \n","for el in x_train: \n","  x_train_trans.append(scaler.transform(el))\n","del x_train \n","\n","x_train = np.asarray(x_train_trans) \n","x_train"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"3DE3PNbC-dmt","executionInfo":{"status":"ok","timestamp":1671208368211,"user_tz":-60,"elapsed":31,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"d383422a-c58d-4f2a-8299-ebcbfe792ec5"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([[[ 3.12954160e-02, -1.26683197e-02, -4.09212952e-02,\n","         -7.67097948e-02, -6.54563660e-02, -1.28408020e-01],\n","        [ 1.85450114e-02, -1.72277786e-02, -3.41257212e-02,\n","         -6.39470377e-02, -9.37877302e-02, -3.99561065e-02],\n","        [-1.61748253e-03, -2.26757961e-02, -2.58370532e-02,\n","         -5.21885500e-02, -8.82578456e-02,  1.26185885e-02],\n","        ...,\n","        [ 2.92121769e-01,  8.64184618e-02,  2.46020690e+00,\n","          1.83637912e+00,  1.86760338e+01,  1.19483326e+00],\n","        [ 2.36585982e-01,  5.49244493e-02,  2.16052902e+00,\n","          1.77476763e+00,  1.92397340e+01,  1.40341531e+00],\n","        [ 1.76763600e-01,  2.95116871e-02,  1.71227815e+00,\n","          1.51622362e+00,  1.64288577e+01,  1.43310326e+00]],\n","\n","       [[-2.91376343e-02, -2.39096432e-02, -4.38676844e-03,\n","         -5.33291264e-02, -4.87959024e-02, -3.18114542e-02],\n","        [-2.96265968e-02, -1.57240242e-02, -1.07058471e-02,\n","         -7.15750697e-02, -5.71934891e-02, -5.07341617e-02],\n","        [-3.04918229e-02, -6.21371102e-03, -1.41969629e-02,\n","         -7.94680731e-02, -6.18868623e-02, -6.48199312e-02],\n","        ...,\n","        [-7.65794573e-02, -3.07442460e-02, -4.69336218e-02,\n","         -4.71158929e-02, -5.64504543e-02, -1.08078961e-02],\n","        [-6.92044608e-02, -2.96311945e-02, -4.86219153e-02,\n","         -4.76368152e-02, -5.26850898e-02, -5.23478824e-03],\n","        [-6.19092205e-02, -2.83441092e-02, -4.83761116e-02,\n","         -4.79556754e-02, -4.99725356e-02, -5.98379385e-03]],\n","\n","       [[ 1.19264428e-01,  8.79664268e-01,  4.27572046e+00,\n","          3.50513506e-03,  4.68840868e-02,  6.39849381e-02],\n","        [ 2.53092227e-01,  1.09018370e+00,  4.70546105e+00,\n","          1.04852399e+00,  8.01475754e-02, -1.34403846e-01],\n","        [ 4.03533614e-01,  1.30760391e+00,  4.87293269e+00,\n","          2.13613451e+00,  6.68530937e-02, -2.04530818e-01],\n","        ...,\n","        [ 5.37988260e-02, -1.74185873e-02, -2.81779624e-02,\n","         -6.34950704e-02, -3.46684546e-02, -1.49385767e-01],\n","        [ 3.93669946e-02, -1.86661199e-02, -2.75021099e-02,\n","         -6.91072570e-02, -3.47029123e-02, -1.62569126e-01],\n","        [ 2.30315200e-02, -1.97068179e-02, -2.92150485e-02,\n","         -6.97861960e-02, -3.47777805e-02, -1.55142761e-01]],\n","\n","       ...,\n","\n","       [[ 1.96470623e+00,  2.09165122e+00,  2.33253208e+00,\n","          2.22194414e+00,  2.07678670e+00,  1.75962667e+00],\n","        [ 1.71424871e+00,  2.47172453e+00,  2.82832228e+00,\n","          2.75811316e+00,  2.60404240e+00,  2.21031762e+00],\n","        [ 1.48456826e+00,  2.62434629e+00,  3.07472239e+00,\n","          3.04332300e+00,  2.89826635e+00,  2.46613519e+00],\n","        ...,\n","        [-4.26267114e-02, -2.01022594e-02, -2.99576455e-02,\n","         -2.77349989e-02, -2.63297084e-02, -3.65894471e-03],\n","        [-4.33031880e-02, -1.93134470e-02, -2.94521596e-02,\n","         -2.81006554e-02, -2.56910491e-02, -1.80802301e-03],\n","        [-4.28255041e-02, -1.91278427e-02, -2.89955964e-02,\n","         -2.81585511e-02, -2.47615168e-02, -2.83897043e-03]],\n","\n","       [[ 3.19285731e-02, -2.05334058e-02, -4.30020722e-02,\n","         -4.72299399e-02, -7.42069568e-02, -3.06413546e-02],\n","        [ 4.26995816e-02, -2.04915198e-02, -4.29852490e-02,\n","         -4.79840060e-02, -7.25376528e-02, -7.23811871e-02],\n","        [ 5.30927921e-02, -2.05017264e-02, -4.27063039e-02,\n","         -4.88582118e-02, -6.73045572e-02, -1.01306171e-01],\n","        ...,\n","        [-2.10052817e-04, -1.70318882e-02, -3.87962774e-03,\n","          2.58271096e-03,  1.77635031e-02, -3.44144567e-02],\n","        [-4.43302259e-03, -1.82248371e-02, -8.56447341e-03,\n","         -4.35769142e-03,  7.04400656e-03, -4.65546416e-02],\n","        [-1.03346501e-02, -1.93829600e-02, -1.45911836e-02,\n","         -1.32540398e-02,  3.61151558e-04, -4.85870440e-02]],\n","\n","       [[-3.46401276e-02, -2.33477888e-02, -3.73647721e-02,\n","         -4.70848932e-02, -4.95453196e-02, -4.22622915e-02],\n","        [-4.69543466e-02, -2.16329020e-02, -3.66413286e-02,\n","         -4.69276823e-02, -5.09532853e-02, -5.09549789e-02],\n","        [-5.33666369e-02, -2.07389306e-02, -3.63849463e-02,\n","         -4.74695777e-02, -5.11445508e-02, -5.15087479e-02],\n","        ...,\n","        [-2.15803825e-02, -2.21561157e-02, -3.83923620e-02,\n","         -4.53495552e-02, -5.02776919e-02, -3.68182983e-02],\n","        [-2.45934831e-02, -2.29348210e-02, -3.79439398e-02,\n","         -4.47448012e-02, -5.07155804e-02, -3.53844671e-02],\n","        [-3.41087638e-02, -2.39507326e-02, -3.73385698e-02,\n","         -4.49690812e-02, -4.94514318e-02, -3.32982197e-02]]])"]},"metadata":{},"execution_count":39}]},{"cell_type":"code","source":["# check \n","print(x_train.shape, y_train.shape) \n","x_train[0][1]\n","y_train[0]"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"BjouEKJoABlN","executionInfo":{"status":"ok","timestamp":1671208368212,"user_tz":-60,"elapsed":26,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"b9115c4c-cb54-4c42-86a7-819b0db9d3fd"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["(2429, 117, 6) (2429, 12)\n"]},{"output_type":"execute_result","data":{"text/plain":["array([1., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], dtype=float32)"]},"metadata":{},"execution_count":40}]},{"cell_type":"markdown","source":["**Defining the training sets**"],"metadata":{"id":"VSP-6T9Vu8BO"}},{"cell_type":"code","source":["x_train1, x_test1, y_train1, y_test1 = model_selection.train_test_split(x_train, y_train, test_size=0.2, random_state=1, stratify=y_train)\n","x_train2, x_test2, y_train2, y_test2 = model_selection.train_test_split(x_train, y_train, test_size=0.2, random_state=2, stratify=y_train)\n","x_train3, x_test3, y_train3, y_test3 = model_selection.train_test_split(x_train, y_train, test_size=0.2, random_state=3, stratify=y_train)\n","x_train4, x_test4, y_train4, y_test4 = model_selection.train_test_split(x_train, y_train, test_size=0.2, random_state=4, stratify=y_train)\n","x_train5, x_test5, y_train5, y_test5 = model_selection.train_test_split(x_train, y_train, test_size=0.2, random_state=5, stratify=y_train)\n","x_train6, x_test6, y_train6, y_test6 = model_selection.train_test_split(x_train, y_train, test_size=0.2, random_state=6, stratify=y_train)\n","x_train7, x_test7, y_train7, y_test7 = model_selection.train_test_split(x_train, y_train, test_size=0.2, random_state=7, stratify=y_train)\n","x_train8, x_test8, y_train8, y_test8 = model_selection.train_test_split(x_train, y_train, test_size=0.2, random_state=8, stratify=y_train)\n","x_train9, x_test9, y_train9, y_test9 = model_selection.train_test_split(x_train, y_train, test_size=0.2, random_state=9, stratify=y_train)\n","x_train10, x_test10, y_train10, y_test10 = model_selection.train_test_split(x_train, y_train, test_size=0.2, random_state=10, stratify=y_train)"],"metadata":{"id":"gaghP_Qxu_id"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["# **LSTM VANILLA**"],"metadata":{"id":"BmMB9Pj6t3RX"}},{"cell_type":"code","source":["input_shape = (117,6)"],"metadata":{"id":"KRI6J1GZuQHw"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","metadata":{"id":"FqTU7SDgjWgV"},"source":["#### 1D Convolutional Neural Network\n","\n","![](https://drive.google.com/uc?export=view&id=10B7c-F58-eNr6CP7D_UZeVp0gTeYMrFQ\n",")"]},{"cell_type":"code","metadata":{"id":"7nxJrPoojt3z"},"source":["def build_ffnn_classifier(input_shape, classes):\n","    input_layer = tfkl.Input(shape=input_shape, name='Input')\n","\n","    cnn = tfkl.Conv1D(256,3,padding='same',activation='relu')(input_layer)\n","    cnn = tfkl.MaxPooling1D()(cnn)\n","    cnn = tfkl.Conv1D(256,3,padding='same',activation='relu')(cnn)\n","    cnn = tfkl.MaxPooling1D()(cnn)\n","    cnn = tfkl.Conv1D(256,3,padding='same',activation='relu')(cnn)\n","    gap = tfkl.GlobalAveragePooling1D()(cnn)\n","    dropout = tfkl.Dropout(.3, seed=seed)(gap)\n","\n","\n","    hidden_layer1 = tfkl.Dense(units=256, activation ='relu', name = 'Hidden1', kernel_regularizer = tfk.regularizers.L2(l2 = 0.0001))(dropout) \n","\n","    hidden_layer2 = tfkl.Dense(units=256, activation ='relu', name = 'Hidden2', kernel_regularizer = tfk.regularizers.L2(l2 = 0.0001))(hidden_layer1) \n","\n","    \n","    output_layer = tfkl.Dense(classes, activation='softmax', \n","                              kernel_initializer=tfk.initializers.GlorotUniform(seed))(hidden_layer2)\n","      \n","    model = tfk.Model(inputs=input_layer, outputs=output_layer, name='model')\n","    model.compile(loss=tfk.losses.CategoricalCrossentropy(), optimizer=tfk.optimizers.Adam(), metrics='accuracy')\n","    return model"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"d1zEtb3Hjt1O","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1671208372054,"user_tz":-60,"elapsed":3149,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"72567cf9-cfa6-4fe3-efa1-909495b7b3df"},"source":["model1 = build_ffnn_classifier(input_shape, 12)\n","model2 = build_ffnn_classifier(input_shape, 12)\n","model3 = build_ffnn_classifier(input_shape, 12)\n","model4 = build_ffnn_classifier(input_shape, 12)\n","model5 = build_ffnn_classifier(input_shape, 12)\n","model6 = build_ffnn_classifier(input_shape, 12)\n","model7 = build_ffnn_classifier(input_shape, 12)\n","model8 = build_ffnn_classifier(input_shape, 12)\n","model9 = build_ffnn_classifier(input_shape, 12)\n","model10 = build_ffnn_classifier(input_shape, 12)\n","\n","model7.summary()"],"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Model: \"model\"\n","_________________________________________________________________\n"," Layer (type)                Output Shape              Param #   \n","=================================================================\n"," Input (InputLayer)          [(None, 117, 6)]          0         \n","                                                                 \n"," conv1d_18 (Conv1D)          (None, 117, 256)          4864      \n","                                                                 \n"," max_pooling1d_12 (MaxPoolin  (None, 58, 256)          0         \n"," g1D)                                                            \n","                                                                 \n"," conv1d_19 (Conv1D)          (None, 58, 256)           196864    \n","                                                                 \n"," max_pooling1d_13 (MaxPoolin  (None, 29, 256)          0         \n"," g1D)                                                            \n","                                                                 \n"," conv1d_20 (Conv1D)          (None, 29, 256)           196864    \n","                                                                 \n"," global_average_pooling1d_6   (None, 256)              0         \n"," (GlobalAveragePooling1D)                                        \n","                                                                 \n"," dropout_6 (Dropout)         (None, 256)               0         \n","                                                                 \n"," Hidden1 (Dense)             (None, 256)               65792     \n","                                                                 \n"," Hidden2 (Dense)             (None, 256)               65792     \n","                                                                 \n"," dense_6 (Dense)             (None, 12)                3084      \n","                                                                 \n","=================================================================\n","Total params: 533,260\n","Trainable params: 533,260\n","Non-trainable params: 0\n","_________________________________________________________________\n"]}]},{"cell_type":"markdown","source":["# training"],"metadata":{"id":"G4OjjxFVvwd9"}},{"cell_type":"code","metadata":{"id":"LQOYU2X0jty5","colab":{"base_uri":"https://localhost:8080/"},"outputId":"e1bcb52f-a80a-4cb0-d66e-4fedf53a6b71","executionInfo":{"status":"ok","timestamp":1671208420764,"user_tz":-60,"elapsed":48718,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}}},"source":["history1 = model1.fit(\n","    x = x_train1,\n","    y = y_train1,\n","    batch_size = 64,\n","    epochs = 200,\n","    validation_data = (x_test1, y_test1),\n","    callbacks = [\n","        tfk.callbacks.EarlyStopping(monitor='val_accuracy', mode='max', patience=30, restore_best_weights=True),\n","        tfk.callbacks.ReduceLROnPlateau(monitor='val_accuracy', mode='max', patience=15, factor=0.5, min_lr=1e-5)\n","    ]\n",").history"],"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/200\n","31/31 [==============================] - 11s 33ms/step - loss: 2.2523 - accuracy: 0.3387 - val_loss: 1.9855 - val_accuracy: 0.3519 - lr: 0.0010\n","Epoch 2/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.9237 - accuracy: 0.3726 - val_loss: 1.8568 - val_accuracy: 0.4033 - lr: 0.0010\n","Epoch 3/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.8624 - accuracy: 0.4035 - val_loss: 1.7913 - val_accuracy: 0.3992 - lr: 0.0010\n","Epoch 4/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.7485 - accuracy: 0.4277 - val_loss: 1.6454 - val_accuracy: 0.4506 - lr: 0.0010\n","Epoch 5/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.6302 - accuracy: 0.4498 - val_loss: 1.5087 - val_accuracy: 0.5041 - lr: 0.0010\n","Epoch 6/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.6597 - accuracy: 0.4637 - val_loss: 1.5617 - val_accuracy: 0.5000 - lr: 0.0010\n","Epoch 7/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.5000 - accuracy: 0.5095 - val_loss: 1.5074 - val_accuracy: 0.4959 - lr: 0.0010\n","Epoch 8/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.4950 - accuracy: 0.5136 - val_loss: 1.3800 - val_accuracy: 0.5247 - lr: 0.0010\n","Epoch 9/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.4020 - accuracy: 0.5203 - val_loss: 1.4143 - val_accuracy: 0.5123 - lr: 0.0010\n","Epoch 10/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3389 - accuracy: 0.5394 - val_loss: 1.3799 - val_accuracy: 0.5247 - lr: 0.0010\n","Epoch 11/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.4316 - accuracy: 0.5322 - val_loss: 1.4639 - val_accuracy: 0.5082 - lr: 0.0010\n","Epoch 12/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.3797 - accuracy: 0.5404 - val_loss: 1.3728 - val_accuracy: 0.5247 - lr: 0.0010\n","Epoch 13/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2855 - accuracy: 0.5584 - val_loss: 1.3072 - val_accuracy: 0.5514 - lr: 0.0010\n","Epoch 14/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3348 - accuracy: 0.5661 - val_loss: 1.2824 - val_accuracy: 0.6029 - lr: 0.0010\n","Epoch 15/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2577 - accuracy: 0.5764 - val_loss: 1.2533 - val_accuracy: 0.6091 - lr: 0.0010\n","Epoch 16/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1893 - accuracy: 0.6125 - val_loss: 1.2412 - val_accuracy: 0.5926 - lr: 0.0010\n","Epoch 17/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2090 - accuracy: 0.5888 - val_loss: 1.3025 - val_accuracy: 0.6111 - lr: 0.0010\n","Epoch 18/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2016 - accuracy: 0.6068 - val_loss: 1.2182 - val_accuracy: 0.6461 - lr: 0.0010\n","Epoch 19/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1623 - accuracy: 0.6161 - val_loss: 1.2318 - val_accuracy: 0.5741 - lr: 0.0010\n","Epoch 20/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1585 - accuracy: 0.6063 - val_loss: 1.1605 - val_accuracy: 0.6502 - lr: 0.0010\n","Epoch 21/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1708 - accuracy: 0.6119 - val_loss: 1.2075 - val_accuracy: 0.6235 - lr: 0.0010\n","Epoch 22/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1204 - accuracy: 0.6356 - val_loss: 1.1593 - val_accuracy: 0.6440 - lr: 0.0010\n","Epoch 23/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0684 - accuracy: 0.6583 - val_loss: 1.1486 - val_accuracy: 0.6461 - lr: 0.0010\n","Epoch 24/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0469 - accuracy: 0.6495 - val_loss: 1.1458 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 25/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0824 - accuracy: 0.6402 - val_loss: 1.1524 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 26/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0442 - accuracy: 0.6536 - val_loss: 1.1535 - val_accuracy: 0.6337 - lr: 0.0010\n","Epoch 27/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0615 - accuracy: 0.6418 - val_loss: 1.1114 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 28/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0408 - accuracy: 0.6655 - val_loss: 1.2378 - val_accuracy: 0.6235 - lr: 0.0010\n","Epoch 29/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0947 - accuracy: 0.6387 - val_loss: 1.1263 - val_accuracy: 0.6440 - lr: 0.0010\n","Epoch 30/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0229 - accuracy: 0.6572 - val_loss: 1.0904 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 31/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9787 - accuracy: 0.6588 - val_loss: 1.1103 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 32/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9484 - accuracy: 0.6819 - val_loss: 1.2963 - val_accuracy: 0.6358 - lr: 0.0010\n","Epoch 33/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9722 - accuracy: 0.6855 - val_loss: 1.0822 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 34/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9643 - accuracy: 0.6752 - val_loss: 1.1069 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 35/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9693 - accuracy: 0.6824 - val_loss: 1.0466 - val_accuracy: 0.6440 - lr: 0.0010\n","Epoch 36/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9374 - accuracy: 0.6819 - val_loss: 1.0313 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 37/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8948 - accuracy: 0.6994 - val_loss: 1.0211 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 38/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8836 - accuracy: 0.6999 - val_loss: 1.0240 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 39/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9011 - accuracy: 0.7056 - val_loss: 1.0783 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 40/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8711 - accuracy: 0.7123 - val_loss: 1.0338 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 41/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8761 - accuracy: 0.7092 - val_loss: 1.1593 - val_accuracy: 0.6461 - lr: 0.0010\n","Epoch 42/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9297 - accuracy: 0.6763 - val_loss: 1.0211 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 43/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8599 - accuracy: 0.7169 - val_loss: 1.0531 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 44/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9038 - accuracy: 0.6989 - val_loss: 1.0899 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 45/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8199 - accuracy: 0.7344 - val_loss: 1.0650 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 46/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9201 - accuracy: 0.7097 - val_loss: 1.1536 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 47/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9223 - accuracy: 0.7051 - val_loss: 1.1483 - val_accuracy: 0.6440 - lr: 0.0010\n","Epoch 48/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9796 - accuracy: 0.6778 - val_loss: 1.1050 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 49/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8747 - accuracy: 0.7190 - val_loss: 1.0750 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 50/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8271 - accuracy: 0.7247 - val_loss: 1.0926 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 51/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7988 - accuracy: 0.7458 - val_loss: 1.1186 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 52/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8205 - accuracy: 0.7355 - val_loss: 0.9949 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 53/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7965 - accuracy: 0.7411 - val_loss: 1.1399 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 54/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8755 - accuracy: 0.7272 - val_loss: 1.1223 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 55/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8460 - accuracy: 0.7334 - val_loss: 1.1026 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 56/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7914 - accuracy: 0.7411 - val_loss: 1.0384 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 57/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7827 - accuracy: 0.7488 - val_loss: 1.0173 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 58/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7980 - accuracy: 0.7401 - val_loss: 0.9884 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 59/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7800 - accuracy: 0.7458 - val_loss: 1.0054 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 60/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8627 - accuracy: 0.7231 - val_loss: 1.1359 - val_accuracy: 0.6317 - lr: 0.0010\n","Epoch 61/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8182 - accuracy: 0.7324 - val_loss: 1.0123 - val_accuracy: 0.6728 - lr: 5.0000e-04\n","Epoch 62/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7168 - accuracy: 0.7638 - val_loss: 1.0050 - val_accuracy: 0.6893 - lr: 5.0000e-04\n","Epoch 63/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6981 - accuracy: 0.7715 - val_loss: 0.9952 - val_accuracy: 0.6955 - lr: 5.0000e-04\n","Epoch 64/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6874 - accuracy: 0.7818 - val_loss: 1.0035 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 65/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6619 - accuracy: 0.7823 - val_loss: 1.0018 - val_accuracy: 0.6770 - lr: 5.0000e-04\n","Epoch 66/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6605 - accuracy: 0.7854 - val_loss: 1.0214 - val_accuracy: 0.6708 - lr: 5.0000e-04\n","Epoch 67/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6661 - accuracy: 0.7844 - val_loss: 0.9740 - val_accuracy: 0.6975 - lr: 5.0000e-04\n","Epoch 68/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6519 - accuracy: 0.7885 - val_loss: 1.0163 - val_accuracy: 0.6852 - lr: 5.0000e-04\n","Epoch 69/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6500 - accuracy: 0.7808 - val_loss: 1.0113 - val_accuracy: 0.6790 - lr: 5.0000e-04\n","Epoch 70/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6500 - accuracy: 0.7895 - val_loss: 1.0024 - val_accuracy: 0.6996 - lr: 5.0000e-04\n","Epoch 71/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6284 - accuracy: 0.7895 - val_loss: 1.0319 - val_accuracy: 0.6749 - lr: 5.0000e-04\n","Epoch 72/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6603 - accuracy: 0.7792 - val_loss: 1.0275 - val_accuracy: 0.6852 - lr: 5.0000e-04\n","Epoch 73/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6431 - accuracy: 0.7957 - val_loss: 1.0099 - val_accuracy: 0.6955 - lr: 5.0000e-04\n","Epoch 74/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6123 - accuracy: 0.7983 - val_loss: 1.0200 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 75/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6275 - accuracy: 0.7921 - val_loss: 1.0525 - val_accuracy: 0.6749 - lr: 5.0000e-04\n","Epoch 76/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6169 - accuracy: 0.7962 - val_loss: 1.0521 - val_accuracy: 0.6852 - lr: 5.0000e-04\n","Epoch 77/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6197 - accuracy: 0.8008 - val_loss: 1.0386 - val_accuracy: 0.6790 - lr: 5.0000e-04\n","Epoch 78/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6068 - accuracy: 0.8008 - val_loss: 1.0557 - val_accuracy: 0.6975 - lr: 5.0000e-04\n","Epoch 79/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6013 - accuracy: 0.8065 - val_loss: 1.0647 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 80/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6076 - accuracy: 0.8044 - val_loss: 1.0615 - val_accuracy: 0.6934 - lr: 5.0000e-04\n","Epoch 81/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5981 - accuracy: 0.8121 - val_loss: 1.1149 - val_accuracy: 0.6543 - lr: 5.0000e-04\n","Epoch 82/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6032 - accuracy: 0.8019 - val_loss: 1.0996 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 83/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5784 - accuracy: 0.8096 - val_loss: 1.1084 - val_accuracy: 0.6852 - lr: 5.0000e-04\n","Epoch 84/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5864 - accuracy: 0.8116 - val_loss: 1.0654 - val_accuracy: 0.6831 - lr: 5.0000e-04\n","Epoch 85/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5890 - accuracy: 0.8116 - val_loss: 1.0683 - val_accuracy: 0.6852 - lr: 5.0000e-04\n","Epoch 86/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5641 - accuracy: 0.8157 - val_loss: 1.0908 - val_accuracy: 0.6934 - lr: 2.5000e-04\n","Epoch 87/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5819 - accuracy: 0.8219 - val_loss: 1.0744 - val_accuracy: 0.6831 - lr: 2.5000e-04\n","Epoch 88/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5599 - accuracy: 0.8240 - val_loss: 1.0683 - val_accuracy: 0.6934 - lr: 2.5000e-04\n","Epoch 89/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5710 - accuracy: 0.8173 - val_loss: 1.0820 - val_accuracy: 0.6996 - lr: 2.5000e-04\n","Epoch 90/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5456 - accuracy: 0.8245 - val_loss: 1.0716 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 91/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5593 - accuracy: 0.8250 - val_loss: 1.0570 - val_accuracy: 0.7037 - lr: 2.5000e-04\n","Epoch 92/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5350 - accuracy: 0.8276 - val_loss: 1.0787 - val_accuracy: 0.7016 - lr: 2.5000e-04\n","Epoch 93/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5335 - accuracy: 0.8322 - val_loss: 1.0688 - val_accuracy: 0.7119 - lr: 2.5000e-04\n","Epoch 94/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5308 - accuracy: 0.8302 - val_loss: 1.0615 - val_accuracy: 0.6996 - lr: 2.5000e-04\n","Epoch 95/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5331 - accuracy: 0.8255 - val_loss: 1.0685 - val_accuracy: 0.7119 - lr: 2.5000e-04\n","Epoch 96/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5224 - accuracy: 0.8317 - val_loss: 1.0757 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 97/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5277 - accuracy: 0.8410 - val_loss: 1.0517 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 98/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5196 - accuracy: 0.8379 - val_loss: 1.0839 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 99/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5221 - accuracy: 0.8307 - val_loss: 1.0483 - val_accuracy: 0.7078 - lr: 2.5000e-04\n","Epoch 100/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5152 - accuracy: 0.8374 - val_loss: 1.0592 - val_accuracy: 0.7140 - lr: 2.5000e-04\n","Epoch 101/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5201 - accuracy: 0.8312 - val_loss: 1.0646 - val_accuracy: 0.7016 - lr: 2.5000e-04\n","Epoch 102/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5040 - accuracy: 0.8425 - val_loss: 1.1043 - val_accuracy: 0.6996 - lr: 2.5000e-04\n","Epoch 103/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5020 - accuracy: 0.8451 - val_loss: 1.1427 - val_accuracy: 0.6811 - lr: 2.5000e-04\n","Epoch 104/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5062 - accuracy: 0.8317 - val_loss: 1.0813 - val_accuracy: 0.7099 - lr: 2.5000e-04\n","Epoch 105/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5023 - accuracy: 0.8384 - val_loss: 1.0896 - val_accuracy: 0.7078 - lr: 2.5000e-04\n","Epoch 106/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4882 - accuracy: 0.8456 - val_loss: 1.1117 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 107/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5143 - accuracy: 0.8425 - val_loss: 1.1592 - val_accuracy: 0.6996 - lr: 2.5000e-04\n","Epoch 108/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5114 - accuracy: 0.8379 - val_loss: 1.0822 - val_accuracy: 0.7181 - lr: 2.5000e-04\n","Epoch 109/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4864 - accuracy: 0.8518 - val_loss: 1.1088 - val_accuracy: 0.7140 - lr: 2.5000e-04\n","Epoch 110/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4946 - accuracy: 0.8405 - val_loss: 1.1108 - val_accuracy: 0.7099 - lr: 2.5000e-04\n","Epoch 111/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4773 - accuracy: 0.8574 - val_loss: 1.1417 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 112/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4897 - accuracy: 0.8482 - val_loss: 1.1394 - val_accuracy: 0.7016 - lr: 2.5000e-04\n","Epoch 113/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4797 - accuracy: 0.8466 - val_loss: 1.1614 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 114/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4688 - accuracy: 0.8528 - val_loss: 1.1432 - val_accuracy: 0.6996 - lr: 2.5000e-04\n","Epoch 115/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4729 - accuracy: 0.8394 - val_loss: 1.1877 - val_accuracy: 0.7016 - lr: 2.5000e-04\n","Epoch 116/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4798 - accuracy: 0.8518 - val_loss: 1.1928 - val_accuracy: 0.7016 - lr: 2.5000e-04\n","Epoch 117/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5132 - accuracy: 0.8487 - val_loss: 1.1919 - val_accuracy: 0.7037 - lr: 2.5000e-04\n","Epoch 118/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4821 - accuracy: 0.8471 - val_loss: 1.2047 - val_accuracy: 0.6893 - lr: 2.5000e-04\n","Epoch 119/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4664 - accuracy: 0.8621 - val_loss: 1.1848 - val_accuracy: 0.7099 - lr: 2.5000e-04\n","Epoch 120/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4650 - accuracy: 0.8605 - val_loss: 1.1647 - val_accuracy: 0.7037 - lr: 2.5000e-04\n","Epoch 121/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4666 - accuracy: 0.8528 - val_loss: 1.1518 - val_accuracy: 0.6872 - lr: 2.5000e-04\n","Epoch 122/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4654 - accuracy: 0.8549 - val_loss: 1.1608 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 123/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4665 - accuracy: 0.8569 - val_loss: 1.1729 - val_accuracy: 0.6996 - lr: 2.5000e-04\n","Epoch 124/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4539 - accuracy: 0.8636 - val_loss: 1.1653 - val_accuracy: 0.7058 - lr: 1.2500e-04\n","Epoch 125/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4477 - accuracy: 0.8559 - val_loss: 1.1700 - val_accuracy: 0.7058 - lr: 1.2500e-04\n","Epoch 126/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4488 - accuracy: 0.8636 - val_loss: 1.2502 - val_accuracy: 0.6934 - lr: 1.2500e-04\n","Epoch 127/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4452 - accuracy: 0.8595 - val_loss: 1.2496 - val_accuracy: 0.6934 - lr: 1.2500e-04\n","Epoch 128/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4352 - accuracy: 0.8595 - val_loss: 1.2576 - val_accuracy: 0.6852 - lr: 1.2500e-04\n","Epoch 129/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4329 - accuracy: 0.8631 - val_loss: 1.2501 - val_accuracy: 0.7037 - lr: 1.2500e-04\n","Epoch 130/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4267 - accuracy: 0.8672 - val_loss: 1.2571 - val_accuracy: 0.7037 - lr: 1.2500e-04\n","Epoch 131/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4299 - accuracy: 0.8682 - val_loss: 1.2187 - val_accuracy: 0.6893 - lr: 1.2500e-04\n","Epoch 132/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4266 - accuracy: 0.8755 - val_loss: 1.2085 - val_accuracy: 0.7037 - lr: 1.2500e-04\n","Epoch 133/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4275 - accuracy: 0.8724 - val_loss: 1.2204 - val_accuracy: 0.7058 - lr: 1.2500e-04\n","Epoch 134/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4282 - accuracy: 0.8672 - val_loss: 1.2455 - val_accuracy: 0.7058 - lr: 1.2500e-04\n","Epoch 135/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4172 - accuracy: 0.8739 - val_loss: 1.2563 - val_accuracy: 0.6955 - lr: 1.2500e-04\n","Epoch 136/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4201 - accuracy: 0.8631 - val_loss: 1.2392 - val_accuracy: 0.7058 - lr: 1.2500e-04\n","Epoch 137/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4341 - accuracy: 0.8657 - val_loss: 1.1930 - val_accuracy: 0.6934 - lr: 1.2500e-04\n","Epoch 138/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4204 - accuracy: 0.8739 - val_loss: 1.2087 - val_accuracy: 0.7099 - lr: 1.2500e-04\n"]}]},{"cell_type":"code","source":["history2 = model2.fit(\n","    x = x_train2,\n","    y = y_train2,\n","    batch_size = 64,\n","    epochs = 200,\n","    validation_data = (x_test2, y_test2),\n","    callbacks = [\n","        tfk.callbacks.EarlyStopping(monitor='val_accuracy', mode='max', patience=30, restore_best_weights=True),\n","        tfk.callbacks.ReduceLROnPlateau(monitor='val_accuracy', mode='max', patience=15, factor=0.5, min_lr=1e-5)\n","    ]\n",").history"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"fDWPFIo6wO1x","executionInfo":{"status":"ok","timestamp":1671208479719,"user_tz":-60,"elapsed":58976,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"75520077-87e8-43d9-8a7d-01181246d94e"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/200\n","31/31 [==============================] - 2s 32ms/step - loss: 2.2665 - accuracy: 0.3345 - val_loss: 1.9695 - val_accuracy: 0.3724 - lr: 0.0010\n","Epoch 2/200\n","31/31 [==============================] - 1s 20ms/step - loss: 1.9386 - accuracy: 0.3850 - val_loss: 1.9211 - val_accuracy: 0.4198 - lr: 0.0010\n","Epoch 3/200\n","31/31 [==============================] - 1s 22ms/step - loss: 1.8731 - accuracy: 0.3953 - val_loss: 1.8601 - val_accuracy: 0.4198 - lr: 0.0010\n","Epoch 4/200\n","31/31 [==============================] - 1s 20ms/step - loss: 1.7560 - accuracy: 0.4364 - val_loss: 1.7599 - val_accuracy: 0.4218 - lr: 0.0010\n","Epoch 5/200\n","31/31 [==============================] - 1s 19ms/step - loss: 1.6452 - accuracy: 0.4632 - val_loss: 1.6421 - val_accuracy: 0.4280 - lr: 0.0010\n","Epoch 6/200\n","31/31 [==============================] - 0s 14ms/step - loss: 1.5693 - accuracy: 0.4936 - val_loss: 1.5578 - val_accuracy: 0.4671 - lr: 0.0010\n","Epoch 7/200\n","31/31 [==============================] - 0s 14ms/step - loss: 1.5779 - accuracy: 0.4802 - val_loss: 1.5831 - val_accuracy: 0.4650 - lr: 0.0010\n","Epoch 8/200\n","31/31 [==============================] - 0s 16ms/step - loss: 1.4301 - accuracy: 0.5229 - val_loss: 1.5178 - val_accuracy: 0.5103 - lr: 0.0010\n","Epoch 9/200\n","31/31 [==============================] - 0s 15ms/step - loss: 1.3558 - accuracy: 0.5363 - val_loss: 1.5851 - val_accuracy: 0.5062 - lr: 0.0010\n","Epoch 10/200\n","31/31 [==============================] - 0s 12ms/step - loss: 1.3779 - accuracy: 0.5409 - val_loss: 1.3336 - val_accuracy: 0.5165 - lr: 0.0010\n","Epoch 11/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.3272 - accuracy: 0.5528 - val_loss: 1.5897 - val_accuracy: 0.5206 - lr: 0.0010\n","Epoch 12/200\n","31/31 [==============================] - 0s 16ms/step - loss: 1.3188 - accuracy: 0.5615 - val_loss: 1.4331 - val_accuracy: 0.5535 - lr: 0.0010\n","Epoch 13/200\n","31/31 [==============================] - 1s 16ms/step - loss: 1.2984 - accuracy: 0.5764 - val_loss: 1.3695 - val_accuracy: 0.5844 - lr: 0.0010\n","Epoch 14/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.3277 - accuracy: 0.5677 - val_loss: 1.5894 - val_accuracy: 0.5350 - lr: 0.0010\n","Epoch 15/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2898 - accuracy: 0.5749 - val_loss: 1.6307 - val_accuracy: 0.5288 - lr: 0.0010\n","Epoch 16/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.2235 - accuracy: 0.5975 - val_loss: 1.5204 - val_accuracy: 0.5967 - lr: 0.0010\n","Epoch 17/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.1587 - accuracy: 0.6212 - val_loss: 1.3852 - val_accuracy: 0.6193 - lr: 0.0010\n","Epoch 18/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1531 - accuracy: 0.6083 - val_loss: 1.2521 - val_accuracy: 0.5844 - lr: 0.0010\n","Epoch 19/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1621 - accuracy: 0.6238 - val_loss: 1.2872 - val_accuracy: 0.6235 - lr: 0.0010\n","Epoch 20/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1232 - accuracy: 0.6315 - val_loss: 1.4751 - val_accuracy: 0.6337 - lr: 0.0010\n","Epoch 21/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1253 - accuracy: 0.6464 - val_loss: 1.4946 - val_accuracy: 0.6276 - lr: 0.0010\n","Epoch 22/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1469 - accuracy: 0.6284 - val_loss: 1.2925 - val_accuracy: 0.6029 - lr: 0.0010\n","Epoch 23/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1482 - accuracy: 0.6155 - val_loss: 1.3523 - val_accuracy: 0.6070 - lr: 0.0010\n","Epoch 24/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2309 - accuracy: 0.5980 - val_loss: 1.4814 - val_accuracy: 0.5535 - lr: 0.0010\n","Epoch 25/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1557 - accuracy: 0.6150 - val_loss: 1.2411 - val_accuracy: 0.6255 - lr: 0.0010\n","Epoch 26/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0794 - accuracy: 0.6438 - val_loss: 1.3479 - val_accuracy: 0.6132 - lr: 0.0010\n","Epoch 27/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0386 - accuracy: 0.6547 - val_loss: 1.2521 - val_accuracy: 0.6358 - lr: 0.0010\n","Epoch 28/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0175 - accuracy: 0.6598 - val_loss: 1.2246 - val_accuracy: 0.6502 - lr: 0.0010\n","Epoch 29/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0015 - accuracy: 0.6701 - val_loss: 1.2371 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 30/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0712 - accuracy: 0.6531 - val_loss: 1.4607 - val_accuracy: 0.6070 - lr: 0.0010\n","Epoch 31/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0470 - accuracy: 0.6588 - val_loss: 1.5394 - val_accuracy: 0.6337 - lr: 0.0010\n","Epoch 32/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9902 - accuracy: 0.6639 - val_loss: 1.2567 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 33/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9774 - accuracy: 0.6830 - val_loss: 1.3706 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 34/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9695 - accuracy: 0.6773 - val_loss: 1.1895 - val_accuracy: 0.6399 - lr: 0.0010\n","Epoch 35/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9443 - accuracy: 0.6881 - val_loss: 1.2293 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 36/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9465 - accuracy: 0.6850 - val_loss: 1.2746 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 37/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9227 - accuracy: 0.6958 - val_loss: 1.2381 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 38/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9138 - accuracy: 0.6927 - val_loss: 1.3167 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 39/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9753 - accuracy: 0.6778 - val_loss: 1.3062 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 40/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9405 - accuracy: 0.6850 - val_loss: 1.3056 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 41/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9208 - accuracy: 0.6897 - val_loss: 1.3935 - val_accuracy: 0.6379 - lr: 0.0010\n","Epoch 42/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9194 - accuracy: 0.7005 - val_loss: 1.5480 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 43/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8836 - accuracy: 0.7118 - val_loss: 1.5921 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 44/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8879 - accuracy: 0.7133 - val_loss: 1.3558 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 45/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8817 - accuracy: 0.7128 - val_loss: 1.2677 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 46/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8642 - accuracy: 0.7195 - val_loss: 1.2419 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 47/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8714 - accuracy: 0.7118 - val_loss: 1.2145 - val_accuracy: 0.6255 - lr: 0.0010\n","Epoch 48/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9349 - accuracy: 0.7030 - val_loss: 1.2741 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 49/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0636 - accuracy: 0.6608 - val_loss: 1.3929 - val_accuracy: 0.5844 - lr: 0.0010\n","Epoch 50/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0594 - accuracy: 0.6485 - val_loss: 1.4076 - val_accuracy: 0.6296 - lr: 0.0010\n","Epoch 51/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8980 - accuracy: 0.7097 - val_loss: 1.2686 - val_accuracy: 0.6523 - lr: 5.0000e-04\n","Epoch 52/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8441 - accuracy: 0.7185 - val_loss: 1.2188 - val_accuracy: 0.6646 - lr: 5.0000e-04\n","Epoch 53/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8238 - accuracy: 0.7293 - val_loss: 1.2649 - val_accuracy: 0.6687 - lr: 5.0000e-04\n","Epoch 54/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8094 - accuracy: 0.7375 - val_loss: 1.2055 - val_accuracy: 0.6584 - lr: 5.0000e-04\n","Epoch 55/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8039 - accuracy: 0.7288 - val_loss: 1.2490 - val_accuracy: 0.6626 - lr: 5.0000e-04\n","Epoch 56/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7925 - accuracy: 0.7401 - val_loss: 1.2729 - val_accuracy: 0.6728 - lr: 5.0000e-04\n","Epoch 57/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7748 - accuracy: 0.7488 - val_loss: 1.1940 - val_accuracy: 0.6770 - lr: 5.0000e-04\n","Epoch 58/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8097 - accuracy: 0.7411 - val_loss: 1.1612 - val_accuracy: 0.6687 - lr: 5.0000e-04\n","Epoch 59/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7931 - accuracy: 0.7365 - val_loss: 1.1894 - val_accuracy: 0.6564 - lr: 5.0000e-04\n","Epoch 60/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7788 - accuracy: 0.7504 - val_loss: 1.1989 - val_accuracy: 0.6687 - lr: 5.0000e-04\n","Epoch 61/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7582 - accuracy: 0.7555 - val_loss: 1.2108 - val_accuracy: 0.6605 - lr: 5.0000e-04\n","Epoch 62/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7471 - accuracy: 0.7560 - val_loss: 1.2121 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 63/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7429 - accuracy: 0.7622 - val_loss: 1.2076 - val_accuracy: 0.6708 - lr: 5.0000e-04\n","Epoch 64/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7280 - accuracy: 0.7597 - val_loss: 1.2812 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 65/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7193 - accuracy: 0.7648 - val_loss: 1.3244 - val_accuracy: 0.6543 - lr: 5.0000e-04\n","Epoch 66/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7265 - accuracy: 0.7669 - val_loss: 1.2567 - val_accuracy: 0.6543 - lr: 5.0000e-04\n","Epoch 67/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7425 - accuracy: 0.7633 - val_loss: 1.2687 - val_accuracy: 0.6790 - lr: 5.0000e-04\n","Epoch 68/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7096 - accuracy: 0.7679 - val_loss: 1.2318 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 69/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7096 - accuracy: 0.7679 - val_loss: 1.1819 - val_accuracy: 0.6770 - lr: 5.0000e-04\n","Epoch 70/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7314 - accuracy: 0.7699 - val_loss: 1.2718 - val_accuracy: 0.6687 - lr: 5.0000e-04\n","Epoch 71/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7146 - accuracy: 0.7658 - val_loss: 1.2480 - val_accuracy: 0.6667 - lr: 5.0000e-04\n","Epoch 72/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7201 - accuracy: 0.7612 - val_loss: 1.1875 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 73/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7085 - accuracy: 0.7725 - val_loss: 1.1667 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 74/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7166 - accuracy: 0.7730 - val_loss: 1.2444 - val_accuracy: 0.6996 - lr: 5.0000e-04\n","Epoch 75/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6886 - accuracy: 0.7756 - val_loss: 1.2099 - val_accuracy: 0.6934 - lr: 5.0000e-04\n","Epoch 76/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6665 - accuracy: 0.7885 - val_loss: 1.2092 - val_accuracy: 0.7058 - lr: 5.0000e-04\n","Epoch 77/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6665 - accuracy: 0.7885 - val_loss: 1.2622 - val_accuracy: 0.6934 - lr: 5.0000e-04\n","Epoch 78/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6787 - accuracy: 0.7777 - val_loss: 1.2883 - val_accuracy: 0.6893 - lr: 5.0000e-04\n","Epoch 79/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6670 - accuracy: 0.7771 - val_loss: 1.2641 - val_accuracy: 0.6831 - lr: 5.0000e-04\n","Epoch 80/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6593 - accuracy: 0.7864 - val_loss: 1.3044 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 81/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6531 - accuracy: 0.7895 - val_loss: 1.3163 - val_accuracy: 0.6975 - lr: 5.0000e-04\n","Epoch 82/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6562 - accuracy: 0.7900 - val_loss: 1.3301 - val_accuracy: 0.6831 - lr: 5.0000e-04\n","Epoch 83/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6421 - accuracy: 0.7890 - val_loss: 1.3032 - val_accuracy: 0.6728 - lr: 5.0000e-04\n","Epoch 84/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6464 - accuracy: 0.7936 - val_loss: 1.3740 - val_accuracy: 0.6955 - lr: 5.0000e-04\n","Epoch 85/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6307 - accuracy: 0.8019 - val_loss: 1.4188 - val_accuracy: 0.6770 - lr: 5.0000e-04\n","Epoch 86/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6387 - accuracy: 0.7967 - val_loss: 1.3925 - val_accuracy: 0.6893 - lr: 5.0000e-04\n","Epoch 87/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6447 - accuracy: 0.7957 - val_loss: 1.5430 - val_accuracy: 0.6996 - lr: 5.0000e-04\n","Epoch 88/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6587 - accuracy: 0.7941 - val_loss: 1.2307 - val_accuracy: 0.6955 - lr: 5.0000e-04\n","Epoch 89/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6285 - accuracy: 0.8101 - val_loss: 1.3243 - val_accuracy: 0.6996 - lr: 5.0000e-04\n","Epoch 90/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6141 - accuracy: 0.8024 - val_loss: 1.5125 - val_accuracy: 0.6852 - lr: 5.0000e-04\n","Epoch 91/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6486 - accuracy: 0.7936 - val_loss: 1.3256 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 92/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5963 - accuracy: 0.8101 - val_loss: 1.3281 - val_accuracy: 0.6934 - lr: 2.5000e-04\n","Epoch 93/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5791 - accuracy: 0.8194 - val_loss: 1.3232 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 94/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5758 - accuracy: 0.8157 - val_loss: 1.3195 - val_accuracy: 0.7119 - lr: 2.5000e-04\n","Epoch 95/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5831 - accuracy: 0.8101 - val_loss: 1.3400 - val_accuracy: 0.6996 - lr: 2.5000e-04\n","Epoch 96/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5723 - accuracy: 0.8147 - val_loss: 1.3115 - val_accuracy: 0.7099 - lr: 2.5000e-04\n","Epoch 97/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5596 - accuracy: 0.8260 - val_loss: 1.3519 - val_accuracy: 0.7037 - lr: 2.5000e-04\n","Epoch 98/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5757 - accuracy: 0.8199 - val_loss: 1.3562 - val_accuracy: 0.7058 - lr: 2.5000e-04\n","Epoch 99/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5753 - accuracy: 0.8137 - val_loss: 1.3684 - val_accuracy: 0.6811 - lr: 2.5000e-04\n","Epoch 100/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5559 - accuracy: 0.8260 - val_loss: 1.3722 - val_accuracy: 0.7078 - lr: 2.5000e-04\n","Epoch 101/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5541 - accuracy: 0.8266 - val_loss: 1.3944 - val_accuracy: 0.7037 - lr: 2.5000e-04\n","Epoch 102/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5419 - accuracy: 0.8266 - val_loss: 1.3694 - val_accuracy: 0.7037 - lr: 2.5000e-04\n","Epoch 103/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5404 - accuracy: 0.8358 - val_loss: 1.4028 - val_accuracy: 0.7078 - lr: 2.5000e-04\n","Epoch 104/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5616 - accuracy: 0.8255 - val_loss: 1.4751 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 105/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5890 - accuracy: 0.8276 - val_loss: 1.3574 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 106/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5784 - accuracy: 0.8101 - val_loss: 1.3822 - val_accuracy: 0.6893 - lr: 2.5000e-04\n","Epoch 107/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5442 - accuracy: 0.8281 - val_loss: 1.3704 - val_accuracy: 0.7160 - lr: 2.5000e-04\n","Epoch 108/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5319 - accuracy: 0.8379 - val_loss: 1.3914 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 109/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5391 - accuracy: 0.8343 - val_loss: 1.3567 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 110/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5304 - accuracy: 0.8363 - val_loss: 1.3179 - val_accuracy: 0.6996 - lr: 2.5000e-04\n","Epoch 111/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5345 - accuracy: 0.8322 - val_loss: 1.3425 - val_accuracy: 0.7119 - lr: 2.5000e-04\n","Epoch 112/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5170 - accuracy: 0.8497 - val_loss: 1.3848 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 113/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5187 - accuracy: 0.8415 - val_loss: 1.4068 - val_accuracy: 0.7016 - lr: 2.5000e-04\n","Epoch 114/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5209 - accuracy: 0.8363 - val_loss: 1.4078 - val_accuracy: 0.6996 - lr: 2.5000e-04\n","Epoch 115/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5109 - accuracy: 0.8446 - val_loss: 1.3935 - val_accuracy: 0.7078 - lr: 2.5000e-04\n","Epoch 116/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5129 - accuracy: 0.8343 - val_loss: 1.4354 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 117/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5063 - accuracy: 0.8405 - val_loss: 1.4350 - val_accuracy: 0.7099 - lr: 2.5000e-04\n","Epoch 118/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5008 - accuracy: 0.8430 - val_loss: 1.4410 - val_accuracy: 0.7160 - lr: 2.5000e-04\n","Epoch 119/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5250 - accuracy: 0.8394 - val_loss: 1.5973 - val_accuracy: 0.6811 - lr: 2.5000e-04\n","Epoch 120/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5026 - accuracy: 0.8415 - val_loss: 1.4363 - val_accuracy: 0.7119 - lr: 2.5000e-04\n","Epoch 121/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4948 - accuracy: 0.8518 - val_loss: 1.5010 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 122/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4920 - accuracy: 0.8533 - val_loss: 1.5080 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 123/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4895 - accuracy: 0.8533 - val_loss: 1.4776 - val_accuracy: 0.7202 - lr: 1.2500e-04\n","Epoch 124/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4771 - accuracy: 0.8482 - val_loss: 1.4763 - val_accuracy: 0.7016 - lr: 1.2500e-04\n","Epoch 125/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4786 - accuracy: 0.8518 - val_loss: 1.4567 - val_accuracy: 0.7078 - lr: 1.2500e-04\n","Epoch 126/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4692 - accuracy: 0.8585 - val_loss: 1.5015 - val_accuracy: 0.7078 - lr: 1.2500e-04\n","Epoch 127/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4700 - accuracy: 0.8590 - val_loss: 1.5181 - val_accuracy: 0.7140 - lr: 1.2500e-04\n","Epoch 128/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4686 - accuracy: 0.8528 - val_loss: 1.5019 - val_accuracy: 0.6893 - lr: 1.2500e-04\n","Epoch 129/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4753 - accuracy: 0.8585 - val_loss: 1.5036 - val_accuracy: 0.6872 - lr: 1.2500e-04\n","Epoch 130/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4607 - accuracy: 0.8641 - val_loss: 1.5254 - val_accuracy: 0.7181 - lr: 1.2500e-04\n","Epoch 131/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4608 - accuracy: 0.8636 - val_loss: 1.5562 - val_accuracy: 0.6975 - lr: 1.2500e-04\n","Epoch 132/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4720 - accuracy: 0.8564 - val_loss: 1.5451 - val_accuracy: 0.7078 - lr: 1.2500e-04\n","Epoch 133/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4619 - accuracy: 0.8574 - val_loss: 1.5803 - val_accuracy: 0.7078 - lr: 1.2500e-04\n","Epoch 134/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4529 - accuracy: 0.8610 - val_loss: 1.5719 - val_accuracy: 0.7160 - lr: 1.2500e-04\n","Epoch 135/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4585 - accuracy: 0.8569 - val_loss: 1.5474 - val_accuracy: 0.6914 - lr: 1.2500e-04\n","Epoch 136/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4509 - accuracy: 0.8677 - val_loss: 1.5603 - val_accuracy: 0.7099 - lr: 1.2500e-04\n","Epoch 137/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4494 - accuracy: 0.8657 - val_loss: 1.5797 - val_accuracy: 0.7078 - lr: 1.2500e-04\n","Epoch 138/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4552 - accuracy: 0.8605 - val_loss: 1.5760 - val_accuracy: 0.6996 - lr: 1.2500e-04\n","Epoch 139/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4399 - accuracy: 0.8698 - val_loss: 1.5773 - val_accuracy: 0.7016 - lr: 6.2500e-05\n","Epoch 140/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4561 - accuracy: 0.8605 - val_loss: 1.5837 - val_accuracy: 0.7058 - lr: 6.2500e-05\n","Epoch 141/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4410 - accuracy: 0.8636 - val_loss: 1.6048 - val_accuracy: 0.7140 - lr: 6.2500e-05\n","Epoch 142/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4481 - accuracy: 0.8652 - val_loss: 1.5934 - val_accuracy: 0.7058 - lr: 6.2500e-05\n","Epoch 143/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4499 - accuracy: 0.8662 - val_loss: 1.5895 - val_accuracy: 0.7078 - lr: 6.2500e-05\n","Epoch 144/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4494 - accuracy: 0.8621 - val_loss: 1.5822 - val_accuracy: 0.7140 - lr: 6.2500e-05\n","Epoch 145/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4485 - accuracy: 0.8646 - val_loss: 1.5780 - val_accuracy: 0.7160 - lr: 6.2500e-05\n","Epoch 146/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4364 - accuracy: 0.8688 - val_loss: 1.5783 - val_accuracy: 0.7202 - lr: 6.2500e-05\n","Epoch 147/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4309 - accuracy: 0.8636 - val_loss: 1.5912 - val_accuracy: 0.7284 - lr: 6.2500e-05\n","Epoch 148/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4435 - accuracy: 0.8652 - val_loss: 1.5860 - val_accuracy: 0.7181 - lr: 6.2500e-05\n","Epoch 149/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4356 - accuracy: 0.8713 - val_loss: 1.6060 - val_accuracy: 0.7058 - lr: 6.2500e-05\n","Epoch 150/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4239 - accuracy: 0.8749 - val_loss: 1.6042 - val_accuracy: 0.7222 - lr: 6.2500e-05\n","Epoch 151/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4249 - accuracy: 0.8724 - val_loss: 1.6085 - val_accuracy: 0.7099 - lr: 6.2500e-05\n","Epoch 152/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4348 - accuracy: 0.8672 - val_loss: 1.6311 - val_accuracy: 0.7099 - lr: 6.2500e-05\n","Epoch 153/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4349 - accuracy: 0.8698 - val_loss: 1.6237 - val_accuracy: 0.7181 - lr: 6.2500e-05\n","Epoch 154/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4307 - accuracy: 0.8729 - val_loss: 1.6072 - val_accuracy: 0.7078 - lr: 6.2500e-05\n","Epoch 155/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4365 - accuracy: 0.8641 - val_loss: 1.6255 - val_accuracy: 0.7160 - lr: 6.2500e-05\n","Epoch 156/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4342 - accuracy: 0.8688 - val_loss: 1.6139 - val_accuracy: 0.6975 - lr: 6.2500e-05\n","Epoch 157/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4342 - accuracy: 0.8693 - val_loss: 1.6234 - val_accuracy: 0.7078 - lr: 6.2500e-05\n","Epoch 158/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4295 - accuracy: 0.8646 - val_loss: 1.6501 - val_accuracy: 0.7119 - lr: 6.2500e-05\n","Epoch 159/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4270 - accuracy: 0.8713 - val_loss: 1.6514 - val_accuracy: 0.7140 - lr: 6.2500e-05\n","Epoch 160/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4278 - accuracy: 0.8816 - val_loss: 1.6386 - val_accuracy: 0.7222 - lr: 6.2500e-05\n","Epoch 161/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4251 - accuracy: 0.8760 - val_loss: 1.6498 - val_accuracy: 0.6975 - lr: 6.2500e-05\n","Epoch 162/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4282 - accuracy: 0.8760 - val_loss: 1.6809 - val_accuracy: 0.7305 - lr: 6.2500e-05\n","Epoch 163/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4266 - accuracy: 0.8755 - val_loss: 1.6565 - val_accuracy: 0.6996 - lr: 6.2500e-05\n","Epoch 164/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4218 - accuracy: 0.8734 - val_loss: 1.6688 - val_accuracy: 0.7181 - lr: 6.2500e-05\n","Epoch 165/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4253 - accuracy: 0.8729 - val_loss: 1.6496 - val_accuracy: 0.7099 - lr: 6.2500e-05\n","Epoch 166/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4213 - accuracy: 0.8775 - val_loss: 1.6824 - val_accuracy: 0.7119 - lr: 6.2500e-05\n","Epoch 167/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4272 - accuracy: 0.8693 - val_loss: 1.6824 - val_accuracy: 0.7140 - lr: 6.2500e-05\n","Epoch 168/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4498 - accuracy: 0.8729 - val_loss: 1.5262 - val_accuracy: 0.7119 - lr: 6.2500e-05\n","Epoch 169/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4285 - accuracy: 0.8682 - val_loss: 1.5409 - val_accuracy: 0.7119 - lr: 6.2500e-05\n","Epoch 170/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4248 - accuracy: 0.8744 - val_loss: 1.5726 - val_accuracy: 0.7119 - lr: 6.2500e-05\n","Epoch 171/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4172 - accuracy: 0.8744 - val_loss: 1.5856 - val_accuracy: 0.7202 - lr: 6.2500e-05\n","Epoch 172/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4177 - accuracy: 0.8796 - val_loss: 1.5998 - val_accuracy: 0.7078 - lr: 6.2500e-05\n","Epoch 173/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4041 - accuracy: 0.8760 - val_loss: 1.6109 - val_accuracy: 0.7099 - lr: 6.2500e-05\n","Epoch 174/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4107 - accuracy: 0.8791 - val_loss: 1.6377 - val_accuracy: 0.7181 - lr: 6.2500e-05\n","Epoch 175/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4142 - accuracy: 0.8796 - val_loss: 1.6217 - val_accuracy: 0.7181 - lr: 6.2500e-05\n","Epoch 176/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4058 - accuracy: 0.8837 - val_loss: 1.6095 - val_accuracy: 0.7078 - lr: 6.2500e-05\n","Epoch 177/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4171 - accuracy: 0.8744 - val_loss: 1.6450 - val_accuracy: 0.7284 - lr: 6.2500e-05\n","Epoch 178/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4175 - accuracy: 0.8780 - val_loss: 1.6306 - val_accuracy: 0.7140 - lr: 3.1250e-05\n","Epoch 179/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4025 - accuracy: 0.8837 - val_loss: 1.6364 - val_accuracy: 0.7016 - lr: 3.1250e-05\n","Epoch 180/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4172 - accuracy: 0.8718 - val_loss: 1.6369 - val_accuracy: 0.7099 - lr: 3.1250e-05\n","Epoch 181/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4181 - accuracy: 0.8703 - val_loss: 1.6440 - val_accuracy: 0.7222 - lr: 3.1250e-05\n","Epoch 182/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4020 - accuracy: 0.8791 - val_loss: 1.6351 - val_accuracy: 0.7099 - lr: 3.1250e-05\n","Epoch 183/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4050 - accuracy: 0.8801 - val_loss: 1.6412 - val_accuracy: 0.7058 - lr: 3.1250e-05\n","Epoch 184/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4136 - accuracy: 0.8775 - val_loss: 1.6462 - val_accuracy: 0.7099 - lr: 3.1250e-05\n","Epoch 185/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4082 - accuracy: 0.8801 - val_loss: 1.6494 - val_accuracy: 0.7016 - lr: 3.1250e-05\n","Epoch 