Machine learning

No account yet ?

Machine learning is considered part of artificial intelligence.

Machine learning includes many algorithms:

Linear, multivariate, polynomial, regularized, logistic regressions, ... which are curves that approximate the data.

The Naïve Bayes algorithm which gives the probability of the prediction, knowing the previous events.

Clustering which, thanks to mathematics, will group the data into packets so that in each packet the data are as close as possible to each other.

Decision trees that answer a number of questions then just follow the branches of the tree to arrive at a result with a probability score.

As well as more advanced algorithms such as: Random Forest, Gradient Boosting, ...

Tested in Anaconda and Python 3.7

import pandas as pd
from sklearn.cluster import MeanShift 
# import seaborn as sns
import matplotlib.pyplot as plt 
 
colleges = pd.read_csv('College_data.csv',index_col = 0)
print(colleges.info())
 
x = colleges[["Apps","Grad.Rate"]].values  
 
print(x)
 
ms = MeanShift() 
y_hc = ms.fit_predict(x) 
 
print(y_hc)
 
plt.scatter(x[y_hc == 0, 0], x[y_hc == 0, 1], s = 100, c = 'red', label = 'Cluster 1')
plt.scatter(x[y_hc == 1, 0], x[y_hc == 1, 1], s = 100, c = 'blue', label = 'Cluster 2')
plt.scatter(x[y_hc == 2, 0], x[y_hc == 2, 1], s = 100, c = 'green', label = 'Cluster 3')
 
plt.title('University Data')
plt.xlabel('Applications')
plt.ylabel('Graduate Rates')
plt.legend()
plt.show()
 

Mean-shift Machine_Learning_Clustering_Algorithms - GitHub

Convolutional Neural Network (CNN)

Implementation of CNN in Python

Tested in Anaconda and Python 3.7

#importing the required libraries
from tensorflow.keras.datasets import mnist
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Conv2D
from tensorflow.keras.layers import MaxPool2D
from tensorflow.keras.layers import Flatten
from tensorflow.keras.layers import Dropout
from tensorflow.keras.layers import Dense
 
#loading data
(X_train,y_train) , (X_test,y_test)=mnist.load_data()
#reshaping data
X_train = X_train.reshape((X_train.shape[0], X_train.shape[1], X_train.shape[2], 1))
X_test = X_test.reshape((X_test.shape[0],X_test.shape[1],X_test.shape[2],1)) 
#checking the shape after reshaping
print(X_train.shape)
print(X_test.shape)
#normalizing the pixel values
X_train=X_train/255
X_test=X_test/255
 
#defining model
model=Sequential()
#adding convolution layer
model.add(Conv2D(32,(3,3),activation='relu',input_shape=(28,28,1)))
#adding pooling layer
model.add(MaxPool2D(2,2))
#adding fully connected layer
model.add(Flatten())
model.add(Dense(100,activation='relu'))
#adding output layer
model.add(Dense(10,activation='softmax'))
#compiling the model
model.compile(loss='sparse_categorical_crossentropy',optimizer='adam',metrics=['accuracy'])
#fitting the model
model.fit(X_train,y_train,epochs=10)
 
#evaluting the model
model.evaluate(X_test,y_test)
 

Source : https://www.analyticsvidhya.com/blog/2021/08/beginners-guide-to-convolutional-neural-network-with-implementation-in-python/

Result

Epoch 1/10
60000/60000 [==============================] - 9s 151us/sample - loss: 0.1630 - acc: 0.9508
Epoch 2/10
60000/60000 [==============================] - 7s 120us/sample - loss: 0.0533 - acc: 0.9837
Epoch 3/10
60000/60000 [==============================] - 7s 123us/sample - loss: 0.0356 - acc: 0.9891
Epoch 4/10
60000/60000 [==============================] - 7s 122us/sample - loss: 0.0242 - acc: 0.9926
Epoch 5/10
60000/60000 [==============================] - 8s 127us/sample - loss: 0.0172 - acc: 0.9945
Epoch 6/10
60000/60000 [==============================] - 8s 133us/sample - loss: 0.0117 - acc: 0.9961
Epoch 7/10
60000/60000 [==============================] - 8s 125us/sample - loss: 0.0100 - acc: 0.9969
Epoch 8/10
60000/60000 [==============================] - 7s 118us/sample - loss: 0.0068 - acc: 0.9977
Epoch 9/10
60000/60000 [==============================] - 8s 125us/sample - loss: 0.0054 - acc: 0.9983
Epoch 10/10
60000/60000 [==============================] - 7s 119us/sample - loss: 0.0051 - acc: 0.9983
10000/10000 [==============================] - 1s 65us/sample - loss: 0.0551 - acc: 0.9875

