cifar10_lr_decay.py

# auto save figures in disk in the background 
import matplotlib
matplotlib.use("Agg")

# import packages
from sklearn.preprocessing import LabelBinarizer
from sklearn.metrics import classification_report
from nn.conv.minivggnet import MiniVGGNet
from keras.callbacks import LearningRateScheduler # custom learning rate
from keras.optimizers import SGD
from keras.datasets import cifar10
import matplotlib.pyplot as plt
import numpy as np
import argparse

# defining our custom rate scheduler
def step_decay(epoch):
    # initialize base learning rate, drop factor, and
    # epochs that imply
    # we'll drop our learning rate by factor of 0.25 every 5 epoch
    initAlpha = 0.01
    factor = 0.25
    dropEvery = 5

    # compute learning rate for the current epoch
    alpha = initAlpha * (factor ** np.floor((1 + epoch) / dropEvery))

    # return learning rate
    return float(alpha)


# parse arguments
ap = argparse.ArgumentParser()
ap.add_argument("-o", "--output", required=True, help="path to output plot")
args = vars(ap.parse_args())

# load data and scale it to [0, 1]
print("[INFO] loading CIFAR-10 data...")
((trainX, trainY), (testX, testY)) = cifar10.load_data()
trainX = trainX.astype("float") / 255.0
testX = testX.astype("float") / 255.0

# convert label from integer to vector
lb = LabelBinarizer()
trainY = lb.fit_transform(trainY)
testY = lb.transform(testY)

# initializes the label names for CIFAR-10 dataset
labelNames = ["airplane", "automobile", "bird", "cat", "deer", "dog", "frog", "horse", "ship", "truck"]

# initialize optimizer and the model
callbacks = [LearningRateScheduler(step_decay)]
opt = SGD(lr=0.01, momentum=0.9, nesterov=True)
model = MiniVGGNet.build(width=32, height=32, depth=3, classes=10)
model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"])

# train the network
print("[INFO] training network...")
H = model.fit(trainX, trainY, validation_data=(testX, testY), batch_size=64, epochs=40, verbose=1)

# evaluate the network
print("[INFO] evaluating network...")
predictions = model.predict(testX, batch_size=64)
print(classification_report(testY.argmax(axis=1), predictions.argmax(axis=1), target_names=labelNames))

# plotting the training loss and accuracy
plt.style.use("ggplot")
plt.figure()
plt.plot(np.arange(0, 40), H.history["loss"], label="train_loss")
plt.plot(np.arange(0, 40), H.history["val_loss"], label="val_loss")
plt.plot(np.arange(0, 40), H.history["acc"], label="train_acc")
plt.plot(np.arange(0, 40), H.history["val_acc"], label="val_acc")
plt.title("Training Loss and Accuracy on CIFAR-10")
plt.xlabel("Epoch #")
plt.ylabel("Loss/Accuracy")
plt.legend()
plt.show()
plt.savefig(args["output"])