186/200\n","31/31 [==============================] - 0s 11ms/step - loss: 0.4036 - accuracy: 0.8821 - val_loss: 1.6506 - val_accuracy: 0.7037 - lr: 3.1250e-05\n","Epoch 187/200\n","31/31 [==============================] - 0s 13ms/step - loss: 0.4144 - accuracy: 0.8749 - val_loss: 1.6457 - val_accuracy: 0.7058 - lr: 3.1250e-05\n","Epoch 188/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4094 - accuracy: 0.8780 - val_loss: 1.6583 - val_accuracy: 0.7160 - lr: 3.1250e-05\n","Epoch 189/200\n","31/31 [==============================] - 0s 12ms/step - loss: 0.4075 - accuracy: 0.8770 - val_loss: 1.6561 - val_accuracy: 0.7119 - lr: 3.1250e-05\n","Epoch 190/200\n","31/31 [==============================] - 0s 12ms/step - loss: 0.3998 - accuracy: 0.8806 - val_loss: 1.6553 - val_accuracy: 0.7078 - lr: 3.1250e-05\n","Epoch 191/200\n","31/31 [==============================] - 0s 12ms/step - loss: 0.4024 - accuracy: 0.8821 - val_loss: 1.6536 - val_accuracy: 0.7078 - lr: 3.1250e-05\n","Epoch 192/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4095 - accuracy: 0.8780 - val_loss: 1.6617 - val_accuracy: 0.7119 - lr: 3.1250e-05\n"]}]},{"cell_type":"code","source":["history3 = model3.fit(\n","    x = x_train3,\n","    y = y_train3,\n","    batch_size = 64,\n","    epochs = 200,\n","    validation_data = (x_test3, y_test3),\n","    callbacks = [\n","        tfk.callbacks.EarlyStopping(monitor='val_accuracy', mode='max', patience=30, restore_best_weights=True),\n","        tfk.callbacks.ReduceLROnPlateau(monitor='val_accuracy', mode='max', patience=15, factor=0.5, min_lr=1e-5)\n","    ]\n",").history"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"iICwi6igwOxQ","executionInfo":{"status":"ok","timestamp":1671208518800,"user_tz":-60,"elapsed":39085,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"c83e6934-9dd0-42d3-d488-ce79f7adf710"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/200\n","31/31 [==============================] - 2s 19ms/step - loss: 2.2787 - accuracy: 0.3299 - val_loss: 2.0211 - val_accuracy: 0.3539 - lr: 0.0010\n","Epoch 2/200\n","31/31 [==============================] - 0s 12ms/step - loss: 1.9396 - accuracy: 0.3736 - val_loss: 1.8829 - val_accuracy: 0.3765 - lr: 0.0010\n","Epoch 3/200\n","31/31 [==============================] - 0s 12ms/step - loss: 1.8369 - accuracy: 0.3937 - val_loss: 1.8117 - val_accuracy: 0.4095 - lr: 0.0010\n","Epoch 4/200\n","31/31 [==============================] - 0s 12ms/step - loss: 1.7762 - accuracy: 0.4179 - val_loss: 1.7198 - val_accuracy: 0.4177 - lr: 0.0010\n","Epoch 5/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.6125 - accuracy: 0.4550 - val_loss: 1.7624 - val_accuracy: 0.4300 - lr: 0.0010\n","Epoch 6/200\n","31/31 [==============================] - 0s 12ms/step - loss: 1.5484 - accuracy: 0.4915 - val_loss: 1.6079 - val_accuracy: 0.4444 - lr: 0.0010\n","Epoch 7/200\n","31/31 [==============================] - 0s 12ms/step - loss: 1.5596 - accuracy: 0.4756 - val_loss: 1.6104 - val_accuracy: 0.4568 - lr: 0.0010\n","Epoch 8/200\n","31/31 [==============================] - 0s 12ms/step - loss: 1.4842 - accuracy: 0.5090 - val_loss: 1.4931 - val_accuracy: 0.4671 - lr: 0.0010\n","Epoch 9/200\n","31/31 [==============================] - 0s 12ms/step - loss: 1.4160 - accuracy: 0.5270 - val_loss: 1.4202 - val_accuracy: 0.5206 - lr: 0.0010\n","Epoch 10/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3624 - accuracy: 0.5461 - val_loss: 1.4684 - val_accuracy: 0.5082 - lr: 0.0010\n","Epoch 11/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.3603 - accuracy: 0.5512 - val_loss: 1.4131 - val_accuracy: 0.4877 - lr: 0.0010\n","Epoch 12/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2910 - accuracy: 0.5548 - val_loss: 1.4540 - val_accuracy: 0.5226 - lr: 0.0010\n","Epoch 13/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2821 - accuracy: 0.5764 - val_loss: 1.4895 - val_accuracy: 0.5412 - lr: 0.0010\n","Epoch 14/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.3395 - accuracy: 0.5569 - val_loss: 1.3083 - val_accuracy: 0.5514 - lr: 0.0010\n","Epoch 15/200\n","31/31 [==============================] - 0s 12ms/step - loss: 1.2312 - accuracy: 0.6063 - val_loss: 1.3308 - val_accuracy: 0.5350 - lr: 0.0010\n","Epoch 16/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2155 - accuracy: 0.5852 - val_loss: 1.4169 - val_accuracy: 0.5288 - lr: 0.0010\n","Epoch 17/200\n","31/31 [==============================] - 0s 12ms/step - loss: 1.1738 - accuracy: 0.6125 - val_loss: 1.3489 - val_accuracy: 0.5432 - lr: 0.0010\n","Epoch 18/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.2230 - accuracy: 0.6001 - val_loss: 1.2578 - val_accuracy: 0.5576 - lr: 0.0010\n","Epoch 19/200\n","31/31 [==============================] - 0s 12ms/step - loss: 1.1436 - accuracy: 0.6176 - val_loss: 1.3156 - val_accuracy: 0.5597 - lr: 0.0010\n","Epoch 20/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.1964 - accuracy: 0.6016 - val_loss: 1.3749 - val_accuracy: 0.5535 - lr: 0.0010\n","Epoch 21/200\n","31/31 [==============================] - 0s 11ms/step - loss: 1.1161 - accuracy: 0.6341 - val_loss: 1.2146 - val_accuracy: 0.5720 - lr: 0.0010\n","Epoch 22/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0852 - accuracy: 0.6413 - val_loss: 1.1650 - val_accuracy: 0.5988 - lr: 0.0010\n","Epoch 23/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0554 - accuracy: 0.6469 - val_loss: 1.2371 - val_accuracy: 0.5617 - lr: 0.0010\n","Epoch 24/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0756 - accuracy: 0.6562 - val_loss: 1.2114 - val_accuracy: 0.5782 - lr: 0.0010\n","Epoch 25/200\n","31/31 [==============================] - 0s 11ms/step - loss: 1.0380 - accuracy: 0.6624 - val_loss: 1.3383 - val_accuracy: 0.5761 - lr: 0.0010\n","Epoch 26/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0446 - accuracy: 0.6572 - val_loss: 1.2477 - val_accuracy: 0.5823 - lr: 0.0010\n","Epoch 27/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0060 - accuracy: 0.6722 - val_loss: 1.1302 - val_accuracy: 0.6111 - lr: 0.0010\n","Epoch 28/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0070 - accuracy: 0.6716 - val_loss: 1.1214 - val_accuracy: 0.6276 - lr: 0.0010\n","Epoch 29/200\n","31/31 [==============================] - 0s 11ms/step - loss: 0.9706 - accuracy: 0.6804 - val_loss: 1.1267 - val_accuracy: 0.6152 - lr: 0.0010\n","Epoch 30/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9602 - accuracy: 0.6809 - val_loss: 1.1389 - val_accuracy: 0.6173 - lr: 0.0010\n","Epoch 31/200\n","31/31 [==============================] - 0s 13ms/step - loss: 1.0182 - accuracy: 0.6598 - val_loss: 2.0135 - val_accuracy: 0.5802 - lr: 0.0010\n","Epoch 32/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1361 - accuracy: 0.6372 - val_loss: 1.8610 - val_accuracy: 0.5700 - lr: 0.0010\n","Epoch 33/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.1979 - accuracy: 0.6233 - val_loss: 1.3618 - val_accuracy: 0.5926 - lr: 0.0010\n","Epoch 34/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0166 - accuracy: 0.6593 - val_loss: 1.1800 - val_accuracy: 0.6008 - lr: 0.0010\n","Epoch 35/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9476 - accuracy: 0.6819 - val_loss: 1.3553 - val_accuracy: 0.6070 - lr: 0.0010\n","Epoch 36/200\n","31/31 [==============================] - 0s 12ms/step - loss: 0.9442 - accuracy: 0.6840 - val_loss: 1.1793 - val_accuracy: 0.6317 - lr: 0.0010\n","Epoch 37/200\n","31/31 [==============================] - 0s 11ms/step - loss: 0.9299 - accuracy: 0.6866 - val_loss: 1.0976 - val_accuracy: 0.6193 - lr: 0.0010\n","Epoch 38/200\n","31/31 [==============================] - 0s 13ms/step - loss: 0.9515 - accuracy: 0.6876 - val_loss: 1.1377 - val_accuracy: 0.6049 - lr: 0.0010\n","Epoch 39/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9251 - accuracy: 0.6850 - val_loss: 1.0971 - val_accuracy: 0.6255 - lr: 0.0010\n","Epoch 40/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8933 - accuracy: 0.6989 - val_loss: 1.1221 - val_accuracy: 0.6111 - lr: 0.0010\n","Epoch 41/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8695 - accuracy: 0.7072 - val_loss: 1.0892 - val_accuracy: 0.6317 - lr: 0.0010\n","Epoch 42/200\n","31/31 [==============================] - 0s 11ms/step - loss: 0.9213 - accuracy: 0.6958 - val_loss: 1.4946 - val_accuracy: 0.6214 - lr: 0.0010\n","Epoch 43/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8843 - accuracy: 0.7020 - val_loss: 1.1116 - val_accuracy: 0.6235 - lr: 0.0010\n","Epoch 44/200\n","31/31 [==============================] - 0s 12ms/step - loss: 0.8715 - accuracy: 0.7087 - val_loss: 1.0762 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 45/200\n","31/31 [==============================] - 0s 11ms/step - loss: 0.8612 - accuracy: 0.7180 - val_loss: 1.0850 - val_accuracy: 0.6255 - lr: 0.0010\n","Epoch 46/200\n","31/31 [==============================] - 0s 15ms/step - loss: 0.8458 - accuracy: 0.7195 - val_loss: 1.0969 - val_accuracy: 0.6235 - lr: 0.0010\n","Epoch 47/200\n","31/31 [==============================] - 0s 14ms/step - loss: 0.8721 - accuracy: 0.7159 - val_loss: 1.0793 - val_accuracy: 0.6317 - lr: 0.0010\n","Epoch 48/200\n","31/31 [==============================] - 0s 12ms/step - loss: 0.8688 - accuracy: 0.7190 - val_loss: 1.1119 - val_accuracy: 0.6379 - lr: 0.0010\n","Epoch 49/200\n","31/31 [==============================] - 0s 11ms/step - loss: 0.8182 - accuracy: 0.7349 - val_loss: 1.0668 - val_accuracy: 0.6379 - lr: 0.0010\n","Epoch 50/200\n","31/31 [==============================] - 0s 15ms/step - loss: 0.8051 - accuracy: 0.7385 - val_loss: 1.0529 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 51/200\n","31/31 [==============================] - 0s 13ms/step - loss: 0.8776 - accuracy: 0.7159 - val_loss: 1.1437 - val_accuracy: 0.6214 - lr: 0.0010\n","Epoch 52/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8187 - accuracy: 0.7262 - val_loss: 1.1659 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 53/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8073 - accuracy: 0.7370 - val_loss: 1.0703 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 54/200\n","31/31 [==============================] - 0s 11ms/step - loss: 0.8357 - accuracy: 0.7313 - val_loss: 1.8739 - val_accuracy: 0.6193 - lr: 0.0010\n","Epoch 55/200\n","31/31 [==============================] - 0s 14ms/step - loss: 0.8758 - accuracy: 0.7298 - val_loss: 1.0470 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 56/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7838 - accuracy: 0.7509 - val_loss: 1.0967 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 57/200\n","31/31 [==============================] - 0s 13ms/step - loss: 0.7521 - accuracy: 0.7519 - val_loss: 1.0525 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 58/200\n","31/31 [==============================] - 0s 13ms/step - loss: 0.7321 - accuracy: 0.7607 - val_loss: 1.0535 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 59/200\n","31/31 [==============================] - 0s 14ms/step - loss: 0.7687 - accuracy: 0.7463 - val_loss: 1.0699 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 60/200\n","31/31 [==============================] - 0s 12ms/step - loss: 0.7681 - accuracy: 0.7432 - val_loss: 1.0779 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 61/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7370 - accuracy: 0.7658 - val_loss: 1.0741 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 62/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6963 - accuracy: 0.7782 - val_loss: 1.0090 - val_accuracy: 0.6811 - lr: 0.0010\n","Epoch 63/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7020 - accuracy: 0.7617 - val_loss: 1.0909 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 64/200\n","31/31 [==============================] - 0s 16ms/step - loss: 0.7742 - accuracy: 0.7468 - val_loss: 1.0938 - val_accuracy: 0.6852 - lr: 0.0010\n","Epoch 65/200\n","31/31 [==============================] - 0s 12ms/step - loss: 0.7244 - accuracy: 0.7638 - val_loss: 1.0900 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 66/200\n","31/31 [==============================] - 0s 13ms/step - loss: 0.7124 - accuracy: 0.7710 - val_loss: 1.0284 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 67/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7554 - accuracy: 0.7612 - val_loss: 1.0952 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 68/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7541 - accuracy: 0.7555 - val_loss: 1.0620 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 69/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7038 - accuracy: 0.7653 - val_loss: 1.0269 - val_accuracy: 0.6811 - lr: 0.0010\n","Epoch 70/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6615 - accuracy: 0.7890 - val_loss: 1.1732 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 71/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6881 - accuracy: 0.7787 - val_loss: 1.2547 - val_accuracy: 0.6811 - lr: 0.0010\n","Epoch 72/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6639 - accuracy: 0.7813 - val_loss: 1.0351 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 73/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6388 - accuracy: 0.7941 - val_loss: 1.0497 - val_accuracy: 0.6975 - lr: 0.0010\n","Epoch 74/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6337 - accuracy: 0.7998 - val_loss: 1.1542 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 75/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6409 - accuracy: 0.7900 - val_loss: 1.2091 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 76/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6103 - accuracy: 0.7993 - val_loss: 1.1540 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 77/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7032 - accuracy: 0.7684 - val_loss: 1.1655 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 78/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7369 - accuracy: 0.7684 - val_loss: 1.1487 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 79/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7592 - accuracy: 0.7633 - val_loss: 1.1347 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 80/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6916 - accuracy: 0.7844 - val_loss: 1.2550 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 81/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6233 - accuracy: 0.8008 - val_loss: 1.1439 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 82/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6060 - accuracy: 0.8044 - val_loss: 1.1187 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 83/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6081 - accuracy: 0.8049 - val_loss: 1.0710 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 84/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6201 - accuracy: 0.8029 - val_loss: 1.0899 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 85/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5740 - accuracy: 0.8142 - val_loss: 1.1484 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 86/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5508 - accuracy: 0.8224 - val_loss: 1.1289 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 87/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5559 - accuracy: 0.8183 - val_loss: 1.2123 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 88/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6104 - accuracy: 0.8044 - val_loss: 1.1310 - val_accuracy: 0.7016 - lr: 0.0010\n","Epoch 89/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5261 - accuracy: 0.8307 - val_loss: 1.1742 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 90/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6736 - accuracy: 0.8055 - val_loss: 1.5135 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 91/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.4462 - accuracy: 0.6686 - val_loss: 1.3352 - val_accuracy: 0.6091 - lr: 0.0010\n","Epoch 92/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8036 - accuracy: 0.7396 - val_loss: 1.1812 - val_accuracy: 0.6337 - lr: 0.0010\n","Epoch 93/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6659 - accuracy: 0.7957 - val_loss: 1.1631 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 94/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6314 - accuracy: 0.7941 - val_loss: 1.1857 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 95/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6003 - accuracy: 0.8096 - val_loss: 1.1366 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 96/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5842 - accuracy: 0.8168 - val_loss: 1.2439 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 97/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5786 - accuracy: 0.8183 - val_loss: 1.1900 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 98/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5423 - accuracy: 0.8240 - val_loss: 1.1722 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 99/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5384 - accuracy: 0.8302 - val_loss: 1.1438 - val_accuracy: 0.6811 - lr: 0.0010\n","Epoch 100/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5336 - accuracy: 0.8369 - val_loss: 1.2306 - val_accuracy: 0.6811 - lr: 0.0010\n","Epoch 101/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5345 - accuracy: 0.8384 - val_loss: 1.3418 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 102/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5243 - accuracy: 0.8430 - val_loss: 1.3233 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 103/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5421 - accuracy: 0.8245 - val_loss: 1.2276 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 104/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4748 - accuracy: 0.8518 - val_loss: 1.2022 - val_accuracy: 0.6749 - lr: 5.0000e-04\n","Epoch 105/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4483 - accuracy: 0.8688 - val_loss: 1.1885 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 106/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4301 - accuracy: 0.8734 - val_loss: 1.2077 - val_accuracy: 0.6728 - lr: 5.0000e-04\n","Epoch 107/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4225 - accuracy: 0.8739 - val_loss: 1.2242 - val_accuracy: 0.6893 - lr: 5.0000e-04\n","Epoch 108/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4227 - accuracy: 0.8780 - val_loss: 1.2857 - val_accuracy: 0.6852 - lr: 5.0000e-04\n","Epoch 109/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4191 - accuracy: 0.8811 - val_loss: 1.2781 - val_accuracy: 0.6790 - lr: 5.0000e-04\n","Epoch 110/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4111 - accuracy: 0.8780 - val_loss: 1.2696 - val_accuracy: 0.6687 - lr: 5.0000e-04\n","Epoch 111/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4159 - accuracy: 0.8682 - val_loss: 1.2680 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 112/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4097 - accuracy: 0.8749 - val_loss: 1.3259 - val_accuracy: 0.6790 - lr: 5.0000e-04\n","Epoch 113/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4088 - accuracy: 0.8775 - val_loss: 1.2647 - val_accuracy: 0.6790 - lr: 5.0000e-04\n","Epoch 114/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4077 - accuracy: 0.8821 - val_loss: 1.3165 - val_accuracy: 0.6626 - lr: 5.0000e-04\n","Epoch 115/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3902 - accuracy: 0.8935 - val_loss: 1.3319 - val_accuracy: 0.6770 - lr: 5.0000e-04\n","Epoch 116/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3908 - accuracy: 0.8863 - val_loss: 1.3728 - val_accuracy: 0.6728 - lr: 5.0000e-04\n","Epoch 117/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4172 - accuracy: 0.8760 - val_loss: 1.2974 - val_accuracy: 0.6852 - lr: 5.0000e-04\n","Epoch 118/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3789 - accuracy: 0.8883 - val_loss: 1.3125 - val_accuracy: 0.6687 - lr: 5.0000e-04\n"]}]},{"cell_type":"code","source":["history4 = model4.fit(\n","    x = x_train4,\n","    y = y_train4,\n","    batch_size = 64,\n","    epochs = 200,\n","    validation_data = (x_test4, y_test4),\n","    callbacks = [\n","        tfk.callbacks.EarlyStopping(monitor='val_accuracy', mode='max', patience=30, restore_best_weights=True),\n","        tfk.callbacks.ReduceLROnPlateau(monitor='val_accuracy', mode='max', patience=15, factor=0.5, min_lr=1e-5)\n","    ]\n",").history"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"4B69wzwBwOtE","executionInfo":{"status":"ok","timestamp":1671208559181,"user_tz":-60,"elapsed":40408,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"f67de47b-b9ef-4244-adf5-d0cd5de1826a"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/200\n","31/31 [==============================] - 1s 18ms/step - loss: 2.2454 - accuracy: 0.3356 - val_loss: 1.9698 - val_accuracy: 0.3621 - lr: 0.0010\n","Epoch 2/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.9079 - accuracy: 0.3865 - val_loss: 1.8393 - val_accuracy: 0.4259 - lr: 0.0010\n","Epoch 3/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.8374 - accuracy: 0.3978 - val_loss: 1.7651 - val_accuracy: 0.4198 - lr: 0.0010\n","Epoch 4/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.7410 - accuracy: 0.4287 - val_loss: 1.7971 - val_accuracy: 0.4362 - lr: 0.0010\n","Epoch 5/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.7036 - accuracy: 0.4385 - val_loss: 1.6387 - val_accuracy: 0.4733 - lr: 0.0010\n","Epoch 6/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.5804 - accuracy: 0.4776 - val_loss: 1.6367 - val_accuracy: 0.4691 - lr: 0.0010\n","Epoch 7/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.4863 - accuracy: 0.4925 - val_loss: 1.4634 - val_accuracy: 0.5082 - lr: 0.0010\n","Epoch 8/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.4684 - accuracy: 0.4889 - val_loss: 1.4959 - val_accuracy: 0.4938 - lr: 0.0010\n","Epoch 9/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.4206 - accuracy: 0.5234 - val_loss: 1.4717 - val_accuracy: 0.5494 - lr: 0.0010\n","Epoch 10/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.3448 - accuracy: 0.5306 - val_loss: 1.5436 - val_accuracy: 0.5267 - lr: 0.0010\n","Epoch 11/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3646 - accuracy: 0.5353 - val_loss: 1.4499 - val_accuracy: 0.5021 - lr: 0.0010\n","Epoch 12/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3224 - accuracy: 0.5553 - val_loss: 1.3342 - val_accuracy: 0.5679 - lr: 0.0010\n","Epoch 13/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.2942 - accuracy: 0.5656 - val_loss: 1.3717 - val_accuracy: 0.5741 - lr: 0.0010\n","Epoch 14/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2127 - accuracy: 0.5965 - val_loss: 1.3536 - val_accuracy: 0.5638 - lr: 0.0010\n","Epoch 15/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.2019 - accuracy: 0.5919 - val_loss: 1.3592 - val_accuracy: 0.5391 - lr: 0.0010\n","Epoch 16/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2016 - accuracy: 0.6063 - val_loss: 1.3235 - val_accuracy: 0.5864 - lr: 0.0010\n","Epoch 17/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.1784 - accuracy: 0.6047 - val_loss: 1.2559 - val_accuracy: 0.6049 - lr: 0.0010\n","Epoch 18/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1195 - accuracy: 0.6320 - val_loss: 1.2394 - val_accuracy: 0.6091 - lr: 0.0010\n","Epoch 19/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1476 - accuracy: 0.6289 - val_loss: 1.3421 - val_accuracy: 0.5844 - lr: 0.0010\n","Epoch 20/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1462 - accuracy: 0.6222 - val_loss: 1.1658 - val_accuracy: 0.6296 - lr: 0.0010\n","Epoch 21/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0634 - accuracy: 0.6562 - val_loss: 1.2596 - val_accuracy: 0.6091 - lr: 0.0010\n","Epoch 22/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0436 - accuracy: 0.6562 - val_loss: 1.2598 - val_accuracy: 0.6337 - lr: 0.0010\n","Epoch 23/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0568 - accuracy: 0.6629 - val_loss: 1.1267 - val_accuracy: 0.6461 - lr: 0.0010\n","Epoch 24/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0356 - accuracy: 0.6593 - val_loss: 1.2489 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 25/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0132 - accuracy: 0.6680 - val_loss: 1.1949 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 26/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0307 - accuracy: 0.6526 - val_loss: 1.1735 - val_accuracy: 0.6276 - lr: 0.0010\n","Epoch 27/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0056 - accuracy: 0.6655 - val_loss: 1.2479 - val_accuracy: 0.6379 - lr: 0.0010\n","Epoch 28/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0688 - accuracy: 0.6480 - val_loss: 1.3390 - val_accuracy: 0.5844 - lr: 0.0010\n","Epoch 29/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0374 - accuracy: 0.6619 - val_loss: 1.2281 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 30/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9524 - accuracy: 0.6783 - val_loss: 1.1269 - val_accuracy: 0.6399 - lr: 0.0010\n","Epoch 31/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9277 - accuracy: 0.6969 - val_loss: 1.1224 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 32/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9126 - accuracy: 0.6886 - val_loss: 1.2370 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 33/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9298 - accuracy: 0.6871 - val_loss: 1.2091 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 34/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9179 - accuracy: 0.6984 - val_loss: 1.1910 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 35/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8960 - accuracy: 0.7010 - val_loss: 1.1694 - val_accuracy: 0.6255 - lr: 0.0010\n","Epoch 36/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8962 - accuracy: 0.6933 - val_loss: 1.1131 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 37/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9011 - accuracy: 0.7072 - val_loss: 1.6958 - val_accuracy: 0.5885 - lr: 0.0010\n","Epoch 38/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9909 - accuracy: 0.6809 - val_loss: 1.1620 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 39/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9005 - accuracy: 0.7005 - val_loss: 1.1179 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 40/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8666 - accuracy: 0.7077 - val_loss: 1.1238 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 41/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8874 - accuracy: 0.7097 - val_loss: 1.1373 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 42/200\n","31/31 [==============================] - 0s 13ms/step - loss: 0.9294 - accuracy: 0.6969 - val_loss: 1.2057 - val_accuracy: 0.6337 - lr: 0.0010\n","Epoch 43/200\n","31/31 [==============================] - 0s 15ms/step - loss: 0.9254 - accuracy: 0.6989 - val_loss: 1.1788 - val_accuracy: 0.6440 - lr: 0.0010\n","Epoch 44/200\n","31/31 [==============================] - 0s 11ms/step - loss: 0.8869 - accuracy: 0.7082 - val_loss: 1.1580 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 45/200\n","31/31 [==============================] - 0s 12ms/step - loss: 0.8574 - accuracy: 0.7231 - val_loss: 1.0926 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 46/200\n","31/31 [==============================] - 0s 14ms/step - loss: 0.8338 - accuracy: 0.7319 - val_loss: 1.0493 - val_accuracy: 0.6914 - lr: 0.0010\n","Epoch 47/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8014 - accuracy: 0.7396 - val_loss: 1.1411 - val_accuracy: 0.6379 - lr: 0.0010\n","Epoch 48/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8060 - accuracy: 0.7334 - val_loss: 1.2280 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 49/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8209 - accuracy: 0.7308 - val_loss: 1.0748 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 50/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7962 - accuracy: 0.7437 - val_loss: 1.1797 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 51/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7857 - accuracy: 0.7391 - val_loss: 1.1768 - val_accuracy: 0.6296 - lr: 0.0010\n","Epoch 52/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8081 - accuracy: 0.7324 - val_loss: 1.2321 - val_accuracy: 0.6440 - lr: 0.0010\n","Epoch 53/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7688 - accuracy: 0.7452 - val_loss: 1.0926 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 54/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7301 - accuracy: 0.7555 - val_loss: 1.0531 - val_accuracy: 0.6852 - lr: 0.0010\n","Epoch 55/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7332 - accuracy: 0.7663 - val_loss: 1.1176 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 56/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7298 - accuracy: 0.7607 - val_loss: 1.1264 - val_accuracy: 0.6852 - lr: 0.0010\n","Epoch 57/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7376 - accuracy: 0.7607 - val_loss: 1.0872 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 58/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7349 - accuracy: 0.7509 - val_loss: 1.1177 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 59/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7393 - accuracy: 0.7560 - val_loss: 1.1410 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 60/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7630 - accuracy: 0.7478 - val_loss: 1.0711 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 61/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7088 - accuracy: 0.7694 - val_loss: 1.1051 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 62/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7087 - accuracy: 0.7689 - val_loss: 1.0576 - val_accuracy: 0.6934 - lr: 0.0010\n","Epoch 63/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7061 - accuracy: 0.7771 - val_loss: 1.0577 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 64/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6631 - accuracy: 0.7844 - val_loss: 1.1005 - val_accuracy: 0.6975 - lr: 0.0010\n","Epoch 65/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6628 - accuracy: 0.7890 - val_loss: 1.0854 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 66/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6803 - accuracy: 0.7890 - val_loss: 1.1647 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 67/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8040 - accuracy: 0.7658 - val_loss: 1.0591 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 68/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7186 - accuracy: 0.7679 - val_loss: 1.6191 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 69/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7067 - accuracy: 0.7782 - val_loss: 1.0110 