Classification SVM model of ozone Machine Learning

I developed 2 machine learning software that predict and classify ozone day and non-ozone day. The working principle of the two is similar but there are differences. I got the dataset from ics.icu. Each software has a different mathematical model, Gaussian RBF and Linear Kernel, and classifications are visualized in different ways. I would be happy to present the software to you!

README.md

@@ -1,2 +1,42 @@
-# Classification-thanks-to-the-SVM-model-with-7-years-of-ozone-data-with-Machine-Learning
-I developed 2 machine learning software that predict and classify ozone day and non-ozone day. The working principle of the two is similar but there are differences. I got the dataset from ics.icu. Each software has a different mathematical model, Gaussian RBF and Linear Kernel, and classifications are visualized in different ways. I would be happy to present the software to you!
+# **Classification thanks to the SVM model with 7 years of ozone data with Machine Learning**
+I developed 2 machine learning software that predict and classify ozone day and non-ozone day. The working principle of the two is similar but there are differences. I got the dataset from ics.icu. Each software has a different mathematical model, Gaussian RBF and Linear Kernel, and classifications are visualized in different ways. I would be happy to present the software to you!
+
+_Example:_ `model_ozone = PCA(n_components=72).fit(X_train)`
+
+`model = svm.SVC(kernel='rbf', gamma=0.05, C=3)`
+
+`model_ozone = svm.SVC(kernel='linear', C=3)
+
+**I am happy to present this software to you!**
+
+Data Source: [DataSource]
+###**The coding language used:**
+
+`Python 3.9.6`
+
+###**Libraries Used:**
+
+`Sklearn`
+
+`Pandas`
+
+`Numpy`
+
+`Pylab`
+
+`Matplotlib`
+### **Developer Information:**
+
+Name-Surname: **Emirhan BULUT**
+
+Contact (Email) : **emirhan.bulut@turkiyeyapayzeka.com**
+
+LinkedIn : **[https://www.linkedin.com/in/artificialintelligencebulut/][LinkedinAccount]**
+
+[LinkedinAccount]: https://www.linkedin.com/in/artificialintelligencebulut/
+
+Official Website: **[https://www.emirhanbulut.com.tr][OfficialWebSite]**
+
+[OfficialWebSite]: https://www.emirhanbulut.com.tr
+
+[DataSource]: https://archive.ics.uci.edu/ml/index.php

svm-ozone-linear.py

@@ -0,0 +1,79 @@
+from sklearn import svm
+from sklearn.model_selection import train_test_split
+import numpy as np
+from sklearn import metrics
+import pandas as pd
+#from sklearn.preprocessing import LabelEncoder
+
+df = pd.read_csv('ozone-data.csv')
+#print(df.head())
+
+#data select
+X = df.iloc[:,1:73]
+#print(X)
+
+#target select [class]
+y = df.iloc[:,73:74]
+#print(y)
+
+#dataframe to numpy array
+yyy = np.array(y).ravel()
+XXX = np.array(X).reshape(2534,-1)
+#print(target)
+#print(connects)
+
+#target_variables = LabelEncoder()
+
+#y['Date_data'] = target_variables.fit_transform(y['Date'])
+
+#y_n = y.drop(['Date'],axis=1)
+
+#print("Features: ", X)
+
+#print("Target: ", y_n)
+
+X_train, X_test, y_train, y_test = train_test_split(XXX,yyy,test_size=0.2880820836621942,random_state=49) # 80% training and 20% test
+
+#creating of SVM model
+model_ozone = svm.SVC(kernel='linear', C=3) # Linear Kernel
+
+#Train the model using the training sets
+model_ozone.fit(X_train, y_train)
+
+#Predict the response for test dataset
+y_test_pred = model_ozone.predict(X_test)
+#print(X_test)
+#print(y_test)
+#Accuracy
+
+#C = 1.0
+
+#h = .02
+
+#rbf_svc = svm.SVC(kernel='rbf', gamma=0.7, C=C).fit(X_train, y_train)
+#poly_svc = svm.SVC(kernel='poly', degree=3, C=C).fit(X_train, y_train)
+#lin_svc = svm.LinearSVC(C=C).fit(X_train, y_train)
+
+
+#print(y_pred)
+print("Accuracy:",metrics.accuracy_score(y_test,y_test_pred))
+#Accuracy: 0.9561643835616438
+
+import matplotlib.pyplot as plt
+
+#plt.scatter(X_train[:,0], X_train[:,1])
+#plt.title('Linearly separable data')
+#plt.xlabel('X1')
+#plt.ylabel('X2')
+#plt.show()
+
+support_vectors = model_ozone.support_vectors_
+
+# Visualize support vectors
+plt.scatter(X_train[:,0], X_train[:,1])
+plt.scatter(support_vectors[:,0], support_vectors[:,1], color='red')
+plt.title('Ozone Day and Normal Day Prediction Software')
+plt.xlabel('Data')
+plt.ylabel('Class [Ozone Day = 1, Normal Day = 0]')
+plt.show()
+

svm-ozone-rbf.py

@@ -0,0 +1,57 @@
+from sklearn.decomposition import PCA
+from sklearn import svm
+from sklearn.model_selection import train_test_split
+import numpy as np
+from sklearn import metrics
+import pandas as pd
+#from sklearn.preprocessing import LabelEncoder
+
+#read data
+df = pd.read_csv('ozone-data.csv')
+#print(df.head())
+
+#data select
+X = df.iloc[:,1:73]
+print(X)
+#target select [class]
+y = df.iloc[:,73:74]
+print(y)
+#target type of Dataframe to type of Numpy Array
+yyy = np.array(y).ravel()
+
+#train,test of creating
+X_train, X_test, y_train, y_test = train_test_split(X,yyy,test_size=0.2880820836621942,random_state=49) # 80% training and 20% test
+
+
+#creationg pca model
+model_ozone = PCA(n_components=72).fit(X_train)
+model_ozone_2d = model_ozone.transform(X_train)
+
+#print(model_ozone_2d)e
+
+
+#creating of model the SVM
+model = svm.SVC(kernel='rbf', gamma=0.05, C=3)
+
+#The model is training in equation
+model.fit(model_ozone_2d, y_train)
+
+#prediction equation is the model.
+y_test_pred = model.predict(X_test)
+
+#accuracy is model
+print("Accuracy:",metrics.accuracy_score(y_test,y_test_pred))
+#Accuracy: 0.9602739726027397
+
+#data visualition
+import pylab as pl
+for i in range(0, model_ozone_2d.shape[0]):
+    if y_train[i] == 0:
+        c1 = pl.scatter(model_ozone_2d[i,0],model_ozone_2d[i,1],color='r',edgecolors='y',marker='*',linewidths=1)
+
+    elif y_train[i] == 1:
+        c2 = pl.scatter(model_ozone_2d[i,0],model_ozone_2d[i,1],color='g',edgecolors='y',marker='o',linewidths=1)
+import matplotlib.pyplot as plt
+pl.legend([c1, c2], ['Ozone Day', 'Normal Day'])
+plt.title('Ozone and Normal Day Classification')
+pl.show()