Artificial Neural Network (ANN)

Implementation of ANN in Python

Tested in Anaconda and Python 3.7

#importing libraries
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
# Ignore the warnings
import warnings
warnings.filterwarnings("ignore")
 
#loading MNIST dataset
from tensorflow.keras.datasets import mnist
(X_train,y_train) , (X_test,y_test)=mnist.load_data()
 
#visualizing the image in train data
plt.imshow(X_train[0])
 
#visualizing the first 20 images in the dataset
 
for i in range(25):
 
    #subplot
 
    plt.subplot(5, 5, i+1)
 
    # plotting pixel data
 
    plt.imshow(X_train[i], cmap=plt.get_cmap('gray'))
 
# show the figure
 
plt.show()
 
print(X_train.shape)
print(X_test.shape)
 
# the image is in pixels which ranges from 0 to 255
X_train[0]
 
X_train_flat=X_train.reshape(len(X_train),28*28)
 
X_test_flat=X_test.reshape(len(X_test),28*28)
 
#checking the shape after flattening
 
print(X_train_flat.shape)
 
print(X_test_flat.shape)
 
#checking the representation of image after flattening
X_train_flat[0]
 
#normalizing the pixel values
X_train_flat=X_train_flat/255
X_test_flat=X_test_flat/255
 
#print this code to check the pixel values after normalization
X_train_flat[0]
 
#Building a simple ANN model without hidden layer
 
#importing necessary libraries
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense
 
#Step 1 : Defining the model
model=Sequential()
model.add(Dense(10,input_shape=(784,),activation='softmax'))
 
#Step 2: Compiling the model
model.compile(loss='sparse_categorical_crossentropy',optimizer='adam',metrics=['accuracy'])
 
#Step 3: Fitting the model
model.fit(X_train_flat,y_train,epochs=10)
 
#Step 4: Evaluating the model
model.evaluate(X_test_flat,y_test)
 
#Step 5 :Making predictions
y_predict = model.predict(X_test_flat)
y_predict[3] #printing the 3rd index
 
# Here we get the index of the maximum value in the above-encoded vector. 
np.argmax(y_predict[3])
 
#checking if the predicting is correct
plt.imshow(X_test[3])
 
y_predict_labels=np.argmax(y_predict,axis=1)
#Confusion matrix
from sklearn.metrics import confusion_matrix
matrix=confusion_matrix(y_test,y_predict_labels)
#visualizaing confusion matrix with heatmap
plt.figure(figsize=(10,7))
sns.heatmap(matrix,annot=True,fmt='d')
 
 
model2=Sequential()
#adding first layer with 100 neurons
model2.add(Dense(100,input_shape=(784,),activation='relu'))
#second layer with 64 neurons
model2.add(Dense(64,activation='relu'))
#third layer with 32 neurons
model2.add(Dense(32,activation='relu'))
#output layer
model2.add(Dense(10,activation='softmax'))
#compliling the model
model2.compile(loss='sparse_categorical_crossentropy',optimizer='adam',metrics=['accuracy'])
#fitting the model
model2.fit(X_train_flat,y_train,epochs=10)
 
#evaluating the model
model2.evaluate(X_test_flat,y_test)
 

Source : https://www.analyticsvidhya.com/blog/2021/08/implementing-artificial-neural-network-on-unstructured-data/