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 70/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6548 - accuracy: 0.7844 - val_loss: 1.0826 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 71/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6525 - accuracy: 0.7946 - val_loss: 1.1075 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 72/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7837 - accuracy: 0.7514 - val_loss: 1.0704 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 73/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6865 - accuracy: 0.7741 - val_loss: 1.2454 - val_accuracy: 0.7016 - lr: 0.0010\n","Epoch 74/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6471 - accuracy: 0.7926 - val_loss: 1.1014 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 75/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6523 - accuracy: 0.7900 - val_loss: 2.7664 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 76/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7398 - accuracy: 0.7566 - val_loss: 1.1024 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 77/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6875 - accuracy: 0.7802 - val_loss: 1.0254 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 78/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5903 - accuracy: 0.8214 - val_loss: 1.0957 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 79/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5997 - accuracy: 0.8132 - val_loss: 1.0656 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 80/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5767 - accuracy: 0.8204 - val_loss: 1.2322 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 81/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5799 - accuracy: 0.8152 - val_loss: 1.1198 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 82/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5611 - accuracy: 0.8296 - val_loss: 1.2123 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 83/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6152 - accuracy: 0.8178 - val_loss: 1.4359 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 84/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6664 - accuracy: 0.7859 - val_loss: 1.3806 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 85/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6438 - accuracy: 0.7946 - val_loss: 1.1376 - val_accuracy: 0.6914 - lr: 0.0010\n","Epoch 86/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5710 - accuracy: 0.8173 - val_loss: 1.1410 - val_accuracy: 0.6975 - lr: 0.0010\n","Epoch 87/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5454 - accuracy: 0.8266 - val_loss: 1.1522 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 88/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5476 - accuracy: 0.8338 - val_loss: 1.1801 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 89/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5652 - accuracy: 0.8250 - val_loss: 1.1487 - val_accuracy: 0.6996 - lr: 5.0000e-04\n","Epoch 90/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5099 - accuracy: 0.8456 - val_loss: 1.1364 - val_accuracy: 0.6955 - lr: 5.0000e-04\n","Epoch 91/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5015 - accuracy: 0.8451 - val_loss: 1.1529 - val_accuracy: 0.6790 - lr: 5.0000e-04\n","Epoch 92/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4786 - accuracy: 0.8600 - val_loss: 1.1651 - val_accuracy: 0.7016 - lr: 5.0000e-04\n","Epoch 93/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4741 - accuracy: 0.8502 - val_loss: 1.1646 - val_accuracy: 0.6955 - lr: 5.0000e-04\n","Epoch 94/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5208 - accuracy: 0.8399 - val_loss: 1.2293 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 95/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4733 - accuracy: 0.8538 - val_loss: 1.1539 - val_accuracy: 0.6831 - lr: 5.0000e-04\n","Epoch 96/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4548 - accuracy: 0.8641 - val_loss: 1.2078 - val_accuracy: 0.7058 - lr: 5.0000e-04\n","Epoch 97/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4408 - accuracy: 0.8688 - val_loss: 1.2020 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 98/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4468 - accuracy: 0.8693 - val_loss: 1.2048 - val_accuracy: 0.6893 - lr: 5.0000e-04\n","Epoch 99/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4305 - accuracy: 0.8718 - val_loss: 1.2083 - val_accuracy: 0.6955 - lr: 5.0000e-04\n","Epoch 100/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4288 - accuracy: 0.8749 - val_loss: 1.2049 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 101/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4495 - accuracy: 0.8703 - val_loss: 1.2424 - val_accuracy: 0.7058 - lr: 5.0000e-04\n","Epoch 102/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4647 - accuracy: 0.8600 - val_loss: 1.3269 - val_accuracy: 0.7016 - lr: 5.0000e-04\n","Epoch 103/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4604 - accuracy: 0.8585 - val_loss: 1.2726 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 104/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4481 - accuracy: 0.8626 - val_loss: 1.2666 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 105/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4268 - accuracy: 0.8688 - val_loss: 1.2768 - val_accuracy: 0.7037 - lr: 5.0000e-04\n","Epoch 106/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3989 - accuracy: 0.8863 - val_loss: 1.3111 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 107/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4154 - accuracy: 0.8785 - val_loss: 1.2759 - val_accuracy: 0.6893 - lr: 5.0000e-04\n","Epoch 108/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4044 - accuracy: 0.8785 - val_loss: 1.3838 - val_accuracy: 0.6996 - lr: 5.0000e-04\n","Epoch 109/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4126 - accuracy: 0.8708 - val_loss: 1.3075 - val_accuracy: 0.6831 - lr: 5.0000e-04\n","Epoch 110/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3862 - accuracy: 0.8847 - val_loss: 1.2960 - val_accuracy: 0.6955 - lr: 5.0000e-04\n","Epoch 111/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3943 - accuracy: 0.8791 - val_loss: 1.3344 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 112/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3700 - accuracy: 0.8929 - val_loss: 1.3259 - val_accuracy: 0.6872 - lr: 2.5000e-04\n","Epoch 113/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3658 - accuracy: 0.8919 - val_loss: 1.3523 - val_accuracy: 0.6893 - lr: 2.5000e-04\n","Epoch 114/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3684 - accuracy: 0.8971 - val_loss: 1.3549 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 115/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3654 - accuracy: 0.8981 - val_loss: 1.3526 - val_accuracy: 0.6872 - lr: 2.5000e-04\n","Epoch 116/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3752 - accuracy: 0.8893 - val_loss: 1.3758 - val_accuracy: 0.7016 - lr: 2.5000e-04\n","Epoch 117/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3473 - accuracy: 0.9053 - val_loss: 1.3734 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 118/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3653 - accuracy: 0.8971 - val_loss: 1.4086 - val_accuracy: 0.6811 - lr: 2.5000e-04\n","Epoch 119/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3749 - accuracy: 0.8935 - val_loss: 1.3782 - val_accuracy: 0.6790 - lr: 2.5000e-04\n","Epoch 120/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3533 - accuracy: 0.8996 - val_loss: 1.3954 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 121/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3373 - accuracy: 0.9007 - val_loss: 1.3957 - val_accuracy: 0.6811 - lr: 2.5000e-04\n","Epoch 122/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3468 - accuracy: 0.9027 - val_loss: 1.4137 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 123/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3469 - accuracy: 0.9027 - val_loss: 1.4122 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 124/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3463 - accuracy: 0.8996 - val_loss: 1.4149 - val_accuracy: 0.6852 - lr: 2.5000e-04\n","Epoch 125/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3415 - accuracy: 0.9032 - val_loss: 1.4167 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 126/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3553 - accuracy: 0.8888 - val_loss: 1.4211 - val_accuracy: 0.6749 - lr: 2.5000e-04\n"]}]},{"cell_type":"code","source":["history5 = model5.fit(\n","    x = x_train5,\n","    y = y_train5,\n","    batch_size = 64,\n","    epochs = 200,\n","    validation_data = (x_test5, y_test5),\n","    callbacks = [\n","        tfk.callbacks.EarlyStopping(monitor='val_accuracy', mode='max', patience=30, restore_best_weights=True),\n","        tfk.callbacks.ReduceLROnPlateau(monitor='val_accuracy', mode='max', patience=15, factor=0.5, min_lr=1e-5)\n","    ]\n",").history"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"pdNG7rdVwOnw","executionInfo":{"status":"ok","timestamp":1671208598089,"user_tz":-60,"elapsed":38946,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"4eaad8c0-114c-4f73-d4d8-c260d2218b3e"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/200\n","31/31 [==============================] - 1s 16ms/step - loss: 2.2474 - accuracy: 0.3423 - val_loss: 2.0772 - val_accuracy: 0.3477 - lr: 0.0010\n","Epoch 2/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.9811 - accuracy: 0.3628 - val_loss: 1.8907 - val_accuracy: 0.4012 - lr: 0.0010\n","Epoch 3/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.8816 - accuracy: 0.3994 - val_loss: 1.8735 - val_accuracy: 0.4095 - lr: 0.0010\n","Epoch 4/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.7853 - accuracy: 0.4246 - val_loss: 1.8004 - val_accuracy: 0.4136 - lr: 0.0010\n","Epoch 5/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.7033 - accuracy: 0.4292 - val_loss: 1.6558 - val_accuracy: 0.4444 - lr: 0.0010\n","Epoch 6/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.5879 - accuracy: 0.4750 - val_loss: 1.5922 - val_accuracy: 0.4609 - lr: 0.0010\n","Epoch 7/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.5180 - accuracy: 0.5018 - val_loss: 1.5560 - val_accuracy: 0.4815 - lr: 0.0010\n","Epoch 8/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.5284 - accuracy: 0.4915 - val_loss: 1.5487 - val_accuracy: 0.4712 - lr: 0.0010\n","Epoch 9/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.4603 - accuracy: 0.5028 - val_loss: 1.5345 - val_accuracy: 0.5082 - lr: 0.0010\n","Epoch 10/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.3895 - accuracy: 0.5296 - val_loss: 1.4176 - val_accuracy: 0.5473 - lr: 0.0010\n","Epoch 11/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.3326 - accuracy: 0.5363 - val_loss: 1.3759 - val_accuracy: 0.5412 - lr: 0.0010\n","Epoch 12/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.3465 - accuracy: 0.5569 - val_loss: 1.5940 - val_accuracy: 0.5247 - lr: 0.0010\n","Epoch 13/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.3314 - accuracy: 0.5476 - val_loss: 1.3677 - val_accuracy: 0.5473 - lr: 0.0010\n","Epoch 14/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2555 - accuracy: 0.5728 - val_loss: 1.3055 - val_accuracy: 0.5802 - lr: 0.0010\n","Epoch 15/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2576 - accuracy: 0.5847 - val_loss: 1.4722 - val_accuracy: 0.5453 - lr: 0.0010\n","Epoch 16/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2575 - accuracy: 0.5713 - val_loss: 1.2946 - val_accuracy: 0.5761 - lr: 0.0010\n","Epoch 17/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1716 - accuracy: 0.6089 - val_loss: 1.3339 - val_accuracy: 0.5802 - lr: 0.0010\n","Epoch 18/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1726 - accuracy: 0.6155 - val_loss: 1.2951 - val_accuracy: 0.5967 - lr: 0.0010\n","Epoch 19/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1660 - accuracy: 0.6109 - val_loss: 1.2702 - val_accuracy: 0.5802 - lr: 0.0010\n","Epoch 20/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.1777 - accuracy: 0.6140 - val_loss: 1.2265 - val_accuracy: 0.6070 - lr: 0.0010\n","Epoch 21/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0864 - accuracy: 0.6428 - val_loss: 1.2376 - val_accuracy: 0.6049 - lr: 0.0010\n","Epoch 22/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0615 - accuracy: 0.6397 - val_loss: 1.1526 - val_accuracy: 0.6296 - lr: 0.0010\n","Epoch 23/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0427 - accuracy: 0.6541 - val_loss: 1.2395 - val_accuracy: 0.6111 - lr: 0.0010\n","Epoch 24/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1196 - accuracy: 0.6330 - val_loss: 1.1936 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 25/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0472 - accuracy: 0.6567 - val_loss: 1.1898 - val_accuracy: 0.6379 - lr: 0.0010\n","Epoch 26/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0295 - accuracy: 0.6557 - val_loss: 1.1788 - val_accuracy: 0.6296 - lr: 0.0010\n","Epoch 27/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0257 - accuracy: 0.6577 - val_loss: 1.1730 - val_accuracy: 0.6440 - lr: 0.0010\n","Epoch 28/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9867 - accuracy: 0.6686 - val_loss: 1.1639 - val_accuracy: 0.6317 - lr: 0.0010\n","Epoch 29/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9623 - accuracy: 0.6752 - val_loss: 1.1246 - val_accuracy: 0.6379 - lr: 0.0010\n","Epoch 30/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0197 - accuracy: 0.6675 - val_loss: 1.3431 - val_accuracy: 0.6029 - lr: 0.0010\n","Epoch 31/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0275 - accuracy: 0.6572 - val_loss: 1.1779 - val_accuracy: 0.6461 - lr: 0.0010\n","Epoch 32/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9570 - accuracy: 0.6763 - val_loss: 1.1058 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 33/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9112 - accuracy: 0.7020 - val_loss: 1.1539 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 34/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9424 - accuracy: 0.6861 - val_loss: 1.1399 - val_accuracy: 0.6379 - lr: 0.0010\n","Epoch 35/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9225 - accuracy: 0.6907 - val_loss: 1.1596 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 36/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9638 - accuracy: 0.6871 - val_loss: 1.1843 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 37/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9817 - accuracy: 0.6794 - val_loss: 1.1506 - val_accuracy: 0.6461 - lr: 0.0010\n","Epoch 38/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9296 - accuracy: 0.6886 - val_loss: 1.1896 - val_accuracy: 0.6235 - lr: 0.0010\n","Epoch 39/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9805 - accuracy: 0.6799 - val_loss: 1.1560 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 40/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8899 - accuracy: 0.7051 - val_loss: 1.1803 - val_accuracy: 0.6070 - lr: 0.0010\n","Epoch 41/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8778 - accuracy: 0.7041 - val_loss: 1.1160 - val_accuracy: 0.6440 - lr: 0.0010\n","Epoch 42/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8494 - accuracy: 0.7123 - val_loss: 1.1021 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 43/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8763 - accuracy: 0.7087 - val_loss: 1.0868 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 44/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9279 - accuracy: 0.7066 - val_loss: 1.1456 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 45/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8632 - accuracy: 0.7138 - val_loss: 1.1340 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 46/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8817 - accuracy: 0.7025 - val_loss: 1.0971 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 47/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8285 - accuracy: 0.7226 - val_loss: 1.0969 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 48/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8287 - accuracy: 0.7164 - val_loss: 1.1417 - val_accuracy: 0.6440 - lr: 0.0010\n","Epoch 49/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8253 - accuracy: 0.7210 - val_loss: 1.1378 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 50/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8056 - accuracy: 0.7334 - val_loss: 1.0992 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 51/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7810 - accuracy: 0.7411 - val_loss: 1.1168 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 52/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7734 - accuracy: 0.7422 - val_loss: 1.0816 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 53/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7579 - accuracy: 0.7581 - val_loss: 1.0937 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 54/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7414 - accuracy: 0.7488 - val_loss: 1.1631 - val_accuracy: 0.6502 - lr: 0.0010\n","Epoch 55/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7679 - accuracy: 0.7488 - val_loss: 1.2069 - val_accuracy: 0.6235 - lr: 0.0010\n","Epoch 56/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7713 - accuracy: 0.7478 - val_loss: 1.0787 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 57/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7195 - accuracy: 0.7540 - val_loss: 1.0705 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 58/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7130 - accuracy: 0.7648 - val_loss: 1.1501 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 59/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6942 - accuracy: 0.7715 - val_loss: 1.1921 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 60/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7866 - accuracy: 0.7499 - val_loss: 1.4058 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 61/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7507 - accuracy: 0.7483 - val_loss: 1.2326 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 62/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7419 - accuracy: 0.7560 - val_loss: 1.1108 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 63/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7398 - accuracy: 0.7566 - val_loss: 1.1433 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 64/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6937 - accuracy: 0.7725 - val_loss: 1.1410 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 65/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7383 - accuracy: 0.7555 - val_loss: 2.0145 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 66/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8980 - accuracy: 0.7483 - val_loss: 1.1303 - val_accuracy: 0.6296 - lr: 0.0010\n","Epoch 67/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7444 - accuracy: 0.7566 - val_loss: 1.1184 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 68/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6966 - accuracy: 0.7720 - val_loss: 1.1667 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 69/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6985 - accuracy: 0.7813 - val_loss: 1.1358 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 70/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7334 - accuracy: 0.7663 - val_loss: 1.1924 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 71/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6742 - accuracy: 0.7854 - val_loss: 1.0951 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 72/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6226 - accuracy: 0.7926 - val_loss: 1.1652 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 73/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5935 - accuracy: 0.8132 - val_loss: 1.1384 - val_accuracy: 0.6728 - lr: 5.0000e-04\n","Epoch 74/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5789 - accuracy: 0.8157 - val_loss: 1.1407 - val_accuracy: 0.6770 - lr: 5.0000e-04\n","Epoch 75/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5754 - accuracy: 0.8209 - val_loss: 1.1588 - val_accuracy: 0.6708 - lr: 5.0000e-04\n","Epoch 76/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5999 - accuracy: 0.8101 - val_loss: 1.1527 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 77/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5617 - accuracy: 0.8178 - val_loss: 1.1134 - val_accuracy: 0.6749 - lr: 5.0000e-04\n","Epoch 78/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5650 - accuracy: 0.8214 - val_loss: 1.1522 - val_accuracy: 0.6831 - lr: 5.0000e-04\n","Epoch 79/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5439 - accuracy: 0.8245 - val_loss: 1.1959 - val_accuracy: 0.6996 - lr: 5.0000e-04\n","Epoch 80/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5727 - accuracy: 0.8178 - val_loss: 1.1946 - val_accuracy: 0.6728 - lr: 5.0000e-04\n","Epoch 81/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5547 - accuracy: 0.8183 - val_loss: 1.2022 - val_accuracy: 0.6831 - lr: 5.0000e-04\n","Epoch 82/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5260 - accuracy: 0.8286 - val_loss: 1.1778 - val_accuracy: 0.6893 - lr: 5.0000e-04\n","Epoch 83/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5397 - accuracy: 0.8343 - val_loss: 1.1634 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 84/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5180 - accuracy: 0.8405 - val_loss: 1.1712 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 85/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5204 - accuracy: 0.8317 - val_loss: 1.2061 - val_accuracy: 0.6708 - lr: 5.0000e-04\n","Epoch 86/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5094 - accuracy: 0.8338 - val_loss: 1.1940 - val_accuracy: 0.6749 - lr: 5.0000e-04\n","Epoch 87/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5225 - accuracy: 0.8410 - val_loss: 1.2426 - val_accuracy: 0.6749 - lr: 5.0000e-04\n","Epoch 88/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5060 - accuracy: 0.8394 - val_loss: 1.2222 - val_accuracy: 0.6852 - lr: 5.0000e-04\n","Epoch 89/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5022 - accuracy: 0.8420 - val_loss: 1.1907 - val_accuracy: 0.6831 - lr: 5.0000e-04\n","Epoch 90/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4942 - accuracy: 0.8379 - val_loss: 1.2104 - val_accuracy: 0.6708 - lr: 5.0000e-04\n","Epoch 91/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4911 - accuracy: 0.8513 - val_loss: 1.2498 - val_accuracy: 0.6728 - lr: 5.0000e-04\n","Epoch 92/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4874 - accuracy: 0.8487 - val_loss: 1.2673 - val_accuracy: 0.6728 - lr: 5.0000e-04\n","Epoch 93/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4841 - accuracy: 0.8549 - val_loss: 1.3003 - val_accuracy: 0.6852 - lr: 5.0000e-04\n","Epoch 94/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4707 - accuracy: 0.8554 - val_loss: 1.2695 - val_accuracy: 0.6790 - lr: 5.0000e-04\n","Epoch 95/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4446 - accuracy: 0.8559 - val_loss: 1.2636 - val_accuracy: 0.6811 - lr: 2.5000e-04\n","Epoch 96/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4498 - accuracy: 0.8600 - val_loss: 1.2722 - val_accuracy: 0.6852 - lr: 2.5000e-04\n","Epoch 97/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4399 - accuracy: 0.8626 - val_loss: 1.2664 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 98/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4425 - accuracy: 0.8621 - val_loss: 1.2438 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 99/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4369 - accuracy: 0.8662 - val_loss: 1.2773 - val_accuracy: 0.6831 - lr: 2.5000e-04\n","Epoch 100/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4364 - accuracy: 0.8688 - val_loss: 1.2612 - val_accuracy: 0.6872 - lr: 2.5000e-04\n","Epoch 101/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4282 - accuracy: 0.8667 - val_loss: 1.3154 - val_accuracy: 0.6893 - lr: 2.5000e-04\n","Epoch 102/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4239 - accuracy: 0.8729 - val_loss: 1.2788 - val_accuracy: 0.6852 - lr: 2.5000e-04\n","Epoch 103/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4239 - accuracy: 0.8718 - val_loss: 1.2690 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 104/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4291 - accuracy: 0.8744 - val_loss: 1.2756 - val_accuracy: 0.6852 - lr: 2.5000e-04\n","Epoch 105/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4201 - accuracy: 0.8770 - val_loss: 1.3692 - val_accuracy: 0.6523 - lr: 2.5000e-04\n","Epoch 106/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4241 - accuracy: 0.8693 - val_loss: 1.2815 - val_accuracy: 0.6831 - lr: 2.5000e-04\n","Epoch 107/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4234 - accuracy: 0.8749 - val_loss: 1.2991 - val_accuracy: 0.6934 - lr: 2.5000e-04\n","Epoch 108/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4070 - accuracy: 0.8785 - val_loss: 1.3159 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 109/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3896 - accuracy: 0.8868 - val_loss: 1.3709 - val_accuracy: 0.7016 - lr: 2.5000e-04\n","Epoch 110/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4023 - accuracy: 0.8821 - val_loss: 1.3510 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 111/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4045 - accuracy: 0.8827 - val_loss: 1.4123 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 112/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3998 - accuracy: 0.8796 - val_loss: 1.4125 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 113/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4026 - accuracy: 0.8821 - val_loss: 1.4093 - val_accuracy: 0.6893 - lr: 2.5000e-04\n","Epoch 114/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4009 - accuracy: 0.8770 - val_loss: 1.3410 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 115/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3844 - accuracy: 0.8873 - val_loss: 1.3388 - val_accuracy: 0.6996 - lr: 2.5000e-04\n","Epoch 116/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3930 - accuracy: 0.8847 - val_loss: 1.3583 - val_accuracy: 0.6934 - lr: 2.5000e-04\n","Epoch 117/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3887 - accuracy: 0.8842 - val_loss: 1.4430 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 118/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3908 - accuracy: 0.8821 - val_loss: 1.3499 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 119/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3807 - accuracy: 0.8888 - val_loss: 1.3821 - val_accuracy: 0.6852 - lr: 2.5000e-04\n","Epoch 120/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3862 - accuracy: 0.8821 - val_loss: 1.4182 - val_accuracy: 0.6934 - lr: 2.5000e-04\n","Epoch 121/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3732 - accuracy: 0.9007 - val_loss: 1.3857 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 122/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3701 - accuracy: 0.8924 - val_loss: 1.3780 - val_accuracy: 0.6872 - lr: 2.5000e-04\n","Epoch 123/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3800 - accuracy: 0.8857 - val_loss: 1.3878 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 124/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3795 - accuracy: 0.8893 - val_loss: 1.4184 - val_accuracy: 0.6852 - lr: 2.5000e-04\n","Epoch 125/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3651 - accuracy: 0.8971 - val_loss: 1.4270 - val_accuracy: 0.6914 - lr: 1.2500e-04\n","Epoch 126/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3678 - accuracy: 0.8909 - val_loss: 1.4546 - val_accuracy: 0.6831 - lr: 1.2500e-04\n","Epoch 127/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3630 - accuracy: 0.8945 - val_loss: 1.4247 - val_accuracy: 0.6914 - lr: 1.2500e-04\n","Epoch 128/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3437 - accuracy: 0.9068 - val_loss: 1.4351 - val_accuracy: 0.6975 - lr: 1.2500e-04\n","Epoch 129/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3480 - accuracy: 0.9043 - val_loss: 1.4446 - val_accuracy: 0.6852 - lr: 1.2500e-04\n","Epoch 130/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3534 - accuracy: 0.8966 - val_loss: 1.4139 - val_accuracy: 0.6934 - lr: 1.2500e-04\n","Epoch 131/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3481 - accuracy: 0.9032 - val_loss: 1.4341 - val_accuracy: 0.6914 - lr: 1.2500e-04\n","Epoch 132/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3505 - accuracy: 0.9058 - val_loss: 1.4481 - val_accuracy: 0.6914 - lr: 1.2500e-04\n","Epoch 133/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3478 - accuracy: 0.8991 - val_loss: 1.4389 - val_accuracy: 0.6914 - lr: 1.2500e-04\n","Epoch 134/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3662 - accuracy: 0.9017 - val_loss: 1.4654 - val_accuracy: 0.6872 - lr: 1.2500e-04\n","Epoch 135/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3604 - accuracy: 0.8971 - val_loss: 1.4090 - val_accuracy: 0.6790 - lr: 1.2500e-04\n","Epoch 136/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3520 - accuracy: 0.9002 - val_loss: 1.4309 - val_accuracy: 0.6831 - lr: 1.2500e-04\n","Epoch 137/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3490 - accuracy: 0.9007 - val_loss: 1.4409 - val_accuracy: 0.6872 - lr: 1.2500e-04\n","Epoch 138/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3519 - accuracy: 0.8981 - val_loss: 1.4506 - val_accuracy: 0.6831 - lr: 1.2500e-04\n","Epoch 139/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3315 - accuracy: 0.9058 - val_loss: 1.4808 - val_accuracy: 0.6914 - lr: 1.2500e-04\n"]}]},{"cell_type":"code","source":["history6 = model6.fit(\n","    x = x_train6,\n","    y = y_train6,\n","    batch_size = 64,\n","    epochs = 200,\n","    validation_data = (x_test6, y_test6),\n","    callbacks = [\n","        tfk.callbacks.EarlyStopping(monitor='val_accuracy', mode='max', patience=30, restore_best_weights=True),\n","        tfk.callbacks.ReduceLROnPlateau(monitor='val_accuracy', mode='max', patience=15, factor=0.5, min_lr=1e-5)\n","    ]\n",").history"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"3dl3HD-_wOlJ","executionInfo":{"status":"ok","timestamp":1671208648819,"user_tz":-60,"elapsed":50754,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"7eda223a-c6b6-41ff-e88e-31690df005aa"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/200\n","31/31 [==============================] - 1s 16ms/step - loss: 2.2947 - accuracy: 0.3227 - val_loss: 2.0244 - val_accuracy: 0.3519 - lr: 0.0010\n","Epoch 2/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.9514 - accuracy: 0.3700 - val_loss: 1.8827 - val_accuracy: 0.3889 - lr: 0.0010\n","Epoch 3/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.8335 - accuracy: 0.4076 - val_loss: 1.9210 - val_accuracy: 0.3827 - lr: 0.0010\n","Epoch 4/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.8277 - accuracy: 0.4174 - val_loss: 1.7854 - val_accuracy: 0.4115 - lr: 0.0010\n","Epoch 5/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.7037 - accuracy: 0.4416 - val_loss: 1.6046 - val_accuracy: 0.4465 - lr: 0.0010\n","Epoch 6/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.6431 - accuracy: 0.4627 - val_loss: 1.6128 - val_accuracy: 0.4733 - lr: 0.0010\n","Epoch 7/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.5746 - accuracy: 0.4627 - val_loss: 1.6156 - val_accuracy: 0.4733 - lr: 0.0010\n","Epoch 8/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.5088 - accuracy: 0.4987 - val_loss: 1.4428 - val_accuracy: 0.5021 - lr: 0.0010\n","Epoch 9/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.4564 - accuracy: 0.5095 - val_loss: 1.4324 - val_accuracy: 0.5247 - lr: 0.0010\n","Epoch 10/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.4958 - accuracy: 0.5095 - val_loss: 1.4738 - val_accuracy: 0.5082 - lr: 0.0010\n","Epoch 11/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.3858 - accuracy: 0.5291 - val_loss: 1.4713 - val_accuracy: 0.5658 - lr: 0.0010\n","Epoch 12/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.3553 - accuracy: 0.5414 - val_loss: 1.4712 - val_accuracy: 0.5782 - lr: 0.0010\n","Epoch 13/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.3437 - accuracy: 0.5419 - val_loss: 1.4192 - val_accuracy: 0.5597 - lr: 0.0010\n","Epoch 14/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2942 - accuracy: 0.5728 - val_loss: 1.3654 - val_accuracy: 0.6049 - lr: 0.0010\n","Epoch 15/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2371 - accuracy: 0.5841 - val_loss: 1.3623 - val_accuracy: 0.5802 - lr: 0.0010\n","Epoch 16/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2649 - accuracy: 0.5677 - val_loss: 1.3408 - val_accuracy: 0.6111 - lr: 0.0010\n","Epoch 17/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1829 - accuracy: 0.6068 - val_loss: 1.2903 - val_accuracy: 0.6091 - lr: 0.0010\n","Epoch 18/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2676 - accuracy: 0.5831 - val_loss: 1.2604 - val_accuracy: 0.6337 - lr: 0.0010\n","Epoch 19/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1711 - accuracy: 0.6042 - val_loss: 1.2903 - val_accuracy: 0.6173 - lr: 0.0010\n","Epoch 20/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1316 - accuracy: 0.6264 - val_loss: 1.3701 - val_accuracy: 0.6255 - lr: 0.0010\n","Epoch 21/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2267 - accuracy: 0.6047 - val_loss: 1.5352 - val_accuracy: 0.5823 - lr: 0.0010\n","Epoch 22/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2298 - accuracy: 0.6047 - val_loss: 1.4141 - val_accuracy: 0.6276 - lr: 0.0010\n","Epoch 23/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1457 - accuracy: 0.6212 - val_loss: 1.3408 - val_accuracy: 0.6296 - lr: 0.0010\n","Epoch 24/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1670 - accuracy: 0.6037 - val_loss: 1.4112 - val_accuracy: 0.6132 - lr: 0.0010\n","Epoch 25/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1053 - accuracy: 0.6284 - val_loss: 1.2961 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 26/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0337 - accuracy: 0.6541 - val_loss: 1.2971 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 27/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0226 - accuracy: 0.6593 - val_loss: 1.3213 - val_accuracy: 0.6358 - lr: 0.0010\n","Epoch 28/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0532 - accuracy: 0.6588 - val_loss: 1.4929 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 29/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0294 - accuracy: 0.6552 - val_loss: 1.2478 - val_accuracy: 0.6461 - lr: 0.0010\n","Epoch 30/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0817 - accuracy: 0.6413 - val_loss: 1.5692 - val_accuracy: 0.6070 - lr: 0.0010\n","Epoch 31/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2259 - accuracy: 0.6217 - val_loss: 1.2060 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 32/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0659 - accuracy: 0.6454 - val_loss: 1.2109 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 33/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9965 - accuracy: 0.6665 - val_loss: 1.2709 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 34/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9705 - accuracy: 0.6773 - val_loss: 1.3158 - val_accuracy: 0.6502 - lr: 0.0010\n","Epoch 35/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9618 - accuracy: 0.6912 - val_loss: 1.2648 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 36/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9614 - accuracy: 0.6830 - val_loss: 1.2839 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 37/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9282 - accuracy: 0.6927 - val_loss: 1.2489 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 38/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9105 - accuracy: 0.6907 - val_loss: 1.2027 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 39/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9373 - accuracy: 0.6871 - val_loss: 1.3974 - val_accuracy: 0.6317 - lr: 0.0010\n","Epoch 40/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9804 - accuracy: 0.6768 - val_loss: 1.3292 - val_accuracy: 0.6255 - lr: 0.0010\n","Epoch 41/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9514 - accuracy: 0.6922 - val_loss: 1.2448 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 42/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9569 - accuracy: 0.6886 - val_loss: 1.3681 - val_accuracy: 0.6852 - lr: 0.0010\n","Epoch 43/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9189 - accuracy: 0.6943 - val_loss: 1.3673 - val_accuracy: 0.6296 - lr: 0.0010\n","Epoch 44/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9600 - accuracy: 0.6773 - val_loss: 1.3188 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 45/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8936 - accuracy: 0.7010 - val_loss: 1.3314 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 46/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9091 - accuracy: 0.7036 - val_loss: 1.1944 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 47/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8837 - accuracy: 0.7082 - val_loss: 1.1652 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 48/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8877 - accuracy: 0.7118 - val_loss: 1.2676 - val_accuracy: 0.6811 - lr: 0.0010\n","Epoch 49/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8676 - accuracy: 0.7082 - val_loss: 1.2564 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 50/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8593 - accuracy: 0.7159 - val_loss: 1.2641 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 51/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8230 - accuracy: 0.7303 - val_loss: 1.1625 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 52/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7946 - accuracy: 0.7406 - val_loss: 1.1996 - val_accuracy: 0.7016 - lr: 5.0000e-04\n","Epoch 53/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7775 - accuracy: 0.7442 - val_loss: 1.1848 - val_accuracy: 0.7058 - lr: 5.0000e-04\n","Epoch 54/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7660 - accuracy: 0.7468 - val_loss: 1.2225 - val_accuracy: 0.7181 - lr: 5.0000e-04\n","Epoch 55/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7507 - accuracy: 0.7612 - val_loss: 1.2217 - val_accuracy: 0.7078 - lr: 5.0000e-04\n","Epoch 56/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7554 - accuracy: 0.7530 - val_loss: 1.3568 - val_accuracy: 0.7140 - lr: 5.0000e-04\n","Epoch 57/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7648 - accuracy: 0.7488 - val_loss: 1.2580 - val_accuracy: 0.7016 - lr: 5.0000e-04\n","Epoch 58/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7591 - accuracy: 0.7530 - val_loss: 1.2655 - val_accuracy: 0.7119 - lr: 5.0000e-04\n","Epoch 59/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7434 - accuracy: 0.7499 - val_loss: 1.2873 - val_accuracy: 0.7037 - lr: 5.0000e-04\n","Epoch 60/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7387 - accuracy: 0.7658 - val_loss: 1.3231 - val_accuracy: 0.7140 - lr: 5.0000e-04\n","Epoch 61/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7422 - accuracy: 0.7648 - val_loss: 1.2534 - val_accuracy: 0.7181 - lr: 5.0000e-04\n","Epoch 62/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7263 - accuracy: 0.7684 - val_loss: 1.2854 - val_accuracy: 0.7222 - lr: 5.0000e-04\n","Epoch 63/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7196 - accuracy: 0.7710 - val_loss: 1.2667 - val_accuracy: 0.6996 - lr: 5.0000e-04\n","Epoch 64/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7525 - accuracy: 0.7638 - val_loss: 1.1556 - val_accuracy: 0.7016 - lr: 5.0000e-04\n","Epoch 65/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7621 - accuracy: 0.7473 - val_loss: 1.2407 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 66/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7308 - accuracy: 0.7638 - val_loss: 1.1815 - val_accuracy: 0.7016 - lr: 5.0000e-04\n","Epoch 67/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7073 - accuracy: 0.7720 - val_loss: 1.1776 - val_accuracy: 0.7119 - lr: 5.0000e-04\n","Epoch 68/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7072 - accuracy: 0.7715 - val_loss: 1.2576 - val_accuracy: 0.7037 - lr: 5.0000e-04\n","Epoch 69/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7066 - accuracy: 0.7597 - val_loss: 1.2919 - val_accuracy: 0.6831 - lr: 5.0000e-04\n","Epoch 70/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7021 - accuracy: 0.7808 - val_loss: 1.2265 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 71/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6857 - accuracy: 0.7782 - val_loss: 1.2662 - val_accuracy: 0.7140 - lr: 5.0000e-04\n","Epoch 72/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6793 - accuracy: 0.7874 - val_loss: 1.2546 - val_accuracy: 0.7078 - lr: 5.0000e-04\n","Epoch 73/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6699 - accuracy: 0.7859 - val_loss: 1.3300 - val_accuracy: 0.7222 - lr: 5.0000e-04\n","Epoch 74/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6709 - accuracy: 0.7844 - val_loss: 1.2633 - val_accuracy: 0.7202 - lr: 5.0000e-04\n","Epoch 75/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6583 - accuracy: 0.7962 - val_loss: 1.2736 - val_accuracy: 0.7243 - lr: 5.0000e-04\n","Epoch 76/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6544 - accuracy: 0.7905 - val_loss: 1.3324 - val_accuracy: 0.7263 - lr: 5.0000e-04\n","Epoch 77/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6544 - accuracy: 0.7777 - val_loss: 1.4128 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 78/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8027 - accuracy: 0.7571 - val_loss: 1.5332 - val_accuracy: 0.7037 - lr: 5.0000e-04\n","Epoch 79/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8182 - accuracy: 0.7494 - val_loss: 1.3004 - val_accuracy: 0.7202 - lr: 5.0000e-04\n","Epoch 80/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6730 - accuracy: 0.7874 - val_loss: 1.3403 - val_accuracy: 0.7263 - lr: 5.0000e-04\n","Epoch 81/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6880 - accuracy: 0.7782 - val_loss: 1.2207 - val_accuracy: 0.7037 - lr: 5.0000e-04\n","Epoch 82/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6869 - accuracy: 0.7766 - val_loss: 1.3603 - val_accuracy: 0.7325 - lr: 5.0000e-04\n","Epoch 83/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6619 - accuracy: 0.7926 - val_loss: 1.3095 - val_accuracy: 0.7181 - lr: 5.0000e-04\n","Epoch 84/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6349 - accuracy: 0.7962 - val_loss: 1.3736 - val_accuracy: 0.7243 - lr: 5.0000e-04\n","Epoch 85/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6448 - accuracy: 0.7900 - val_loss: 1.2900 - val_accuracy: 0.7284 - lr: 5.0000e-04\n","Epoch 86/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6229 - accuracy: 0.8024 - val_loss: 1.3474 - val_accuracy: 0.7243 - lr: 5.0000e-04\n","Epoch 87/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6188 - accuracy: 0.8008 - val_loss: 1.3395 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 88/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6269 - accuracy: 0.7967 - val_loss: 1.3219 - val_accuracy: 0.7263 - lr: 5.0000e-04\n","Epoch 89/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6205 - accuracy: 0.7998 - val_loss: 1.3663 - val_accuracy: 0.7243 - lr: 5.0000e-04\n","Epoch 90/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6182 - accuracy: 0.7967 - val_loss: 1.3123 - val_accuracy: 0.7181 - lr: 5.0000e-04\n","Epoch 91/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6225 - accuracy: 0.7993 - val_loss: 1.3098 - val_accuracy: 0.7222 - lr: 5.0000e-04\n","Epoch 92/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6043 - accuracy: 0.8019 - val_loss: 1.4525 - val_accuracy: 0.7202 - lr: 5.0000e-04\n","Epoch 93/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6199 - accuracy: 0.7998 - val_loss: 1.4069 - val_accuracy: 0.7222 - lr: 5.0000e-04\n","Epoch 94/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5965 - accuracy: 0.8091 - val_loss: 1.3522 - val_accuracy: 0.7243 - lr: 5.0000e-04\n","Epoch 95/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5985 - accuracy: 0.8034 - val_loss: 1.4057 - val_accuracy: 0.7284 - lr: 5.0000e-04\n","Epoch 96/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6010 - accuracy: 0.8065 - val_loss: 1.3850 - val_accuracy: 0.7160 - lr: 5.0000e-04\n","Epoch 97/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6070 - accuracy: 0.8019 - val_loss: 1.4297 - val_accuracy: 0.7243 - lr: 5.0000e-04\n","Epoch 98/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5764 - accuracy: 0.8152 - val_loss: 1.4721 - val_accuracy: 0.7263 - lr: 2.5000e-04\n","Epoch 99/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5515 - accuracy: 0.8286 - val_loss: 1.4352 - val_accuracy: 0.7366 - lr: 2.5000e-04\n","Epoch 100/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5425 - accuracy: 0.8250 - val_loss: 1.4134 - val_accuracy: 0.7325 - lr: 2.5000e-04\n","Epoch 101/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5397 - accuracy: 0.8281 - val_loss: 1.4672 - val_accuracy: 0.7222 - lr: 2.5000e-04\n","Epoch 102/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5410 - accuracy: 0.8332 - val_loss: 1.4991 - val_accuracy: 0.7263 - lr: 2.5000e-04\n","Epoch 103/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5442 - accuracy: 0.8245 - val_loss: 1.4708 - val_accuracy: 0.7325 - lr: 2.5000e-04\n","Epoch 104/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5368 - accuracy: 0.8296 - val_loss: 1.4850 - val_accuracy: 0.7449 - lr: 2.5000e-04\n","Epoch 105/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5422 - accuracy: 0.8260 - val_loss: 1.4735 - val_accuracy: 0.7407 - lr: 2.5000e-04\n","Epoch 106/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5177 - accuracy: 0.8425 - val_loss: 1.4913 - val_accuracy: 0.7325 - lr: 2.5000e-04\n","Epoch 107/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5249 - accuracy: 0.8332 - val_loss: 1.4909 - val_accuracy: 0.7263 - lr: 2.5000e-04\n","Epoch 108/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5319 - accuracy: 0.8322 - val_loss: 1.4766 - val_accuracy: 0.7305 - lr: 2.5000e-04\n","Epoch 109/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5325 - accuracy: 0.8291 - val_loss: 1.4843 - val_accuracy: 0.7325 - lr: 2.5000e-04\n","Epoch 110/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5251 - accuracy: 0.8369 - val_loss: 1.4549 - val_accuracy: 0.7181 - lr: 2.5000e-04\n","Epoch 111/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5154 - accuracy: 0.8338 - val_loss: 1.5011 - val_accuracy: 0.7449 - lr: 2.5000e-04\n","Epoch 112/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5050 - accuracy: 0.8399 - val_loss: 1.5517 - val_accuracy: 0.7346 - lr: 2.5000e-04\n","Epoch 113/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5117 - accuracy: 0.8363 - val_loss: 1.5358 - val_accuracy: 0.7305 - lr: 2.5000e-04\n","Epoch 114/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5191 - accuracy: 0.8343 - val_loss: 1.5891 - val_accuracy: 0.7305 - lr: 2.5000e-04\n","Epoch 115/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5241 - accuracy: 0.8353 - val_loss: 1.5026 - val_accuracy: 0.7387 - lr: 2.5000e-04\n","Epoch 116/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5110 - accuracy: 0.8446 - val_loss: 1.4942 - val_accuracy: 0.7366 - lr: 2.5000e-04\n","Epoch 117/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5086 - accuracy: 0.8420 - val_loss: 1.4930 - val_accuracy: 0.7325 - lr: 2.5000e-04\n","Epoch 118/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5114 - accuracy: 0.8358 - val_loss: 1.5703 - val_accuracy: 0.7305 - lr: 2.5000e-04\n","Epoch 119/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4971 - accuracy: 0.8456 - val_loss: 1.4772 - val_accuracy: 0.7243 - lr: 2.5000e-04\n","Epoch 120/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4874 - accuracy: 0.8507 - val_loss: 1.4694 - val_accuracy: 0.7387 - lr: 1.2500e-04\n","Epoch 121/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4731 - accuracy: 0.8549 - val_loss: 1.5238 - val_accuracy: 0.7449 - lr: 1.2500e-04\n","Epoch 122/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4742 - accuracy: 0.8590 - val_loss: 1.5292 - val_accuracy: 0.7305 - lr: 1.2500e-04\n","Epoch 123/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4739 - accuracy: 0.8574 - val_loss: 1.5497 - val_accuracy: 0.7449 - lr: 1.2500e-04\n","Epoch 124/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4719 - accuracy: 0.8585 - val_loss: 1.5969 - val_accuracy: 0.7387 - lr: 1.2500e-04\n","Epoch 125/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4739 - accuracy: 0.8523 - val_loss: 1.5742 - val_accuracy: 0.7305 - lr: 1.2500e-04\n","Epoch 126/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4726 - accuracy: 0.8523 - val_loss: 1.5671 - val_accuracy: 0.7325 - lr: 1.2500e-04\n","Epoch 127/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4617 - accuracy: 0.8564 - val_loss: 1.5873 - val_accuracy: 0.7387 - lr: 1.2500e-04\n","Epoch 128/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4569 - accuracy: 0.8590 - val_loss: 1.5851 - val_accuracy: 0.7449 - lr: 1.2500e-04\n","Epoch 129/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4673 - accuracy: 0.8528 - val_loss: 1.6160 - val_accuracy: 0.7325 - lr: 1.2500e-04\n","Epoch 130/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4677 - accuracy: 0.8585 - val_loss: 1.6627 - val_accuracy: 0.7325 - lr: 1.2500e-04\n","Epoch 131/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4722 - accuracy: 0.8518 - val_loss: 1.6292 - val_accuracy: 0.7469 - lr: 1.2500e-04\n","Epoch 132/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4567 - accuracy: 0.8626 - val_loss: 1.6168 - val_accuracy: 0.7366 - lr: 1.2500e-04\n","Epoch 133/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4570 - accuracy: 0.8641 - val_loss: 1.6388 - val_accuracy: 0.7490 - lr: 1.2500e-04\n","Epoch 134/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4519 - accuracy: 0.8605 - val_loss: 1.6284 - val_accuracy: 0.7387 - lr: 1.2500e-04\n","Epoch 135/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4612 - accuracy: 0.8631 - val_loss: 1.6538 - val_accuracy: 0.7428 - lr: 1.2500e-04\n","Epoch 136/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4531 - accuracy: 0.8569 - val_loss: 1.6381 - val_accuracy: 0.7346 - lr: 1.2500e-04\n","Epoch 137/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4478 - accuracy: 0.8652 - val_loss: 1.6591 - val_accuracy: 0.7346 - lr: 1.2500e-04\n","Epoch 138/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4507 - accuracy: 0.8605 - val_loss: 1.6739 - val_accuracy: 0.7428 - lr: 1.2500e-04\n","Epoch 139/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4546 - accuracy: 0.8636 - val_loss: 1.6435 - val_accuracy: 0.7325 - lr: 1.2500e-04\n","Epoch 140/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4516 - accuracy: 0.8595 - val_loss: 1.6546 - val_accuracy: 0.7449 - lr: 1.2500e-04\n","Epoch 141/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4524 - accuracy: 0.8564 - val_loss: 1.6688 - val_accuracy: 0.7346 - lr: 1.2500e-04\n","Epoch 142/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4522 - accuracy: 0.8646 - val_loss: 1.6586 - val_accuracy: 0.7284 - lr: 1.2500e-04\n","Epoch 143/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4478 - accuracy: 0.8605 - val_loss: 1.6471 - val_accuracy: 0.7366 - lr: 1.2500e-04\n","Epoch 144/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4467 - accuracy: 0.8641 - val_loss: 1.6821 - val_accuracy: 0.7325 - lr: 1.2500e-04\n","Epoch 145/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4467 - accuracy: 0.8646 - val_loss: 1.6849 - val_accuracy: 0.7366 - lr: 1.2500e-04\n","Epoch 146/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4377 - accuracy: 0.8703 - val_loss: 1.7253 - val_accuracy: 0.7366 - lr: 1.2500e-04\n","Epoch 147/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4435 - accuracy: 0.8662 - val_loss: 1.7140 - val_accuracy: 0.7407 - lr: 1.2500e-04\n","Epoch 148/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4425 - accuracy: 0.8682 - val_loss: 1.7317 - val_accuracy: 0.7449 - lr: 1.2500e-04\n","Epoch 149/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4345 - accuracy: 0.8693 - val_loss: 1.7187 - val_accuracy: 0.7531 - lr: 6.2500e-05\n","Epoch 150/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4378 - accuracy: 0.8631 - val_loss: 1.6914 - val_accuracy: 0.7449 - lr: 6.2500e-05\n","Epoch 151/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4341 - accuracy: 0.8708 - val_loss: 1.6963 - val_accuracy: 0.7387 - lr: 6.2500e-05\n","Epoch 152/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4524 - accuracy: 0.8616 - val_loss: 1.6538 - val_accuracy: 0.7428 - lr: 6.2500e-05\n","Epoch 153/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4386 - accuracy: 0.8693 - val_loss: 1.7095 - val_accuracy: 0.7469 - lr: 6.2500e-05\n","Epoch 154/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4299 - accuracy: 0.8708 - val_loss: 1.7661 - val_accuracy: 0.7469 - lr: 6.2500e-05\n","Epoch 155/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4306 - accuracy: 0.8718 - val_loss: 1.7660 - val_accuracy: 0.7387 - lr: 6.2500e-05\n","Epoch 156/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4255 - accuracy: 0.8765 - val_loss: 1.7492 - val_accuracy: 0.7469 - lr: 6.2500e-05\n","Epoch 157/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4259 - accuracy: 0.8703 - val_loss: 1.7373 - val_accuracy: 0.7510 - lr: 6.2500e-05\n","Epoch 158/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4178 - accuracy: 0.8739 - val_loss: 1.7570 - val_accuracy: 0.7469 - lr: 6.2500e-05\n","Epoch 159/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4297 - accuracy: 0.8744 - val_loss: 1.7399 - val_accuracy: 0.7428 - lr: 6.2500e-05\n","Epoch 160/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4223 - accuracy: 0.8760 - val_loss: 1.7539 - val_accuracy: 0.7469 - lr: 6.2500e-05\n","Epoch 161/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4228 - accuracy: 0.8791 - val_loss: 1.7406 - val_accuracy: 0.7407 - lr: 6.2500e-05\n","Epoch 162/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4187 - accuracy: 0.8811 - val_loss: 1.7587 - val_accuracy: 0.7469 - lr: 6.2500e-05\n","Epoch 163/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4271 - accuracy: 0.8755 - val_loss: 1.7668 - val_accuracy: 0.7346 - lr: 6.2500e-05\n","Epoch 164/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4166 - accuracy: 0.8760 - val_loss: 1.7635 - val_accuracy: 0.7469 - lr: 6.2500e-05\n","Epoch 165/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4219 - accuracy: 0.8827 - val_loss: 1.7554 - val_accuracy: 0.7407 - lr: 3.1250e-05\n","Epoch 166/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4216 - accuracy: 0.8770 - val_loss: 1.7467 - val_accuracy: 0.7387 - lr: 3.1250e-05\n","Epoch 167/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4196 - accuracy: 0.8791 - val_loss: 1.7747 - val_accuracy: 0.7387 - lr: 3.1250e-05\n","Epoch 168/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4231 - accuracy: 0.8816 - val_loss: 1.7534 - val_accuracy: 0.7366 - lr: 3.1250e-05\n","Epoch 169/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4115 - accuracy: 0.8816 - val_loss: 1.7843 - val_accuracy: 0.7469 - lr: 3.1250e-05\n","Epoch 170/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4238 - accuracy: 0.8749 - val_loss: 1.7829 - val_accuracy: 0.7407 - lr: 3.1250e-05\n","Epoch 171/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4156 - accuracy: 0.8796 - val_loss: 1.7598 - val_accuracy: 0.7428 - lr: 3.1250e-05\n","Epoch 172/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4102 - accuracy: 0.8847 - val_loss: 1.7601 - val_accuracy: 0.7449 - lr: 3.1250e-05\n","Epoch 173/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4115 - accuracy: 0.8770 - val_loss: 1.7679 - val_accuracy: 0.7407 - lr: 3.1250e-05\n","Epoch 174/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4226 - accuracy: 0.8770 - val_loss: 1.7667 - val_accuracy: 0.7490 - lr: 3.1250e-05\n","Epoch 175/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4157 - accuracy: 0.8713 - val_loss: 1.7795 - val_accuracy: 0.7449 - lr: 3.1250e-05\n","Epoch 176/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4056 - accuracy: 0.8785 - val_loss: 1.7897 - val_accuracy: 0.7325 - lr: 3.1250e-05\n","Epoch 177/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4113 - accuracy: 0.8806 - val_loss: 1.8142 - val_accuracy: 0.7449 - lr: 3.1250e-05\n","Epoch 178/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4187 - accuracy: 0.8780 - val_loss: 1.8244 - val_accuracy: 0.7407 - lr: 3.1250e-05\n","Epoch 179/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4152 - accuracy: 0.8821 - val_loss: 1.8323 - val_accuracy: 0.7469 - lr: 3.1250e-05\n"]}]},{"cell_type":"code","source":["history7 = model7.fit(\n","    x = x_train7,\n","    y = y_train7,\n","    batch_size = 64,\n","    epochs = 200,\n","    validation_data = (x_test7, y_test7),\n","    callbacks = [\n","        tfk.callbacks.EarlyStopping(monitor='val_accuracy', mode='max', patience=30, restore_best_weights=True),\n","        tfk.callbacks.ReduceLROnPlateau(monitor='val_accuracy', mode='max', patience=15, factor=0.5, min_lr=1e-5)\n","    ]\n",").history"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"8qRkDoU-wRL6","executionInfo":{"status":"ok","timestamp":1671208682590,"user_tz":-60,"elapsed":33778,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"5cdfc52d-8a63-43da-dfe6-4e50f1cc7501"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/200\n","31/31 [==============================] - 1s 15ms/step - loss: 2.2583 - accuracy: 0.3325 - val_loss: 2.0090 - val_accuracy: 0.3560 - lr: 0.0010\n","Epoch 2/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.9318 - accuracy: 0.3793 - val_loss: 1.8871 - val_accuracy: 0.3765 - lr: 0.0010\n","Epoch 3/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.8462 - accuracy: 0.4035 - val_loss: 1.8684 - val_accuracy: 0.4198 - lr: 0.0010\n","Epoch 4/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.7328 - accuracy: 0.4297 - val_loss: 2.0329 - val_accuracy: 0.4547 - lr: 0.0010\n","Epoch 5/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.6560 - accuracy: 0.4375 - val_loss: 1.6307 - val_accuracy: 0.4835 - lr: 0.0010\n","Epoch 6/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.5172 - accuracy: 0.5028 - val_loss: 1.5879 - val_accuracy: 0.5123 - lr: 0.0010\n","Epoch 7/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.4463 - accuracy: 0.5136 - val_loss: 1.8109 - val_accuracy: 0.5185 - lr: 0.0010\n","Epoch 8/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.5147 - accuracy: 0.5152 - val_loss: 1.4858 - val_accuracy: 0.5309 - lr: 0.0010\n","Epoch 9/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3953 - accuracy: 0.5404 - val_loss: 1.4910 - val_accuracy: 0.5041 - lr: 0.0010\n","Epoch 10/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3341 - accuracy: 0.5466 - val_loss: 1.4471 - val_accuracy: 0.5309 - lr: 0.0010\n","Epoch 11/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.2950 - accuracy: 0.5656 - val_loss: 1.4781 - val_accuracy: 0.5967 - lr: 0.0010\n","Epoch 12/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2647 - accuracy: 0.5785 - val_loss: 1.4872 - val_accuracy: 0.5782 - lr: 0.0010\n","Epoch 13/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2111 - accuracy: 0.5888 - val_loss: 1.4462 - val_accuracy: 0.5885 - lr: 0.0010\n","Epoch 14/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1869 - accuracy: 0.6104 - val_loss: 1.4151 - val_accuracy: 0.6111 - lr: 0.0010\n","Epoch 15/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1792 - accuracy: 0.6104 - val_loss: 1.3695 - val_accuracy: 0.5802 - lr: 0.0010\n","Epoch 16/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1763 - accuracy: 0.6155 - val_loss: 1.6008 - val_accuracy: 0.5885 - lr: 0.0010\n","Epoch 17/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1242 - accuracy: 0.6289 - val_loss: 1.5582 - val_accuracy: 0.5947 - lr: 0.0010\n","Epoch 18/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2205 - accuracy: 0.6022 - val_loss: 1.5375 - val_accuracy: 0.5741 - lr: 0.0010\n","Epoch 19/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2079 - accuracy: 0.6047 - val_loss: 1.5458 - val_accuracy: 0.5679 - lr: 0.0010\n","Epoch 20/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1385 - accuracy: 0.6186 - val_loss: 1.4355 - val_accuracy: 0.6049 - lr: 0.0010\n","Epoch 21/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1042 - accuracy: 0.6356 - val_loss: 1.3075 - val_accuracy: 0.6091 - lr: 0.0010\n","Epoch 22/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0677 - accuracy: 0.6495 - val_loss: 1.3131 - val_accuracy: 0.6379 - lr: 0.0010\n","Epoch 23/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0444 - accuracy: 0.6562 - val_loss: 1.5103 - val_accuracy: 0.5967 - lr: 0.0010\n","Epoch 24/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1029 - accuracy: 0.6361 - val_loss: 1.3049 - val_accuracy: 0.6152 - lr: 0.0010\n","Epoch 25/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0320 - accuracy: 0.6670 - val_loss: 1.2727 - val_accuracy: 0.6358 - lr: 0.0010\n","Epoch 26/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0107 - accuracy: 0.6562 - val_loss: 1.4621 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 27/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9727 - accuracy: 0.6737 - val_loss: 1.3500 - val_accuracy: 0.6379 - lr: 0.0010\n","Epoch 28/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0084 - accuracy: 0.6665 - val_loss: 1.2597 - val_accuracy: 0.6440 - lr: 0.0010\n","Epoch 29/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9914 - accuracy: 0.6773 - val_loss: 1.4301 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 30/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9454 - accuracy: 0.6891 - val_loss: 1.5682 - val_accuracy: 0.6399 - lr: 0.0010\n","Epoch 31/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9703 - accuracy: 0.6824 - val_loss: 1.5478 - val_accuracy: 0.6173 - lr: 0.0010\n","Epoch 32/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9660 - accuracy: 0.6788 - val_loss: 1.4759 - val_accuracy: 0.6317 - lr: 0.0010\n","Epoch 33/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9817 - accuracy: 0.6835 - val_loss: 1.2135 - val_accuracy: 0.6379 - lr: 0.0010\n","Epoch 34/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0181 - accuracy: 0.6660 - val_loss: 1.2037 - val_accuracy: 0.6276 - lr: 0.0010\n","Epoch 35/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0164 - accuracy: 0.6716 - val_loss: 1.3520 - val_accuracy: 0.6337 - lr: 0.0010\n","Epoch 36/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9875 - accuracy: 0.6727 - val_loss: 1.3396 - val_accuracy: 0.6276 - lr: 0.0010\n","Epoch 37/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9296 - accuracy: 0.6891 - val_loss: 1.4279 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 38/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9163 - accuracy: 0.7051 - val_loss: 1.4739 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 39/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9407 - accuracy: 0.6835 - val_loss: 1.4184 - val_accuracy: 0.6070 - lr: 0.0010\n","Epoch 40/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9463 - accuracy: 0.6886 - val_loss: 1.3934 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 41/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8927 - accuracy: 0.6974 - val_loss: 1.3560 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 42/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8842 - accuracy: 0.7144 - val_loss: 1.2144 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 43/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8708 - accuracy: 0.7087 - val_loss: 1.2423 - val_accuracy: 0.6296 - lr: 0.0010\n","Epoch 44/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8705 - accuracy: 0.7092 - val_loss: 1.2716 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 45/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8458 - accuracy: 0.7102 - val_loss: 1.4823 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 46/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8545 - accuracy: 0.7262 - val_loss: 1.3321 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 47/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8418 - accuracy: 0.7190 - val_loss: 1.3623 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 48/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8557 - accuracy: 0.7118 - val_loss: 1.4686 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 49/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8296 - accuracy: 0.7174 - val_loss: 1.6818 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 50/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8917 - accuracy: 0.6999 - val_loss: 1.2485 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 51/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8356 - accuracy: 0.7113 - val_loss: 1.3667 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 52/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7878 - accuracy: 0.7375 - val_loss: 1.5430 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 53/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7890 - accuracy: 0.7298 - val_loss: 1.5552 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 54/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8166 - accuracy: 0.7329 - val_loss: 1.5629 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 55/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7648 - accuracy: 0.7478 - val_loss: 1.5611 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 56/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7517 - accuracy: 0.7545 - val_loss: 1.3076 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 57/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8086 - accuracy: 0.7370 - val_loss: 1.4403 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 58/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7824 - accuracy: 0.7401 - val_loss: 1.2061 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 59/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8036 - accuracy: 0.7329 - val_loss: 1.3036 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 60/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8460 - accuracy: 0.7411 - val_loss: 1.6767 - val_accuracy: 0.6811 - lr: 0.0010\n","Epoch 61/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7877 - accuracy: 0.7416 - val_loss: 1.3813 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 62/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7897 - accuracy: 0.7406 - val_loss: 1.2783 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 63/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7799 - accuracy: 0.7514 - val_loss: 1.3179 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 64/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7319 - accuracy: 0.7591 - val_loss: 1.1499 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 65/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7045 - accuracy: 0.7705 - val_loss: 1.2193 - val_accuracy: 0.6975 - lr: 0.0010\n","Epoch 66/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7066 - accuracy: 0.7756 - val_loss: 1.3036 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 67/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6753 - accuracy: 0.7838 - val_loss: 1.2825 - val_accuracy: 0.6996 - lr: 0.0010\n","Epoch 68/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6945 - accuracy: 0.7766 - val_loss: 1.1540 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 69/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7265 - accuracy: 0.7643 - val_loss: 1.2996 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 70/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7206 - accuracy: 0.7684 - val_loss: 1.2930 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 71/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7575 - accuracy: 0.7725 - val_loss: 1.1642 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 72/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7613 - accuracy: 0.7607 - val_loss: 1.0592 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 73/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7421 - accuracy: 0.7602 - val_loss: 1.1786 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 74/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7024 - accuracy: 0.7751 - val_loss: 1.2277 - val_accuracy: 0.6852 - lr: 0.0010\n","Epoch 75/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6755 - accuracy: 0.7838 - val_loss: 1.1482 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 76/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7177 - accuracy: 0.7653 - val_loss: 1.1554 - val_accuracy: 0.6975 - lr: 0.0010\n","Epoch 77/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6544 - accuracy: 0.7946 - val_loss: 1.1645 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 78/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6204 - accuracy: 0.7988 - val_loss: 1.2389 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 79/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6291 - accuracy: 0.8019 - val_loss: 1.2143 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 80/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6118 - accuracy: 0.8075 - val_loss: 1.2357 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 81/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5808 - accuracy: 0.8183 - val_loss: 1.2699 - val_accuracy: 0.6934 - lr: 0.0010\n","Epoch 82/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6082 - accuracy: 0.8106 - val_loss: 1.3960 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 83/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5670 - accuracy: 0.8188 - val_loss: 1.3780 - val_accuracy: 0.6934 - lr: 5.0000e-04\n","Epoch 84/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5291 - accuracy: 0.8435 - val_loss: 1.2657 - val_accuracy: 0.7078 - lr: 5.0000e-04\n","Epoch 85/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5418 - accuracy: 0.8255 - val_loss: 1.3941 - val_accuracy: 0.6790 - lr: 5.0000e-04\n","Epoch 86/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5380 - accuracy: 0.8307 - val_loss: 1.4495 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 87/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5235 - accuracy: 0.8461 - val_loss: 1.4980 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 88/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5281 - accuracy: 0.8389 - val_loss: 1.2499 - val_accuracy: 0.7140 - lr: 5.0000e-04\n","Epoch 89/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5073 - accuracy: 0.8456 - val_loss: 1.5156 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 90/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5253 - accuracy: 0.8266 - val_loss: 1.3583 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 91/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5215 - accuracy: 0.8425 - val_loss: 1.3161 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 92/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5105 - accuracy: 0.8456 - val_loss: 1.4436 - val_accuracy: 0.6975 - lr: 5.0000e-04\n","Epoch 93/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4945 - accuracy: 0.8528 - val_loss: 1.4833 - val_accuracy: 0.6996 - lr: 5.0000e-04\n","Epoch 94/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4877 - accuracy: 0.8456 - val_loss: 1.5578 - val_accuracy: 0.6975 - lr: 5.0000e-04\n","Epoch 95/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4901 - accuracy: 0.8513 - val_loss: 1.5065 - val_accuracy: 0.6955 - lr: 5.0000e-04\n","Epoch 96/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4877 - accuracy: 0.8502 - val_loss: 1.5424 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 97/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4952 - accuracy: 0.8538 - val_loss: 1.6381 - val_accuracy: 0.6852 - lr: 5.0000e-04\n","Epoch 98/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4698 - accuracy: 0.8554 - val_loss: 1.6764 - val_accuracy: 0.6934 - lr: 5.0000e-04\n","Epoch 99/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4655 - accuracy: 0.8585 - val_loss: 1.7682 - val_accuracy: 0.6934 - lr: 5.0000e-04\n","Epoch 100/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4729 - accuracy: 0.8528 - val_loss: 1.7147 - val_accuracy: 0.7016 - lr: 5.0000e-04\n","Epoch 101/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4591 - accuracy: 0.8713 - val_loss: 1.7778 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 102/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4541 - accuracy: 0.8621 - val_loss: 1.7188 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 103/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4624 - accuracy: 0.8564 - val_loss: 1.6529 - val_accuracy: 0.6996 - lr: 5.0000e-04\n","Epoch 104/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4296 - accuracy: 0.8734 - val_loss: 1.7279 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 105/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4187 - accuracy: 0.8791 - val_loss: 1.8014 - val_accuracy: 0.7016 - lr: 2.5000e-04\n","Epoch 106/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4217 - accuracy: 0.8796 - val_loss: 1.7733 - val_accuracy: 0.6934 - lr: 2.5000e-04\n","Epoch 107/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4366 - accuracy: 0.8698 - val_loss: 1.7497 - val_accuracy: 0.6996 - lr: 2.5000e-04\n","Epoch 108/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4211 - accuracy: 0.8672 - val_loss: 1.8075 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 109/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4135 - accuracy: 0.8811 - val_loss: 1.7539 - val_accuracy: 0.6872 - lr: 2.5000e-04\n","Epoch 110/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4227 - accuracy: 0.8749 - val_loss: 1.8011 - val_accuracy: 0.7140 - lr: 2.5000e-04\n","Epoch 111/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4159 - accuracy: 0.8734 - val_loss: 1.7699 - val_accuracy: 0.7099 - lr: 2.5000e-04\n","Epoch 112/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4092 - accuracy: 0.8791 - val_loss: 1.7109 - val_accuracy: 0.6934 - lr: 2.5000e-04\n","Epoch 113/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4093 - accuracy: 0.8811 - val_loss: 1.8490 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 114/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4096 - accuracy: 0.8785 - val_loss: 1.8307 - val_accuracy: 0.6872 - lr: 2.5000e-04\n","Epoch 115/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4169 - accuracy: 0.8791 - val_loss: 1.8961 - val_accuracy: 0.7016 - lr: 2.5000e-04\n","Epoch 116/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3974 - accuracy: 0.8816 - val_loss: 1.8596 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 117/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4071 - accuracy: 0.8842 - val_loss: 1.8375 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 118/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4140 - accuracy: 0.8770 - val_loss: 1.8958 - val_accuracy: 0.6934 - lr: 2.5000e-04\n"]}]},{"cell_type":"code","source":["history8 = model8.fit(\n","    x = x_train8,\n","    y = y_train8,\n","    batch_size = 64,\n","    epochs = 200,\n","    validation_data = (x_test8, y_test8),\n","    callbacks = [\n","        tfk.callbacks.EarlyStopping(monitor='val_accuracy', mode='max', patience=30, restore_best_weights=True),\n","        tfk.callbacks.ReduceLROnPlateau(monitor='val_accuracy', mode='max', patience=15, factor=0.5, min_lr=1e-5)\n","    ]\n",").history"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"fnZgHNiTwRGs","executionInfo":{"status":"ok","timestamp":1671208725295,"user_tz":-60,"elapsed":42739,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"72338349-9589-47ae-c979-ada79deed43e"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/200\n","31/31 [==============================] - 1s 16ms/step - loss: 2.2501 - accuracy: 0.3227 - val_loss: 1.9926 - val_accuracy: 0.3745 - lr: 0.0010\n","Epoch 2/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.9358 - accuracy: 0.3747 - val_loss: 1.8585 - val_accuracy: 0.4074 - lr: 0.0010\n","Epoch 3/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.8481 - accuracy: 0.3942 - val_loss: 1.7869 - val_accuracy: 0.4053 - lr: 0.0010\n","Epoch 4/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.7389 - accuracy: 0.4256 - val_loss: 1.7194 - val_accuracy: 0.4403 - lr: 0.0010\n","Epoch 5/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.6161 - accuracy: 0.4663 - val_loss: 1.6986 - val_accuracy: 0.4300 - lr: 0.0010\n","Epoch 6/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.5906 - accuracy: 0.4802 - val_loss: 1.5134 - val_accuracy: 0.4733 - lr: 0.0010\n","Epoch 7/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.5010 - accuracy: 0.5039 - val_loss: 1.5754 - val_accuracy: 0.4691 - lr: 0.0010\n","Epoch 8/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.4244 - accuracy: 0.5203 - val_loss: 1.5752 - val_accuracy: 0.4979 - lr: 0.0010\n","Epoch 9/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.4048 - accuracy: 0.5332 - val_loss: 1.4828 - val_accuracy: 0.5185 - lr: 0.0010\n","Epoch 10/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3570 - accuracy: 0.5450 - val_loss: 1.6022 - val_accuracy: 0.4856 - lr: 0.0010\n","Epoch 11/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.3876 - accuracy: 0.5497 - val_loss: 1.5050 - val_accuracy: 0.5082 - lr: 0.0010\n","Epoch 12/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2904 - accuracy: 0.5749 - val_loss: 1.4132 - val_accuracy: 0.5597 - lr: 0.0010\n","Epoch 13/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3555 - accuracy: 0.5620 - val_loss: 1.3809 - val_accuracy: 0.5206 - lr: 0.0010\n","Epoch 14/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.2883 - accuracy: 0.5764 - val_loss: 1.3863 - val_accuracy: 0.5576 - lr: 0.0010\n","Epoch 15/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2221 - accuracy: 0.5944 - val_loss: 1.3467 - val_accuracy: 0.6008 - lr: 0.0010\n","Epoch 16/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1829 - accuracy: 0.5955 - val_loss: 1.3642 - val_accuracy: 0.5802 - lr: 0.0010\n","Epoch 17/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1926 - accuracy: 0.6155 - val_loss: 1.3339 - val_accuracy: 0.5926 - lr: 0.0010\n","Epoch 18/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1244 - accuracy: 0.6315 - val_loss: 1.3344 - val_accuracy: 0.6132 - lr: 0.0010\n","Epoch 19/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0975 - accuracy: 0.6423 - val_loss: 1.3563 - val_accuracy: 0.6008 - lr: 0.0010\n","Epoch 20/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0922 - accuracy: 0.6351 - val_loss: 1.3354 - val_accuracy: 0.6214 - lr: 0.0010\n","Epoch 21/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0633 - accuracy: 0.6526 - val_loss: 1.4792 - val_accuracy: 0.5782 - lr: 0.0010\n","Epoch 22/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0706 - accuracy: 0.6382 - val_loss: 1.3751 - val_accuracy: 0.5988 - lr: 0.0010\n","Epoch 23/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0525 - accuracy: 0.6531 - val_loss: 1.3018 - val_accuracy: 0.6193 - lr: 0.0010\n","Epoch 24/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0233 - accuracy: 0.6572 - val_loss: 1.2602 - val_accuracy: 0.6317 - lr: 0.0010\n","Epoch 25/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9803 - accuracy: 0.6788 - val_loss: 1.3095 - val_accuracy: 0.6440 - lr: 0.0010\n","Epoch 26/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1124 - accuracy: 0.6449 - val_loss: 1.3629 - val_accuracy: 0.6029 - lr: 0.0010\n","Epoch 27/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0469 - accuracy: 0.6526 - val_loss: 1.2564 - val_accuracy: 0.6255 - lr: 0.0010\n","Epoch 28/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.1473 - accuracy: 0.6500 - val_loss: 1.3034 - val_accuracy: 0.5988 - lr: 0.0010\n","Epoch 29/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0939 - accuracy: 0.6547 - val_loss: 1.2910 - val_accuracy: 0.6317 - lr: 0.0010\n","Epoch 30/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9952 - accuracy: 0.6763 - val_loss: 1.3995 - val_accuracy: 0.6132 - lr: 0.0010\n","Epoch 31/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9722 - accuracy: 0.6706 - val_loss: 1.3734 - val_accuracy: 0.6132 - lr: 0.0010\n","Epoch 32/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9660 - accuracy: 0.6758 - val_loss: 1.2123 - val_accuracy: 0.6214 - lr: 0.0010\n","Epoch 33/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9572 - accuracy: 0.6824 - val_loss: 1.2327 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 34/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9297 - accuracy: 0.6886 - val_loss: 1.3042 - val_accuracy: 0.6358 - lr: 0.0010\n","Epoch 35/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9225 - accuracy: 0.6876 - val_loss: 1.2592 - val_accuracy: 0.6399 - lr: 0.0010\n","Epoch 36/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9331 - accuracy: 0.6886 - val_loss: 1.2722 - val_accuracy: 0.6399 - lr: 0.0010\n","Epoch 37/200\n","31/31 [==============================] - 0s 11ms/step - loss: 0.8946 - accuracy: 0.7010 - val_loss: 1.2759 - val_accuracy: 0.6502 - lr: 0.0010\n","Epoch 38/200\n","31/31 [==============================] - 0s 11ms/step - loss: 0.9417 - accuracy: 0.6886 - val_loss: 1.3014 - val_accuracy: 0.6461 - lr: 0.0010\n","Epoch 39/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9312 - accuracy: 0.6912 - val_loss: 1.2837 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 40/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9115 - accuracy: 0.6897 - val_loss: 1.2630 - val_accuracy: 0.6358 - lr: 0.0010\n","Epoch 41/200\n","31/31 [==============================] - 0s 11ms/step - loss: 0.9181 - accuracy: 0.6933 - val_loss: 1.2277 - val_accuracy: 0.6379 - lr: 0.0010\n","Epoch 42/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8843 - accuracy: 0.7092 - val_loss: 1.2559 - val_accuracy: 0.6276 - lr: 0.0010\n","Epoch 43/200\n","31/31 [==============================] - 0s 11ms/step - loss: 0.8530 - accuracy: 0.7113 - val_loss: 1.3514 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 44/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8249 - accuracy: 0.7277 - val_loss: 1.2620 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 45/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8344 - accuracy: 0.7195 - val_loss: 1.3026 - val_accuracy: 0.6214 - lr: 0.0010\n","Epoch 46/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8268 - accuracy: 0.7210 - val_loss: 1.3187 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 47/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8195 - accuracy: 0.7319 - val_loss: 1.3173 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 48/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8328 - accuracy: 0.7159 - val_loss: 1.2840 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 49/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7949 - accuracy: 0.7355 - val_loss: 1.3569 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 50/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7959 - accuracy: 0.7334 - val_loss: 1.2612 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 51/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7841 - accuracy: 0.7416 - val_loss: 1.3628 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 52/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7889 - accuracy: 0.7375 - val_loss: 1.4077 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 53/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7705 - accuracy: 0.7458 - val_loss: 1.3787 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 54/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7547 - accuracy: 0.7442 - val_loss: 1.3886 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 55/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7937 - accuracy: 0.7432 - val_loss: 1.2949 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 56/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8118 - accuracy: 0.7396 - val_loss: 1.2934 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 57/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7690 - accuracy: 0.7432 - val_loss: 1.2346 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 58/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7563 - accuracy: 0.7432 - val_loss: 1.4147 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 59/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8891 - accuracy: 0.7416 - val_loss: 1.3118 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 60/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8905 - accuracy: 0.7324 - val_loss: 1.1055 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 61/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7492 - accuracy: 0.7432 - val_loss: 1.1876 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 62/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7242 - accuracy: 0.7633 - val_loss: 1.2281 - val_accuracy: 0.6852 - lr: 0.0010\n","Epoch 63/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7002 - accuracy: 0.7689 - val_loss: 1.2617 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 64/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7481 - accuracy: 0.7494 - val_loss: 1.2210 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 65/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7088 - accuracy: 0.7627 - val_loss: 1.2654 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 66/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6778 - accuracy: 0.7869 - val_loss: 1.2733 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 67/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6764 - accuracy: 0.7771 - val_loss: 1.3081 - val_accuracy: 0.6914 - lr: 0.0010\n","Epoch 68/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7136 - accuracy: 0.7715 - val_loss: 1.2988 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 69/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6972 - accuracy: 0.7787 - val_loss: 1.2821 - val_accuracy: 0.6646 - lr: 0.0010\n","Epoch 70/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6591 - accuracy: 0.7905 - val_loss: 1.2670 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 71/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6853 - accuracy: 0.7864 - val_loss: 1.3910 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 72/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6742 - accuracy: 0.7813 - val_loss: 1.3496 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 73/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7513 - accuracy: 0.7602 - val_loss: 1.1251 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 74/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7379 - accuracy: 0.7622 - val_loss: 1.2512 - val_accuracy: 0.6317 - lr: 0.0010\n","Epoch 75/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7172 - accuracy: 0.7669 - val_loss: 1.3814 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 76/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7330 - accuracy: 0.7669 - val_loss: 1.1721 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 77/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6586 - accuracy: 0.7921 - val_loss: 1.2340 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 78/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6199 - accuracy: 0.8029 - val_loss: 1.2386 - val_accuracy: 0.7016 - lr: 0.0010\n","Epoch 79/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6448 - accuracy: 0.7874 - val_loss: 1.2712 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 80/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6465 - accuracy: 0.7926 - val_loss: 1.2140 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 81/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6373 - accuracy: 0.8034 - val_loss: 1.2502 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 82/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6141 - accuracy: 0.8008 - val_loss: 1.2419 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 83/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5581 - accuracy: 0.8240 - val_loss: 1.3546 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 84/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5641 - accuracy: 0.8199 - val_loss: 1.3960 - val_accuracy: 0.6934 - lr: 0.0010\n","Epoch 85/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5676 - accuracy: 0.8168 - val_loss: 1.4378 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 86/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5765 - accuracy: 0.8116 - val_loss: 1.4024 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 87/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6192 - accuracy: 0.8111 - val_loss: 1.3509 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 88/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6573 - accuracy: 0.7869 - val_loss: 1.3747 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 89/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5937 - accuracy: 0.8091 - val_loss: 1.3613 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 90/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5650 - accuracy: 0.8250 - val_loss: 1.4430 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 91/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5924 - accuracy: 0.8250 - val_loss: 1.2382 - val_accuracy: 0.7037 - lr: 0.0010\n","Epoch 92/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5518 - accuracy: 0.8374 - val_loss: 1.2268 - val_accuracy: 0.6852 - lr: 0.0010\n","Epoch 93/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6034 - accuracy: 0.8209 - val_loss: 1.2908 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 94/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5602 - accuracy: 0.8281 - val_loss: 1.3897 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 95/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5196 - accuracy: 0.8281 - val_loss: 1.2712 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 96/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4979 - accuracy: 0.8523 - val_loss: 1.4071 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 97/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4975 - accuracy: 0.8487 - val_loss: 1.4649 - val_accuracy: 0.6914 - lr: 0.0010\n","Epoch 98/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4939 - accuracy: 0.8507 - val_loss: 1.4894 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 99/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5757 - accuracy: 0.8240 - val_loss: 1.3103 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 100/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5751 - accuracy: 0.8168 - val_loss: 1.3870 - val_accuracy: 0.6934 - lr: 0.0010\n","Epoch 101/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5195 - accuracy: 0.8343 - val_loss: 1.3987 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 102/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4729 - accuracy: 0.8580 - val_loss: 1.4069 - val_accuracy: 0.6914 - lr: 0.0010\n","Epoch 103/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4726 - accuracy: 0.8652 - val_loss: 1.4263 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 104/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4677 - accuracy: 0.8605 - val_loss: 1.6853 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 105/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4294 - accuracy: 0.8765 - val_loss: 1.6897 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 106/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4881 - accuracy: 0.8595 - val_loss: 1.4659 - val_accuracy: 0.6934 - lr: 0.0010\n","Epoch 107/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4375 - accuracy: 0.8662 - val_loss: 1.5255 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 108/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4195 - accuracy: 0.8770 - val_loss: 1.4626 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 109/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3814 - accuracy: 0.8883 - val_loss: 1.5552 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 110/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3731 - accuracy: 0.8899 - val_loss: 1.5533 - val_accuracy: 0.7037 - lr: 5.0000e-04\n","Epoch 111/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3634 - accuracy: 0.8940 - val_loss: 1.5206 - val_accuracy: 0.6852 - lr: 5.0000e-04\n","Epoch 112/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3763 - accuracy: 0.8899 - val_loss: 1.7205 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 