Result

Epoch 1/10
60000/60000 [==============================] - 4s 70us/sample - loss: 0.4720 - acc: 0.8763
Epoch 2/10
60000/60000 [==============================] - 4s 66us/sample - loss: 0.3036 - acc: 0.9160
Epoch 3/10
60000/60000 [==============================] - 4s 70us/sample - loss: 0.2829 - acc: 0.9215
Epoch 4/10
60000/60000 [==============================] - 4s 67us/sample - loss: 0.2733 - acc: 0.9234
Epoch 5/10
60000/60000 [==============================] - 4s 67us/sample - loss: 0.2669 - acc: 0.9256
Epoch 6/10
60000/60000 [==============================] - 4s 67us/sample - loss: 0.2623 - acc: 0.9269
Epoch 7/10
60000/60000 [==============================] - 4s 70us/sample - loss: 0.2585 - acc: 0.9285
Epoch 8/10
60000/60000 [==============================] - 4s 70us/sample - loss: 0.2554 - acc: 0.9295 3s - loss: 0.2484 - acc: 0.9305
Epoch 9/10
60000/60000 [==============================] - 4s 69us/sample - loss: 0.2528 - acc: 0.9304
Epoch 10/10
60000/60000 [==============================] - 4s 71us/sample - loss: 0.2509 - acc: 0.9304
10000/10000 [==============================] - 0s 49us/sample - loss: 0.2653 - acc: 0.9268
Epoch 1/10
60000/60000 [==============================] - 6s 102us/sample - loss: 0.2697 - acc: 0.9196
Epoch 2/10
60000/60000 [==============================] - 6s 94us/sample - loss: 0.1174 - acc: 0.9646
Epoch 3/10
60000/60000 [==============================] - 6s 100us/sample - loss: 0.0847 - acc: 0.9736
Epoch 4/10
60000/60000 [==============================] - 6s 98us/sample - loss: 0.0692 - acc: 0.9783
Epoch 5/10
60000/60000 [==============================] - 6s 99us/sample - loss: 0.0542 - acc: 0.9829: 5s - loss: 0.0581 - acc: 0.9826
Epoch 6/10
60000/60000 [==============================] - 6s 106us/sample - loss: 0.0457 - acc: 0.9854
Epoch 7/10
60000/60000 [==============================] - 6s 100us/sample - loss: 0.0396 - acc: 0.9865
Epoch 8/10
60000/60000 [==============================] - 6s 97us/sample - loss: 0.0327 - acc: 0.9896
Epoch 9/10
60000/60000 [==============================] - 6s 99us/sample - loss: 0.0302 - acc: 0.9898
Epoch 10/10
60000/60000 [==============================] - 6s 96us/sample - loss: 0.0260 - acc: 0.991860000 [==========>...................] - ETA: 3s - loss: 0.0193 - acc: 0.9935
10000/10000 [==============================] - 1s 55us/sample - loss: 0.0920 - acc: 0.9768

Developing an Image Classification Model Using CNN

Tested in Anaconda and Python 3.7

# importing necessary libraries
import numpy as np
import matplotlib.pyplot as plt
 
# To convert to categorical data
from tensorflow.keras.utils import to_categorical
#libraries for building model
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Conv2D, MaxPool2D, Dropout,Flatten
from tensorflow.keras.datasets import cifar10
 
#loading the data
(X_train, y_train), (X_test, y_test) = cifar10.load_data()
 
#shape of the dataset
print(X_train.shape)
print(y_train.shape)
print(X_test.shape)
print(y_test.shape)
 
#checking the labels 
np.unique(y_train)
 
#first image of training data
plt.subplot(121)
plt.imshow(X_train[0])
plt.show()
plt.title("Label : {}".format(y_train[0]))
#first image of test data
plt.subplot(122)
plt.imshow(X_test[0])
plt.show()
plt.title("Label : {}".format(y_test[0]));
 
#visualizing the first 20 images in the dataset
for i in range(20):
    #subplot
    plt.subplot(5, 5, i+1)
    # plotting pixel data
    plt.imshow(X_train[i], cmap=plt.get_cmap('gray'))
# show the figure
plt.show()
 