113/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3706 - accuracy: 0.8909 - val_loss: 1.7211 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 114/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3809 - accuracy: 0.8888 - val_loss: 1.6533 - val_accuracy: 0.6893 - lr: 5.0000e-04\n","Epoch 115/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3733 - accuracy: 0.8857 - val_loss: 1.7581 - val_accuracy: 0.6790 - lr: 5.0000e-04\n","Epoch 116/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3701 - accuracy: 0.8976 - val_loss: 1.7291 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 117/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3645 - accuracy: 0.8940 - val_loss: 1.6828 - val_accuracy: 0.6749 - lr: 5.0000e-04\n","Epoch 118/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3422 - accuracy: 0.9058 - val_loss: 1.7674 - val_accuracy: 0.7078 - lr: 5.0000e-04\n","Epoch 119/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3404 - accuracy: 0.8945 - val_loss: 1.8177 - val_accuracy: 0.7058 - lr: 5.0000e-04\n","Epoch 120/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3303 - accuracy: 0.9032 - val_loss: 1.8459 - val_accuracy: 0.6893 - lr: 5.0000e-04\n","Epoch 121/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3284 - accuracy: 0.9104 - val_loss: 1.8652 - val_accuracy: 0.6955 - lr: 5.0000e-04\n","Epoch 122/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3145 - accuracy: 0.9115 - val_loss: 1.8862 - val_accuracy: 0.6811 - lr: 5.0000e-04\n","Epoch 123/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3549 - accuracy: 0.9063 - val_loss: 1.8479 - val_accuracy: 0.6831 - lr: 5.0000e-04\n","Epoch 124/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3934 - accuracy: 0.8981 - val_loss: 1.8565 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 125/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3734 - accuracy: 0.8955 - val_loss: 1.7094 - val_accuracy: 0.6893 - lr: 5.0000e-04\n","Epoch 126/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3372 - accuracy: 0.9068 - val_loss: 1.7805 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 127/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3309 - accuracy: 0.9079 - val_loss: 1.7409 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 128/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3175 - accuracy: 0.9130 - val_loss: 1.9454 - val_accuracy: 0.6893 - lr: 5.0000e-04\n","Epoch 129/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3074 - accuracy: 0.9146 - val_loss: 1.8406 - val_accuracy: 0.6996 - lr: 5.0000e-04\n","Epoch 130/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3051 - accuracy: 0.9228 - val_loss: 1.9223 - val_accuracy: 0.6831 - lr: 5.0000e-04\n","Epoch 131/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3194 - accuracy: 0.9094 - val_loss: 1.8798 - val_accuracy: 0.6872 - lr: 5.0000e-04\n","Epoch 132/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3087 - accuracy: 0.9141 - val_loss: 1.8287 - val_accuracy: 0.6955 - lr: 5.0000e-04\n","Epoch 133/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3175 - accuracy: 0.9110 - val_loss: 1.9379 - val_accuracy: 0.6934 - lr: 5.0000e-04\n","Epoch 134/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3042 - accuracy: 0.9156 - val_loss: 1.8970 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 135/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.2803 - accuracy: 0.9259 - val_loss: 1.9025 - val_accuracy: 0.6934 - lr: 2.5000e-04\n","Epoch 136/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.2751 - accuracy: 0.9238 - val_loss: 1.9194 - val_accuracy: 0.6893 - lr: 2.5000e-04\n","Epoch 137/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.2710 - accuracy: 0.9249 - val_loss: 1.9059 - val_accuracy: 0.7058 - lr: 2.5000e-04\n","Epoch 138/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.2627 - accuracy: 0.9249 - val_loss: 1.9563 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 139/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.2602 - accuracy: 0.9326 - val_loss: 1.9950 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 140/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.2691 - accuracy: 0.9274 - val_loss: 2.0267 - val_accuracy: 0.6914 - lr: 2.5000e-04\n","Epoch 141/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.2639 - accuracy: 0.9326 - val_loss: 2.0074 - val_accuracy: 0.7016 - lr: 2.5000e-04\n","Epoch 142/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.2727 - accuracy: 0.9295 - val_loss: 2.0790 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 143/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.2597 - accuracy: 0.9321 - val_loss: 2.1023 - val_accuracy: 0.6934 - lr: 2.5000e-04\n","Epoch 144/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.2598 - accuracy: 0.9300 - val_loss: 2.0544 - val_accuracy: 0.6893 - lr: 2.5000e-04\n","Epoch 145/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.2644 - accuracy: 0.9264 - val_loss: 2.0569 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 146/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.2613 - accuracy: 0.9295 - val_loss: 2.0817 - val_accuracy: 0.6934 - lr: 2.5000e-04\n","Epoch 147/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.2492 - accuracy: 0.9352 - val_loss: 2.0643 - val_accuracy: 0.6893 - lr: 2.5000e-04\n","Epoch 148/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.2575 - accuracy: 0.9290 - val_loss: 2.0585 - val_accuracy: 0.6996 - lr: 2.5000e-04\n"]}]},{"cell_type":"code","source":["history9 = model9.fit(\n","    x = x_train9,\n","    y = y_train9,\n","    batch_size = 64,\n","    epochs = 200,\n","    validation_data = (x_test9, y_test9),\n","    callbacks = [\n","        tfk.callbacks.EarlyStopping(monitor='val_accuracy', mode='max', patience=30, restore_best_weights=True),\n","        tfk.callbacks.ReduceLROnPlateau(monitor='val_accuracy', mode='max', patience=15, factor=0.5, min_lr=1e-5)\n","    ]\n",").history"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"ZEMxC6eDwRCk","executionInfo":{"status":"ok","timestamp":1671208803070,"user_tz":-60,"elapsed":77800,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"e9003683-34ca-471a-95cf-a4da85c95096"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/200\n","31/31 [==============================] - 1s 19ms/step - loss: 2.2761 - accuracy: 0.3227 - val_loss: 1.9394 - val_accuracy: 0.3786 - lr: 0.0010\n","Epoch 2/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.9174 - accuracy: 0.3757 - val_loss: 1.8481 - val_accuracy: 0.4012 - lr: 0.0010\n","Epoch 3/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.8430 - accuracy: 0.4143 - val_loss: 1.7977 - val_accuracy: 0.4177 - lr: 0.0010\n","Epoch 4/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.7121 - accuracy: 0.4375 - val_loss: 1.6733 - val_accuracy: 0.4568 - lr: 0.0010\n","Epoch 5/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.6355 - accuracy: 0.4611 - val_loss: 1.6315 - val_accuracy: 0.4506 - lr: 0.0010\n","Epoch 6/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.5115 - accuracy: 0.4910 - val_loss: 1.5150 - val_accuracy: 0.5473 - lr: 0.0010\n","Epoch 7/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.4911 - accuracy: 0.5039 - val_loss: 1.4547 - val_accuracy: 0.5350 - lr: 0.0010\n","Epoch 8/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.4340 - accuracy: 0.5142 - val_loss: 1.4082 - val_accuracy: 0.5473 - lr: 0.0010\n","Epoch 9/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3881 - accuracy: 0.5466 - val_loss: 1.4773 - val_accuracy: 0.4733 - lr: 0.0010\n","Epoch 10/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.4143 - accuracy: 0.5224 - val_loss: 1.5196 - val_accuracy: 0.5000 - lr: 0.0010\n","Epoch 11/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.3227 - accuracy: 0.5579 - val_loss: 1.3484 - val_accuracy: 0.5761 - lr: 0.0010\n","Epoch 12/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.2717 - accuracy: 0.5852 - val_loss: 1.2765 - val_accuracy: 0.5782 - lr: 0.0010\n","Epoch 13/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2740 - accuracy: 0.5841 - val_loss: 1.3349 - val_accuracy: 0.5535 - lr: 0.0010\n","Epoch 14/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2272 - accuracy: 0.5836 - val_loss: 1.2317 - val_accuracy: 0.6152 - lr: 0.0010\n","Epoch 15/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1819 - accuracy: 0.6047 - val_loss: 1.3072 - val_accuracy: 0.5700 - lr: 0.0010\n","Epoch 16/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2331 - accuracy: 0.6006 - val_loss: 1.1847 - val_accuracy: 0.6111 - lr: 0.0010\n","Epoch 17/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1791 - accuracy: 0.6089 - val_loss: 1.2148 - val_accuracy: 0.5905 - lr: 0.0010\n","Epoch 18/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.2156 - accuracy: 0.6042 - val_loss: 1.2764 - val_accuracy: 0.5658 - lr: 0.0010\n","Epoch 19/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1248 - accuracy: 0.6217 - val_loss: 1.1466 - val_accuracy: 0.6255 - lr: 0.0010\n","Epoch 20/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1427 - accuracy: 0.6186 - val_loss: 1.1786 - val_accuracy: 0.6337 - lr: 0.0010\n","Epoch 21/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1595 - accuracy: 0.6274 - val_loss: 1.2566 - val_accuracy: 0.6235 - lr: 0.0010\n","Epoch 22/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.1019 - accuracy: 0.6423 - val_loss: 1.1763 - val_accuracy: 0.5905 - lr: 0.0010\n","Epoch 23/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0589 - accuracy: 0.6490 - val_loss: 1.2736 - val_accuracy: 0.6070 - lr: 0.0010\n","Epoch 24/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0535 - accuracy: 0.6567 - val_loss: 1.1687 - val_accuracy: 0.6399 - lr: 0.0010\n","Epoch 25/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0457 - accuracy: 0.6572 - val_loss: 1.1910 - val_accuracy: 0.6173 - lr: 0.0010\n","Epoch 26/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0272 - accuracy: 0.6624 - val_loss: 1.1709 - val_accuracy: 0.6296 - lr: 0.0010\n","Epoch 27/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0215 - accuracy: 0.6603 - val_loss: 1.1173 - val_accuracy: 0.6420 - lr: 0.0010\n","Epoch 28/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0139 - accuracy: 0.6758 - val_loss: 1.1267 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 29/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9914 - accuracy: 0.6644 - val_loss: 1.1211 - val_accuracy: 0.6502 - lr: 0.0010\n","Epoch 30/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9946 - accuracy: 0.6701 - val_loss: 1.2343 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 31/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0263 - accuracy: 0.6665 - val_loss: 1.0495 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 32/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0552 - accuracy: 0.6402 - val_loss: 1.1282 - val_accuracy: 0.6276 - lr: 0.0010\n","Epoch 33/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9736 - accuracy: 0.6716 - val_loss: 1.0736 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 34/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9489 - accuracy: 0.6886 - val_loss: 1.0517 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 35/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9308 - accuracy: 0.6897 - val_loss: 1.1705 - val_accuracy: 0.6440 - lr: 0.0010\n","Epoch 36/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9536 - accuracy: 0.6737 - val_loss: 1.0595 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 37/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9093 - accuracy: 0.7041 - val_loss: 1.0731 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 38/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8731 - accuracy: 0.7087 - val_loss: 1.0556 - val_accuracy: 0.6811 - lr: 0.0010\n","Epoch 39/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9179 - accuracy: 0.6943 - val_loss: 1.0949 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 40/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9492 - accuracy: 0.6840 - val_loss: 1.0581 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 41/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8982 - accuracy: 0.7015 - val_loss: 1.0725 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 42/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9202 - accuracy: 0.6994 - val_loss: 1.3448 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 43/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8839 - accuracy: 0.7118 - val_loss: 1.0522 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 44/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8457 - accuracy: 0.7159 - val_loss: 1.0827 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 45/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8676 - accuracy: 0.7092 - val_loss: 1.0860 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 46/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8247 - accuracy: 0.7241 - val_loss: 1.0376 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 47/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8098 - accuracy: 0.7252 - val_loss: 1.1067 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 48/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8289 - accuracy: 0.7324 - val_loss: 1.0118 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 49/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8243 - accuracy: 0.7231 - val_loss: 1.0997 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 50/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8531 - accuracy: 0.7267 - val_loss: 1.1626 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 51/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8263 - accuracy: 0.7216 - val_loss: 1.0934 - val_accuracy: 0.7058 - lr: 0.0010\n","Epoch 52/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8090 - accuracy: 0.7355 - val_loss: 1.1068 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 53/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9750 - accuracy: 0.7056 - val_loss: 1.2733 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 54/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9672 - accuracy: 0.7113 - val_loss: 1.2303 - val_accuracy: 0.6173 - lr: 0.0010\n","Epoch 55/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.9705 - accuracy: 0.6799 - val_loss: 1.1013 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 56/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9553 - accuracy: 0.6984 - val_loss: 0.9889 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 57/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8171 - accuracy: 0.7293 - val_loss: 0.9665 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 58/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7785 - accuracy: 0.7442 - val_loss: 1.1026 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 59/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7836 - accuracy: 0.7437 - val_loss: 1.0198 - val_accuracy: 0.7140 - lr: 0.0010\n","Epoch 60/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7508 - accuracy: 0.7591 - val_loss: 1.0346 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 61/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7517 - accuracy: 0.7524 - val_loss: 1.0860 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 62/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7532 - accuracy: 0.7586 - val_loss: 1.0458 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 63/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7164 - accuracy: 0.7535 - val_loss: 1.0206 - val_accuracy: 0.7058 - lr: 0.0010\n","Epoch 64/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7869 - accuracy: 0.7530 - val_loss: 1.2274 - val_accuracy: 0.6667 - lr: 0.0010\n","Epoch 65/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7738 - accuracy: 0.7406 - val_loss: 1.0956 - val_accuracy: 0.7016 - lr: 0.0010\n","Epoch 66/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7237 - accuracy: 0.7638 - val_loss: 1.0625 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 67/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7734 - accuracy: 0.7509 - val_loss: 1.2026 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 68/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7540 - accuracy: 0.7478 - val_loss: 1.0820 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 69/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7254 - accuracy: 0.7674 - val_loss: 1.0397 - val_accuracy: 0.6975 - lr: 0.0010\n","Epoch 70/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7237 - accuracy: 0.7648 - val_loss: 1.0852 - val_accuracy: 0.7078 - lr: 0.0010\n","Epoch 71/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6743 - accuracy: 0.7735 - val_loss: 1.1195 - val_accuracy: 0.6975 - lr: 0.0010\n","Epoch 72/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6513 - accuracy: 0.7828 - val_loss: 1.1584 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 73/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6446 - accuracy: 0.7895 - val_loss: 1.1545 - val_accuracy: 0.7016 - lr: 0.0010\n","Epoch 74/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7147 - accuracy: 0.7741 - val_loss: 1.4339 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 75/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7423 - accuracy: 0.7576 - val_loss: 1.1264 - val_accuracy: 0.7078 - lr: 5.0000e-04\n","Epoch 76/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6506 - accuracy: 0.7854 - val_loss: 1.0959 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 77/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6039 - accuracy: 0.8065 - val_loss: 1.0551 - val_accuracy: 0.7222 - lr: 5.0000e-04\n","Epoch 78/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5894 - accuracy: 0.8116 - val_loss: 1.1095 - val_accuracy: 0.7202 - lr: 5.0000e-04\n","Epoch 79/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5842 - accuracy: 0.8080 - val_loss: 1.0866 - val_accuracy: 0.7119 - lr: 5.0000e-04\n","Epoch 80/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5813 - accuracy: 0.8214 - val_loss: 1.1189 - val_accuracy: 0.7140 - lr: 5.0000e-04\n","Epoch 81/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5883 - accuracy: 0.8049 - val_loss: 1.1258 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 82/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5668 - accuracy: 0.8245 - val_loss: 1.1102 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 83/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5466 - accuracy: 0.8271 - val_loss: 1.1622 - val_accuracy: 0.7078 - lr: 5.0000e-04\n","Epoch 84/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5581 - accuracy: 0.8183 - val_loss: 1.1834 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 85/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5513 - accuracy: 0.8281 - val_loss: 1.1896 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 86/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5650 - accuracy: 0.8178 - val_loss: 1.2383 - val_accuracy: 0.7078 - lr: 5.0000e-04\n","Epoch 87/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5369 - accuracy: 0.8302 - val_loss: 1.1606 - val_accuracy: 0.7181 - lr: 5.0000e-04\n","Epoch 88/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5298 - accuracy: 0.8322 - val_loss: 1.1812 - val_accuracy: 0.7037 - lr: 5.0000e-04\n","Epoch 89/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5252 - accuracy: 0.8379 - val_loss: 1.1626 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 90/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5251 - accuracy: 0.8410 - val_loss: 1.1923 - val_accuracy: 0.7119 - lr: 5.0000e-04\n","Epoch 91/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5084 - accuracy: 0.8487 - val_loss: 1.1305 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 92/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5238 - accuracy: 0.8358 - val_loss: 1.1632 - val_accuracy: 0.7140 - lr: 5.0000e-04\n","Epoch 93/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4960 - accuracy: 0.8466 - val_loss: 1.1953 - val_accuracy: 0.6996 - lr: 2.5000e-04\n","Epoch 94/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4886 - accuracy: 0.8564 - val_loss: 1.1925 - val_accuracy: 0.6955 - lr: 2.5000e-04\n","Epoch 95/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4867 - accuracy: 0.8518 - val_loss: 1.1920 - val_accuracy: 0.7119 - lr: 2.5000e-04\n","Epoch 96/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4791 - accuracy: 0.8543 - val_loss: 1.2000 - val_accuracy: 0.7140 - lr: 2.5000e-04\n","Epoch 97/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4757 - accuracy: 0.8605 - val_loss: 1.2362 - val_accuracy: 0.7037 - lr: 2.5000e-04\n","Epoch 98/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4750 - accuracy: 0.8554 - val_loss: 1.1882 - val_accuracy: 0.7160 - lr: 2.5000e-04\n","Epoch 99/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4739 - accuracy: 0.8554 - val_loss: 1.2198 - val_accuracy: 0.7140 - lr: 2.5000e-04\n","Epoch 100/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4742 - accuracy: 0.8580 - val_loss: 1.2318 - val_accuracy: 0.7181 - lr: 2.5000e-04\n","Epoch 101/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4657 - accuracy: 0.8580 - val_loss: 1.2242 - val_accuracy: 0.7202 - lr: 2.5000e-04\n","Epoch 102/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4651 - accuracy: 0.8574 - val_loss: 1.2046 - val_accuracy: 0.7078 - lr: 2.5000e-04\n","Epoch 103/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4578 - accuracy: 0.8677 - val_loss: 1.2294 - val_accuracy: 0.7243 - lr: 2.5000e-04\n","Epoch 104/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4536 - accuracy: 0.8672 - val_loss: 1.2738 - val_accuracy: 0.7099 - lr: 2.5000e-04\n","Epoch 105/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4518 - accuracy: 0.8657 - val_loss: 1.2632 - val_accuracy: 0.7099 - lr: 2.5000e-04\n","Epoch 106/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4423 - accuracy: 0.8626 - val_loss: 1.2707 - val_accuracy: 0.7058 - lr: 2.5000e-04\n","Epoch 107/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4418 - accuracy: 0.8682 - val_loss: 1.2511 - val_accuracy: 0.7058 - lr: 2.5000e-04\n","Epoch 108/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4453 - accuracy: 0.8626 - val_loss: 1.2492 - val_accuracy: 0.7181 - lr: 2.5000e-04\n","Epoch 109/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4413 - accuracy: 0.8646 - val_loss: 1.2573 - val_accuracy: 0.6934 - lr: 2.5000e-04\n","Epoch 110/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4403 - accuracy: 0.8646 - val_loss: 1.2575 - val_accuracy: 0.6975 - lr: 2.5000e-04\n","Epoch 111/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4354 - accuracy: 0.8713 - val_loss: 1.3033 - val_accuracy: 0.7140 - lr: 2.5000e-04\n","Epoch 112/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4209 - accuracy: 0.8755 - val_loss: 1.2695 - val_accuracy: 0.7078 - lr: 2.5000e-04\n","Epoch 113/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4283 - accuracy: 0.8734 - val_loss: 1.2931 - val_accuracy: 0.7099 - lr: 2.5000e-04\n","Epoch 114/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4286 - accuracy: 0.8657 - val_loss: 1.3149 - val_accuracy: 0.7078 - lr: 2.5000e-04\n","Epoch 115/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4253 - accuracy: 0.8729 - val_loss: 1.2579 - val_accuracy: 0.7099 - lr: 2.5000e-04\n","Epoch 116/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4176 - accuracy: 0.8718 - val_loss: 1.3096 - val_accuracy: 0.7058 - lr: 2.5000e-04\n","Epoch 117/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4259 - accuracy: 0.8693 - val_loss: 1.3139 - val_accuracy: 0.7119 - lr: 2.5000e-04\n","Epoch 118/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4129 - accuracy: 0.8760 - val_loss: 1.3131 - val_accuracy: 0.7119 - lr: 2.5000e-04\n","Epoch 119/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4055 - accuracy: 0.8816 - val_loss: 1.3226 - val_accuracy: 0.7016 - lr: 1.2500e-04\n","Epoch 120/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3998 - accuracy: 0.8816 - val_loss: 1.3064 - val_accuracy: 0.7099 - lr: 1.2500e-04\n","Epoch 121/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3951 - accuracy: 0.8785 - val_loss: 1.3008 - val_accuracy: 0.7119 - lr: 1.2500e-04\n","Epoch 122/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3954 - accuracy: 0.8780 - val_loss: 1.3235 - val_accuracy: 0.7037 - lr: 1.2500e-04\n","Epoch 123/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3902 - accuracy: 0.8796 - val_loss: 1.3670 - val_accuracy: 0.7058 - lr: 1.2500e-04\n","Epoch 124/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3990 - accuracy: 0.8749 - val_loss: 1.3553 - val_accuracy: 0.7037 - lr: 1.2500e-04\n","Epoch 125/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4009 - accuracy: 0.8755 - val_loss: 1.3326 - val_accuracy: 0.7202 - lr: 1.2500e-04\n","Epoch 126/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4009 - accuracy: 0.8806 - val_loss: 1.3246 - val_accuracy: 0.7078 - lr: 1.2500e-04\n","Epoch 127/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3929 - accuracy: 0.8811 - val_loss: 1.3544 - val_accuracy: 0.7016 - lr: 1.2500e-04\n","Epoch 128/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3934 - accuracy: 0.8837 - val_loss: 1.3458 - val_accuracy: 0.7099 - lr: 1.2500e-04\n","Epoch 129/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3884 - accuracy: 0.8770 - val_loss: 1.3309 - val_accuracy: 0.7078 - lr: 1.2500e-04\n","Epoch 130/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3900 - accuracy: 0.8893 - val_loss: 1.3634 - val_accuracy: 0.7140 - lr: 1.2500e-04\n","Epoch 131/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3803 - accuracy: 0.8852 - val_loss: 1.3369 - val_accuracy: 0.7037 - lr: 1.2500e-04\n","Epoch 132/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3926 - accuracy: 0.8837 - val_loss: 1.3443 - val_accuracy: 0.7016 - lr: 1.2500e-04\n","Epoch 133/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3902 - accuracy: 0.8868 - val_loss: 1.3369 - val_accuracy: 0.7140 - lr: 1.2500e-04\n"]}]},{"cell_type":"code","source":["history10 = model10.fit(\n","    x = x_train10,\n","    y = y_train10,\n","    batch_size = 64,\n","    epochs = 200,\n","    validation_data = (x_test10, y_test10),\n","    callbacks = [\n","        tfk.callbacks.EarlyStopping(monitor='val_accuracy', mode='max', patience=30, restore_best_weights=True),\n","        tfk.callbacks.ReduceLROnPlateau(monitor='val_accuracy', mode='max', patience=15, factor=0.5, min_lr=1e-5)\n","    ]\n",").history"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"vyk24WNlwRAB","executionInfo":{"status":"ok","timestamp":1671208847699,"user_tz":-60,"elapsed":44651,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"2acbd366-9ebb-425c-d87b-54f0c3dc0e19"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/200\n","31/31 [==============================] - 1s 16ms/step - loss: 2.2452 - accuracy: 0.3248 - val_loss: 2.0305 - val_accuracy: 0.3621 - lr: 0.0010\n","Epoch 2/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.9361 - accuracy: 0.3793 - val_loss: 1.9156 - val_accuracy: 0.3971 - lr: 0.0010\n","Epoch 3/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.8540 - accuracy: 0.3932 - val_loss: 1.7960 - val_accuracy: 0.4259 - lr: 0.0010\n","Epoch 4/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.7301 - accuracy: 0.4370 - val_loss: 1.7523 - val_accuracy: 0.4136 - lr: 0.0010\n","Epoch 5/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.6122 - accuracy: 0.4658 - val_loss: 1.7528 - val_accuracy: 0.3971 - lr: 0.0010\n","Epoch 6/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.5569 - accuracy: 0.4797 - val_loss: 1.4865 - val_accuracy: 0.4918 - lr: 0.0010\n","Epoch 7/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.4691 - accuracy: 0.4951 - val_loss: 1.5324 - val_accuracy: 0.4835 - lr: 0.0010\n","Epoch 8/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.4629 - accuracy: 0.5121 - val_loss: 1.4326 - val_accuracy: 0.5309 - lr: 0.0010\n","Epoch 9/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.4067 - accuracy: 0.5332 - val_loss: 1.3907 - val_accuracy: 0.5453 - lr: 0.0010\n","Epoch 10/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.4251 - accuracy: 0.5327 - val_loss: 1.4700 - val_accuracy: 0.5000 - lr: 0.0010\n","Epoch 11/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3843 - accuracy: 0.5404 - val_loss: 1.3916 - val_accuracy: 0.5350 - lr: 0.0010\n","Epoch 12/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.2893 - accuracy: 0.5723 - val_loss: 1.3346 - val_accuracy: 0.5514 - lr: 0.0010\n","Epoch 13/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3792 - accuracy: 0.5425 - val_loss: 1.3889 - val_accuracy: 0.5782 - lr: 0.0010\n","Epoch 14/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.3015 - accuracy: 0.5620 - val_loss: 1.2906 - val_accuracy: 0.6132 - lr: 0.0010\n","Epoch 15/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2224 - accuracy: 0.5934 - val_loss: 1.2352 - val_accuracy: 0.6049 - lr: 0.0010\n","Epoch 16/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1872 - accuracy: 0.5996 - val_loss: 1.2105 - val_accuracy: 0.6193 - lr: 0.0010\n","Epoch 17/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2550 - accuracy: 0.6032 - val_loss: 1.2141 - val_accuracy: 0.5761 - lr: 0.0010\n","Epoch 18/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.2378 - accuracy: 0.5980 - val_loss: 1.2732 - val_accuracy: 0.5617 - lr: 0.0010\n","Epoch 19/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1499 - accuracy: 0.6171 - val_loss: 1.1957 - val_accuracy: 0.6111 - lr: 0.0010\n","Epoch 20/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1003 - accuracy: 0.6325 - val_loss: 1.1558 - val_accuracy: 0.6276 - lr: 0.0010\n","Epoch 21/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.1400 - accuracy: 0.6284 - val_loss: 1.4232 - val_accuracy: 0.5638 - lr: 0.0010\n","Epoch 22/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.1719 - accuracy: 0.6155 - val_loss: 1.2006 - val_accuracy: 0.6111 - lr: 0.0010\n","Epoch 23/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0947 - accuracy: 0.6289 - val_loss: 1.1338 - val_accuracy: 0.6276 - lr: 0.0010\n","Epoch 24/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0649 - accuracy: 0.6377 - val_loss: 1.1628 - val_accuracy: 0.6193 - lr: 0.0010\n","Epoch 25/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0172 - accuracy: 0.6634 - val_loss: 1.1142 - val_accuracy: 0.6070 - lr: 0.0010\n","Epoch 26/200\n","31/31 [==============================] - 0s 10ms/step - loss: 1.0344 - accuracy: 0.6511 - val_loss: 1.1423 - val_accuracy: 0.6358 - lr: 0.0010\n","Epoch 27/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0231 - accuracy: 0.6454 - val_loss: 1.1186 - val_accuracy: 0.6193 - lr: 0.0010\n","Epoch 28/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9946 - accuracy: 0.6706 - val_loss: 1.1119 - val_accuracy: 0.6481 - lr: 0.0010\n","Epoch 29/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0184 - accuracy: 0.6505 - val_loss: 1.1490 - val_accuracy: 0.6070 - lr: 0.0010\n","Epoch 30/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0753 - accuracy: 0.6330 - val_loss: 1.2002 - val_accuracy: 0.6214 - lr: 0.0010\n","Epoch 31/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0081 - accuracy: 0.6670 - val_loss: 1.0982 - val_accuracy: 0.6502 - lr: 0.0010\n","Epoch 32/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9855 - accuracy: 0.6691 - val_loss: 1.1369 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 33/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9812 - accuracy: 0.6768 - val_loss: 1.1933 - val_accuracy: 0.6091 - lr: 0.0010\n","Epoch 34/200\n","31/31 [==============================] - 0s 8ms/step - loss: 1.0546 - accuracy: 0.6516 - val_loss: 1.1593 - val_accuracy: 0.6091 - lr: 0.0010\n","Epoch 35/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9955 - accuracy: 0.6511 - val_loss: 1.0843 - val_accuracy: 0.6564 - lr: 0.0010\n","Epoch 36/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9401 - accuracy: 0.6891 - val_loss: 1.0618 - val_accuracy: 0.6523 - lr: 0.0010\n","Epoch 37/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.9150 - accuracy: 0.6840 - val_loss: 1.0298 - val_accuracy: 0.6502 - lr: 0.0010\n","Epoch 38/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9064 - accuracy: 0.6933 - val_loss: 1.0312 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 39/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8898 - accuracy: 0.7056 - val_loss: 1.0625 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 40/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8994 - accuracy: 0.7030 - val_loss: 1.1225 - val_accuracy: 0.6358 - lr: 0.0010\n","Epoch 41/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9077 - accuracy: 0.6953 - val_loss: 1.0935 - val_accuracy: 0.6461 - lr: 0.0010\n","Epoch 42/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8911 - accuracy: 0.6994 - val_loss: 1.0650 - val_accuracy: 0.6605 - lr: 0.0010\n","Epoch 43/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8766 - accuracy: 0.7138 - val_loss: 1.0247 - val_accuracy: 0.6728 - lr: 0.0010\n","Epoch 44/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8750 - accuracy: 0.7118 - val_loss: 1.0869 - val_accuracy: 0.6399 - lr: 0.0010\n","Epoch 45/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.9147 - accuracy: 0.6927 - val_loss: 1.0946 - val_accuracy: 0.6543 - lr: 0.0010\n","Epoch 46/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8544 - accuracy: 0.7144 - val_loss: 1.0236 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 47/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8416 - accuracy: 0.7216 - val_loss: 1.0427 - val_accuracy: 0.6626 - lr: 0.0010\n","Epoch 48/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.8189 - accuracy: 0.7205 - val_loss: 1.0096 - val_accuracy: 0.6811 - lr: 0.0010\n","Epoch 49/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8012 - accuracy: 0.7319 - val_loss: 1.0291 - val_accuracy: 0.6749 - lr: 0.0010\n","Epoch 50/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8849 - accuracy: 0.7200 - val_loss: 1.1217 - val_accuracy: 0.6502 - lr: 0.0010\n","Epoch 51/200\n","31/31 [==============================] - 0s 9ms/step - loss: 1.0074 - accuracy: 0.6969 - val_loss: 1.0800 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 52/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8919 - accuracy: 0.7097 - val_loss: 1.1680 - val_accuracy: 0.6399 - lr: 0.0010\n","Epoch 53/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.8506 - accuracy: 0.7159 - val_loss: 1.0773 - val_accuracy: 0.6770 - lr: 0.0010\n","Epoch 54/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8251 - accuracy: 0.7226 - val_loss: 1.0422 - val_accuracy: 0.6687 - lr: 0.0010\n","Epoch 55/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7801 - accuracy: 0.7391 - val_loss: 1.0401 - val_accuracy: 0.6790 - lr: 0.0010\n","Epoch 56/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7690 - accuracy: 0.7344 - val_loss: 1.0431 - val_accuracy: 0.6708 - lr: 0.0010\n","Epoch 57/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7334 - accuracy: 0.7612 - val_loss: 1.0479 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 58/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7589 - accuracy: 0.7488 - val_loss: 1.0264 - val_accuracy: 0.6914 - lr: 0.0010\n","Epoch 59/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7487 - accuracy: 0.7530 - val_loss: 1.0359 - val_accuracy: 0.7037 - lr: 0.0010\n","Epoch 60/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7741 - accuracy: 0.7560 - val_loss: 1.1078 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 61/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.8015 - accuracy: 0.7334 - val_loss: 1.0619 - val_accuracy: 0.6914 - lr: 0.0010\n","Epoch 62/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7337 - accuracy: 0.7540 - val_loss: 1.0357 - val_accuracy: 0.7037 - lr: 0.0010\n","Epoch 63/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7516 - accuracy: 0.7509 - val_loss: 1.0824 - val_accuracy: 0.6934 - lr: 0.0010\n","Epoch 64/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7224 - accuracy: 0.7617 - val_loss: 1.0698 - val_accuracy: 0.6831 - lr: 0.0010\n","Epoch 65/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7278 - accuracy: 0.7576 - val_loss: 1.0698 - val_accuracy: 0.6852 - lr: 0.0010\n","Epoch 66/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.7488 - accuracy: 0.7519 - val_loss: 1.0723 - val_accuracy: 0.6584 - lr: 0.0010\n","Epoch 67/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.7212 - accuracy: 0.7648 - val_loss: 1.0988 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 68/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6800 - accuracy: 0.7813 - val_loss: 0.9968 - val_accuracy: 0.7099 - lr: 0.0010\n","Epoch 69/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6636 - accuracy: 0.7849 - val_loss: 1.0383 - val_accuracy: 0.7078 - lr: 0.0010\n","Epoch 70/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6652 - accuracy: 0.7885 - val_loss: 1.0322 - val_accuracy: 0.6996 - lr: 0.0010\n","Epoch 71/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6441 - accuracy: 0.7926 - val_loss: 1.1331 - val_accuracy: 0.6893 - lr: 0.0010\n","Epoch 72/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.6849 - accuracy: 0.7833 - val_loss: 1.0377 - val_accuracy: 0.7058 - lr: 0.0010\n","Epoch 73/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6788 - accuracy: 0.7782 - val_loss: 1.0461 - val_accuracy: 0.6975 - lr: 0.0010\n","Epoch 74/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6282 - accuracy: 0.7957 - val_loss: 1.0930 - val_accuracy: 0.6934 - lr: 0.0010\n","Epoch 75/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6416 - accuracy: 0.7838 - val_loss: 1.0416 - val_accuracy: 0.7078 - lr: 0.0010\n","Epoch 76/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.6049 - accuracy: 0.8060 - val_loss: 1.0832 - val_accuracy: 0.7160 - lr: 0.0010\n","Epoch 77/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6715 - accuracy: 0.7813 - val_loss: 1.1161 - val_accuracy: 0.6811 - lr: 0.0010\n","Epoch 78/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.7163 - accuracy: 0.7756 - val_loss: 1.0668 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 79/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6166 - accuracy: 0.8019 - val_loss: 1.1079 - val_accuracy: 0.6872 - lr: 0.0010\n","Epoch 80/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6051 - accuracy: 0.7988 - val_loss: 1.1129 - val_accuracy: 0.7078 - lr: 0.0010\n","Epoch 81/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5794 - accuracy: 0.8199 - val_loss: 1.0567 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 82/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5770 - accuracy: 0.8142 - val_loss: 1.1061 - val_accuracy: 0.6996 - lr: 0.0010\n","Epoch 83/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5973 - accuracy: 0.8085 - val_loss: 1.1208 - val_accuracy: 0.7099 - lr: 0.0010\n","Epoch 84/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5755 - accuracy: 0.8106 - val_loss: 1.0913 - val_accuracy: 0.7202 - lr: 0.0010\n","Epoch 85/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6034 - accuracy: 0.8121 - val_loss: 1.2087 - val_accuracy: 0.7016 - lr: 0.0010\n","Epoch 86/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.6168 - accuracy: 0.8111 - val_loss: 1.1704 - val_accuracy: 0.6934 - lr: 0.0010\n","Epoch 87/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5506 - accuracy: 0.8322 - val_loss: 1.0465 - val_accuracy: 0.7099 - lr: 0.0010\n","Epoch 88/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5311 - accuracy: 0.8327 - val_loss: 1.1407 - val_accuracy: 0.7099 - lr: 0.0010\n","Epoch 89/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.5195 - accuracy: 0.8379 - val_loss: 1.1803 - val_accuracy: 0.7058 - lr: 0.0010\n","Epoch 90/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.5173 - accuracy: 0.8369 - val_loss: 1.2339 - val_accuracy: 0.7037 - lr: 0.0010\n","Epoch 91/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5193 - accuracy: 0.8332 - val_loss: 1.2512 - val_accuracy: 0.7037 - lr: 0.0010\n","Epoch 92/200\n","31/31 [==============================] - 0s 14ms/step - loss: 0.5172 - accuracy: 0.8286 - val_loss: 1.1857 - val_accuracy: 0.7202 - lr: 0.0010\n","Epoch 93/200\n","31/31 [==============================] - 0s 14ms/step - loss: 0.4892 - accuracy: 0.8477 - val_loss: 1.2075 - val_accuracy: 0.7037 - lr: 0.0010\n","Epoch 94/200\n","31/31 [==============================] - 0s 14ms/step - loss: 0.4619 - accuracy: 0.8585 - val_loss: 1.3013 - val_accuracy: 0.7140 - lr: 0.0010\n","Epoch 95/200\n","31/31 [==============================] - 0s 14ms/step - loss: 0.4656 - accuracy: 0.8616 - val_loss: 1.2348 - val_accuracy: 0.6996 - lr: 0.0010\n","Epoch 96/200\n","31/31 [==============================] - 0s 13ms/step - loss: 0.5111 - accuracy: 0.8332 - val_loss: 1.3068 - val_accuracy: 0.7140 - lr: 0.0010\n","Epoch 97/200\n","31/31 [==============================] - 0s 15ms/step - loss: 0.4946 - accuracy: 0.8446 - val_loss: 1.2950 - val_accuracy: 0.7140 - lr: 0.0010\n","Epoch 98/200\n","31/31 [==============================] - 0s 12ms/step - loss: 0.4895 - accuracy: 0.8523 - val_loss: 1.3197 - val_accuracy: 0.6955 - lr: 0.0010\n","Epoch 99/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.5218 - accuracy: 0.8250 - val_loss: 1.2564 - val_accuracy: 0.6852 - lr: 0.0010\n","Epoch 100/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4367 - accuracy: 0.8667 - val_loss: 1.1815 - val_accuracy: 0.7202 - lr: 5.0000e-04\n","Epoch 101/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4152 - accuracy: 0.8770 - val_loss: 1.1999 - val_accuracy: 0.7222 - lr: 5.0000e-04\n","Epoch 102/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4013 - accuracy: 0.8811 - val_loss: 1.2523 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 103/200\n","31/31 [==============================] - 0s 13ms/step - loss: 0.3930 - accuracy: 0.8770 - val_loss: 1.3112 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 104/200\n","31/31 [==============================] - 0s 16ms/step - loss: 0.3962 - accuracy: 0.8816 - val_loss: 1.3025 - val_accuracy: 0.7263 - lr: 5.0000e-04\n","Epoch 105/200\n","31/31 [==============================] - 0s 15ms/step - loss: 0.3795 - accuracy: 0.8919 - val_loss: 1.3927 - val_accuracy: 0.6996 - lr: 5.0000e-04\n","Epoch 106/200\n","31/31 [==============================] - 0s 13ms/step - loss: 0.3812 - accuracy: 0.8914 - val_loss: 1.3985 - val_accuracy: 0.7263 - lr: 5.0000e-04\n","Epoch 107/200\n","31/31 [==============================] - 0s 14ms/step - loss: 0.3732 - accuracy: 0.8873 - val_loss: 1.3105 - val_accuracy: 0.7222 - lr: 5.0000e-04\n","Epoch 108/200\n","31/31 [==============================] - 1s 19ms/step - loss: 0.3780 - accuracy: 0.8935 - val_loss: 1.4003 - val_accuracy: 0.7202 - lr: 5.0000e-04\n","Epoch 109/200\n","31/31 [==============================] - 1s 19ms/step - loss: 0.3799 - accuracy: 0.8863 - val_loss: 1.3763 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 110/200\n","31/31 [==============================] - 0s 15ms/step - loss: 0.3764 - accuracy: 0.8878 - val_loss: 1.4510 - val_accuracy: 0.7263 - lr: 5.0000e-04\n","Epoch 111/200\n","31/31 [==============================] - 0s 13ms/step - loss: 0.3952 - accuracy: 0.8852 - val_loss: 1.3510 - val_accuracy: 0.6914 - lr: 5.0000e-04\n","Epoch 112/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.4142 - accuracy: 0.8755 - val_loss: 1.3345 - val_accuracy: 0.7078 - lr: 5.0000e-04\n","Epoch 113/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.4080 - accuracy: 0.8765 - val_loss: 1.5018 - val_accuracy: 0.6975 - lr: 5.0000e-04\n","Epoch 114/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.4042 - accuracy: 0.8940 - val_loss: 1.3993 - val_accuracy: 0.7058 - lr: 5.0000e-04\n","Epoch 115/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3658 - accuracy: 0.8950 - val_loss: 1.3701 - val_accuracy: 0.7181 - lr: 5.0000e-04\n","Epoch 116/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3540 - accuracy: 0.8950 - val_loss: 1.4019 - val_accuracy: 0.7078 - lr: 5.0000e-04\n","Epoch 117/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3443 - accuracy: 0.8976 - val_loss: 1.4086 - val_accuracy: 0.7305 - lr: 5.0000e-04\n","Epoch 118/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3381 - accuracy: 0.9068 - val_loss: 1.4406 - val_accuracy: 0.7140 - lr: 5.0000e-04\n","Epoch 119/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3249 - accuracy: 0.9120 - val_loss: 1.4236 - val_accuracy: 0.7181 - lr: 5.0000e-04\n","Epoch 120/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3421 - accuracy: 0.9032 - val_loss: 1.4224 - val_accuracy: 0.7037 - lr: 5.0000e-04\n","Epoch 121/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3327 - accuracy: 0.9074 - val_loss: 1.5205 - val_accuracy: 0.7078 - lr: 5.0000e-04\n","Epoch 122/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3326 - accuracy: 0.9079 - val_loss: 1.4655 - val_accuracy: 0.7222 - lr: 5.0000e-04\n","Epoch 123/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3219 - accuracy: 0.9038 - val_loss: 1.5251 - val_accuracy: 0.7160 - lr: 5.0000e-04\n","Epoch 124/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3244 - accuracy: 0.9079 - val_loss: 1.5096 - val_accuracy: 0.7140 - lr: 5.0000e-04\n","Epoch 125/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3210 - accuracy: 0.9110 - val_loss: 1.5002 - val_accuracy: 0.7058 - lr: 5.0000e-04\n","Epoch 126/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3035 - accuracy: 0.9207 - val_loss: 1.5568 - val_accuracy: 0.7160 - lr: 5.0000e-04\n","Epoch 127/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3028 - accuracy: 0.9151 - val_loss: 1.5984 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 128/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3009 - accuracy: 0.9228 - val_loss: 1.5795 - val_accuracy: 0.7099 - lr: 5.0000e-04\n","Epoch 129/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.3152 - accuracy: 0.9166 - val_loss: 1.5529 - val_accuracy: 0.7140 - lr: 5.0000e-04\n","Epoch 130/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.3075 - accuracy: 0.9182 - val_loss: 1.6039 - val_accuracy: 0.7181 - lr: 5.0000e-04\n","Epoch 131/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3120 - accuracy: 0.9156 - val_loss: 1.5742 - val_accuracy: 0.7037 - lr: 5.0000e-04\n","Epoch 132/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.3144 - accuracy: 0.9182 - val_loss: 1.5603 - val_accuracy: 0.7078 - lr: 5.0000e-04\n","Epoch 133/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.2854 - accuracy: 0.9202 - val_loss: 1.5628 - val_accuracy: 0.7119 - lr: 2.5000e-04\n","Epoch 134/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.2650 - accuracy: 0.9341 - val_loss: 1.6132 - val_accuracy: 0.7140 - lr: 2.5000e-04\n","Epoch 135/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.2735 - accuracy: 0.9269 - val_loss: 1.6069 - val_accuracy: 0.7140 - lr: 2.5000e-04\n","Epoch 136/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.2664 - accuracy: 0.9295 - val_loss: 1.6513 - val_accuracy: 0.7222 - lr: 2.5000e-04\n","Epoch 137/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.2618 - accuracy: 0.9279 - val_loss: 1.6815 - val_accuracy: 0.7099 - lr: 2.5000e-04\n","Epoch 138/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.2636 - accuracy: 0.9321 - val_loss: 1.7094 - val_accuracy: 0.7078 - lr: 2.5000e-04\n","Epoch 139/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.2576 - accuracy: 0.9367 - val_loss: 1.6878 - val_accuracy: 0.7119 - lr: 2.5000e-04\n","Epoch 140/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.2561 - accuracy: 0.9321 - val_loss: 1.6341 - val_accuracy: 0.7078 - lr: 2.5000e-04\n","Epoch 141/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.2587 - accuracy: 0.9336 - val_loss: 1.6722 - val_accuracy: 0.7160 - lr: 2.5000e-04\n","Epoch 142/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.2463 - accuracy: 0.9398 - val_loss: 1.7159 - val_accuracy: 0.7099 - lr: 2.5000e-04\n","Epoch 143/200\n","31/31 [==============================] - 0s 9ms/step - loss: 0.2493 - accuracy: 0.9346 - val_loss: 1.7462 - val_accuracy: 0.7181 - lr: 2.5000e-04\n","Epoch 144/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.2577 - accuracy: 0.9326 - val_loss: 1.7285 - val_accuracy: 0.7160 - lr: 2.5000e-04\n","Epoch 145/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.2406 - accuracy: 0.9429 - val_loss: 1.7433 - val_accuracy: 0.7119 - lr: 2.5000e-04\n","Epoch 146/200\n","31/31 [==============================] - 0s 8ms/step - loss: 0.2589 - accuracy: 0.9315 - val_loss: 1.7542 - val_accuracy: 0.7181 - lr: 2.5000e-04\n","Epoch 147/200\n","31/31 [==============================] - 0s 10ms/step - loss: 0.2450 - accuracy: 0.9393 - val_loss: 1.7511 - val_accuracy: 0.7181 - lr: 2.5000e-04\n"]}]},{"cell_type":"markdown","source":["# evaluation"],"metadata":{"id":"4g67kGedwESd"}},{"cell_type":"markdown","source":["**evaluation of a single model**"],"metadata":{"id":"LSVaquMTVlYH"}},{"cell_type":"code","metadata":{"id":"rzrqDHgijtwW","colab":{"base_uri":"https://localhost:8080/","height":764},"executionInfo":{"status":"ok","timestamp":1671208849337,"user_tz":-60,"elapsed":1651,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"25be1d7e-ac17-4329-fd5b-3f0015fef78e"},"source":["best_epoch = np.argmax(history1['val_accuracy'])\n","plt.figure(figsize=(17,4))\n","plt.plot(history1['loss'], label='Training loss', alpha=.8, color='#ff7f0e')\n","plt.plot(history1['val_loss'], label='Validation loss', alpha=.9, color='#5a9aa5')\n","plt.axvline(x=best_epoch, label='Best epoch', alpha=.3, ls='--', color='#5a9aa5')\n","plt.title('Categorical Crossentropy')\n","plt.legend()\n","plt.grid(alpha=.3)\n","plt.show()\n","\n","plt.figure(figsize=(17,4))\n","plt.plot(history1['accuracy'], label='Training accuracy', alpha=.8, color='#ff7f0e')\n","plt.plot(history1['val_accuracy'], label='Validation accuracy', alpha=.9, color='#5a9aa5')\n","plt.axvline(x=best_epoch, label='Best epoch', alpha=.3, ls='--', color='#5a9aa5')\n","plt.title('Accuracy')\n","plt.legend()\n","plt.grid(alpha=.3)\n","plt.show()\n","\n","plt.figure(figsize=(18,3))\n","plt.plot(history1['lr'], label='Learning Rate', alpha=.8, color='#ff7f0e')\n","plt.axvline(x=best_epoch, label='Best epoch', alpha=.3, ls='--', color='#5a9aa5')\n","plt.legend()\n","plt.grid(alpha=.3)\n","plt.show()"],"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1224x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 1224x288 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 1296x216 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"code","metadata":{"id":"ZmNPtx9VjtuC"},"source":["#model.save('1DCNN')"],"execution_count":null,"outputs":[]},{"cell_type":"code","metadata":{"id":"fvKkbmnujtrS","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1671208864338,"user_tz":-60,"elapsed":662,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"3d1aef4a-b469-4ae2-f869-124639c8a4a2"},"source":["predictions = model1.predict(x_test1)\n","predictions.shape"],"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["16/16 [==============================] - 0s 9ms/step\n"]},{"output_type":"execute_result","data":{"text/plain":["(486, 12)"]},"metadata":{},"execution_count":58}]},{"cell_type":"code","metadata":{"id":"KRdpbCqCjtpR","colab":{"base_uri":"https://localhost:8080/","height":1000},"executionInfo":{"status":"ok","timestamp":1671208870759,"user_tz":-60,"elapsed":1535,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"282b132b-46e7-4ce2-8141-520af2b9c441"},"source":["metrics = model1.evaluate(x_test1, return_dict=True)\n","print(metrics)\n","\n","labels = y_test1 \n","predictions = model1.predict(x_test1) \n","assert len(labels) == len(predictions)\n","\n","# Compute the confusion matrix\n","cm = confusion_matrix(np.argmax(labels, axis=-1), np.argmax(predictions, axis=-1))\n","\n","# Compute the classification metrics\n","accuracy = accuracy_score(np.argmax(labels, axis=-1), np.argmax(predictions, axis=-1))\n","precision = precision_score(np.argmax(labels, axis=-1), np.argmax(predictions, axis=-1), average='macro')\n","recall = recall_score(np.argmax(labels, axis=-1), np.argmax(predictions, axis=-1), average='macro')\n","f1 = f1_score(np.argmax(labels, axis=-1), np.argmax(predictions, axis=-1), average='macro')\n","f1_classes = f1_score(np.argmax(labels, axis=-1), np.argmax(predictions, axis=-1), average=None)\n","\n","print('Accuracy:',accuracy.round(4))\n","print('Precision:',precision.round(4))\n","print('Recall:',recall.round(4))\n","print('F1:',f1.round(4))\n","print('F1 (per class):' ,f1_classes)\n","\n","# Plot the confusion matrix\n","ig, ax = plt.subplots(figsize=(12,12))\n","ax.matshow(cm, cmap=plt.cm.Blues, alpha=0.3)\n","for i in range(cm.shape[0]):\n","    for j in range(cm.shape[1]):\n","        ax.text(x=j, y=i,s=cm[i, j], va='center', ha='center', size='xx-large')\n","\n","        \n","plt.figure(figsize=(10,8))\n","sns.heatmap(cm.T, cmap='Blues')\n","plt.xlabel('True labels')\n","plt.ylabel('Predicted labels')\n","plt.show()"],"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["16/16 [==============================] - 0s 2ms/step - loss: 0.0533 - accuracy: 0.0000e+00\n","{'loss': 0.05330146849155426, 'accuracy': 0.0}\n","16/16 [==============================] - 0s 2ms/step\n","Accuracy: 0.7181\n","Precision: 0.6846\n","Recall: 0.5991\n","F1: 0.6249\n","F1 (per class): [0.72727273 0.57142857 0.97196262 0.54545455 0.         0.41935484\n"," 0.88888889 0.41666667 0.93617021 0.74846626 0.96551724 0.30769231]\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 864x864 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 720x576 with 2 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["**Model averaging test (obviously the following evaluation is not reliable)**"],"metadata":{"id":"L_oQuh-ByagJ"}},{"cell_type":"code","source":["labels = y_test1 #for example \n","p1 = model1.predict(x_test1)\n","p2 = model2.predict(x_test1)\n","p3 = model3.predict(x_test1)\n","p4 = model4.predict(x_test1)\n","p5 = model5.predict(x_test1)\n","p6 = model6.predict(x_test1)\n","p7 = model7.predict(x_test1)\n","p8 = model8.predict(x_test1)\n","p9 = model9.predict(x_test1)\n","p10 = model10.predict(x_test1)\n","\n","predictions = (p1+p2+p3+p4+p5+p6+p7+p8+p9+p10)/10.0\n","assert len(labels) == len(predictions)\n","\n","# Compute the confusion matrix\n","cm = confusion_matrix(np.argmax(labels, axis=-1), np.argmax(predictions, axis=-1))\n","\n","# Compute the classification metrics\n","accuracy = accuracy_score(np.argmax(labels, axis=-1), np.argmax(predictions, axis=-1))\n","precision = precision_score(np.argmax(labels, axis=-1), np.argmax(predictions, axis=-1), average='macro')\n","recall = recall_score(np.argmax(labels, axis=-1), np.argmax(predictions, axis=-1), average='macro')\n","f1 = f1_score(np.argmax(labels, axis=-1), np.argmax(predictions, axis=-1), average='macro')\n","f1_classes = f1_score(np.argmax(labels, axis=-1), np.argmax(predictions, axis=-1), average=None)\n","\n","print('Accuracy:',accuracy.round(4))\n","print('Precision:',precision.round(4))\n","print('Recall:',recall.round(4))\n","print('F1:',f1.round(4))\n","print('F1 (per class):' ,f1_classes)\n","\n","# Plot the confusion matrix\n","ig, ax = plt.subplots(figsize=(12,12))\n","ax.matshow(cm, cmap=plt.cm.Blues, alpha=0.3)\n","for i in range(cm.shape[0]):\n","    for j in range(cm.shape[1]):\n","        ax.text(x=j, y=i,s=cm[i, j], va='center', ha='center', size='xx-large')\n","\n","        \n","plt.figure(figsize=(10,8))\n","sns.heatmap(cm.T, cmap='Blues')\n","plt.xlabel('True labels')\n","plt.ylabel('Predicted labels')\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"id":"46Qu5_8MyaEG","executionInfo":{"status":"ok","timestamp":1671208969247,"user_tz":-60,"elapsed":2994,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"630e0c24-673c-4559-a589-414e24b066d8"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["16/16 [==============================] - 0s 2ms/step\n","16/16 [==============================] - 0s 2ms/step\n","16/16 [==============================] - 0s 3ms/step\n","16/16 [==============================] - 0s 2ms/step\n","16/16 [==============================] - 0s 2ms/step\n","16/16 [==============================] - 0s 2ms/step\n","16/16 [==============================] - 0s 2ms/step\n","16/16 [==============================] - 0s 2ms/step\n","16/16 [==============================] - 0s 3ms/step\n","16/16 [==============================] - 0s 2ms/step\n","Accuracy: 0.8807\n","Precision: 0.9129\n","Recall: 0.7767\n","F1: 0.8185\n","F1 (per class): [0.92307692 0.92       1.         0.77124183 0.375      0.7037037\n"," 0.984375   0.69565217 0.9787234  0.89880952 1.         0.57142857]\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 864x864 with 1 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}},{"output_type":"display_data","data":{"text/plain":["<Figure size 720x576 with 2 Axes>"],"image/png":"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\n"},"metadata":{"needs_background":"light"}}]},{"cell_type":"markdown","source":["# save"],"metadata":{"id":"e3luJb6Ev2jU"}},{"cell_type":"code","source":["%cd /content/drive/MyDrive/ann_challenge2/modelli_bagging"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"YzOgmwW-1ksQ","executionInfo":{"status":"ok","timestamp":1671209412739,"user_tz":-60,"elapsed":4,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"f91102a2-673d-41d1-9bea-915c2aebf6f8"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["/content/drive/MyDrive/ann_challenge2/modelli_bagging\n"]}]},{"cell_type":"code","source":["model1.save(\"bagging_model1\")\n","model2.save(\"bagging_model2\")\n","model3.save(\"bagging_model3\")\n","model4.save(\"bagging_model4\")\n","model5.save(\"bagging_model5\")\n","model6.save(\"bagging_model6\")\n","model7.save(\"bagging_model7\")\n","model8.save(\"bagging_model8\")\n","model9.save(\"bagging_model9\")\n","model10.save(\"bagging_model10\")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"xofA0OW72DaZ","executionInfo":{"status":"ok","timestamp":1671209514510,"user_tz":-60,"elapsed":13508,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"445e24c8-8db4-45b3-ac12-3770427d9e7f"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stderr","text":["WARNING:absl:Function `_wrapped_model` contains input name(s) Input with unsupported characters which will be renamed to input in the SavedModel.\n","WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 3 of 3). These functions will not be directly callable after loading.\n","WARNING:absl:Function `_wrapped_model` contains input name(s) Input with unsupported characters which will be renamed to input in the SavedModel.\n","WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 3 of 3). These functions will not be directly callable after loading.\n","WARNING:absl:Function `_wrapped_model` contains input name(s) Input with unsupported characters which will be renamed to input in the SavedModel.\n","WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 3 of 3). These functions will not be directly callable after loading.\n","WARNING:absl:Function `_wrapped_model` contains input name(s) Input with unsupported characters which will be renamed to input in the SavedModel.\n","WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 3 of 3). These functions will not be directly callable after loading.\n","WARNING:absl:Function `_wrapped_model` contains input name(s) Input with unsupported characters which will be renamed to input in the SavedModel.\n","WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 3 of 3). These functions will not be directly callable after loading.\n","WARNING:absl:Function `_wrapped_model` contains input name(s) Input with unsupported characters which will be renamed to input in the SavedModel.\n","WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 3 of 3). These functions will not be directly callable after loading.\n","WARNING:absl:Function `_wrapped_model` contains input name(s) Input with unsupported characters which will be renamed to input in the SavedModel.\n","WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 3 of 3). These functions will not be directly callable after loading.\n","WARNING:absl:Function `_wrapped_model` contains input name(s) Input with unsupported characters which will be renamed to input in the SavedModel.\n","WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 3 of 3). These functions will not be directly callable after loading.\n","WARNING:absl:Function `_wrapped_model` contains input name(s) Input with unsupported characters which will be renamed to input in the SavedModel.\n","WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 3 of 3). These functions will not be directly callable after loading.\n","WARNING:absl:Function `_wrapped_model` contains input name(s) Input with unsupported characters which will be renamed to input in the SavedModel.\n","WARNING:absl:Found untraced functions such as _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 3 of 3). These functions will not be directly callable after loading.\n"]}]},{"cell_type":"code","source":["scaler.__dict__"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"TrLGlogx_6cJ","executionInfo":{"status":"ok","timestamp":1671210052374,"user_tz":-60,"elapsed":465,"user":{"displayName":"enrico staiano","userId":"10756654379627178087"}},"outputId":"1f518fc9-4436-452b-9080-db393ce182b3"},"execution_count":null,"outputs":[{"output_type":"execute_result","data":{"text/plain":["{'with_mean': True,\n"," 'with_std': True,\n"," 'copy': True,\n"," 'n_features_in_': 6,\n"," 'n_samples_seen_': 284193,\n"," 'mean_': array([ 9.7380162 , 18.01140827, 26.70841189, 32.42400051, 33.8400954 ,\n","        31.34904435]),\n"," 'var_': array([ 63062.52096298, 603339.45305708, 484803.54050911, 461821.14745811,\n","        442735.39552485, 527209.68247186]),\n"," 'scale_': array([251.12252182, 776.74928584, 696.27834988, 679.57423984,\n","        665.38364537, 726.09206198])}"]},"metadata":{},"execution_count":68}]},{"cell_type":"code","source":[],"metadata":{"id":"2AwLrrTc_6Zw"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":[],"metadata":{"id":"-y2x8lap_6Xk"},"execution_count":null,"outputs":[]}]}