# Scale the data to lie between 0 to 1
X_train = X_train/255
X_test = X_test/255
print(X_train)
 
#reshaping the train and test lables to 1D
y_train = y_train.reshape(-1,)
y_test = y_test.reshape(-1,)
 
model=Sequential()
#adding the first Convolution layer
model.add(Conv2D(32,(3,3),activation='relu',input_shape=(32,32,3)))
#adding Max pooling layer
model.add(MaxPool2D(2,2))
#adding another Convolution layer
model.add(Conv2D(64,(3,3),activation='relu'))
model.add(MaxPool2D(2,2))
model.add(Flatten())
#adding dense layer
model.add(Dense(216,activation='relu'))
#adding output layer
model.add(Dense(10,activation='softmax'))
 
model.compile(optimizer='rmsprop',loss='sparse_categorical_crossentropy',metrics=['accuracy'])
 
model.fit(X_train,y_train,epochs=10)
 
model.evaluate(X_test,y_test)
 
pred=model.predict(X_test)
#printing the first element from predicted data
print(pred[0])
#printing the index of 
print('Index:',np.argmax(pred[0]))
 
y_classes = [np.argmax(element) for element in pred]
print('Predicted_values:',y_classes[:10])
print('Actual_values:',y_test[:10])
 
model4=Sequential()
#adding the first Convolution layer
model4.add(Conv2D(32,(3,3),activation='relu',input_shape=(32,32,3)))
#adding Max pooling layer
model4.add(MaxPool2D(2,2))
#adding dropout
model4.add(Dropout(0.2))
#adding another Convolution layer
model4.add(Conv2D(64,(3,3),activation='relu'))
model4.add(MaxPool2D(2,2))
#adding dropout
model4.add(Dropout(0.2))
model4.add(Flatten())
#adding dense layer
model4.add(Dense(216,activation='relu'))
#adding dropout
model4.add(Dropout(0.2))
#adding output layer
model4.add(Dense(10,activation='softmax'))
model4.compile(optimizer='adam',loss='sparse_categorical_crossentropy',metrics=['accuracy'])
model4.fit(X_train,y_train,epochs=10)
 
model4.evaluate(X_test,y_test)
 

Source : https://www.analyticsvidhya.com/blog/2021/08/developing-an-image-classification-model-using-cnn/

Result

(50000, 32, 32, 3)
(50000, 1)
(10000, 32, 32, 3)
(10000, 1)
[[[[0.23137255 0.24313725 0.24705882]
[0.16862745 0.18039216 0.17647059]
[0.19607843 0.18823529 0.16862745]
...
[0.61960784 0.51764706 0.42352941]
[0.59607843 0.49019608 0.4 ]
[0.58039216 0.48627451 0.40392157]]

[[0.0627451 0.07843137 0.07843137]
[0. 0. 0. ]
[0.07058824 0.03137255 0. ]
...
[0.48235294 0.34509804 0.21568627]
[0.46666667 0.3254902 0.19607843]
[0.47843137 0.34117647 0.22352941]]

[[0.09803922 0.09411765 0.08235294]
[0.0627451 0.02745098 0. ]
[0.19215686 0.10588235 0.03137255]
...
[0.4627451 0.32941176 0.19607843]
[0.47058824 0.32941176 0.19607843]
[0.42745098 0.28627451 0.16470588]]

...

[[0.81568627 0.66666667 0.37647059]
[0.78823529 0.6 0.13333333]
[0.77647059 0.63137255 0.10196078]
...
[0.62745098 0.52156863 0.2745098 ]
[0.21960784 0.12156863 0.02745098]
[0.20784314 0.13333333 0.07843137]]

[[0.70588235 0.54509804 0.37647059]
[0.67843137 0.48235294 0.16470588]
[0.72941176 0.56470588 0.11764706]
...
[0.72156863 0.58039216 0.36862745]
[0.38039216 0.24313725 0.13333333]
[0.3254902 0.20784314 0.13333333]]

[[0.69411765 0.56470588 0.45490196]
[0.65882353 0.50588235 0.36862745]
[0.70196078 0.55686275 0.34117647]
...
[0.84705882 0.72156863 0.54901961]
[0.59215686 0.4627451 0.32941176]
[0.48235294 0.36078431 0.28235294]]]


[[[0.60392157 0.69411765 0.73333333]
[0.49411765 0.5372549 0.53333333]
[0.41176471 0.40784314 0.37254902]
...
[0.35686275 0.37254902 0.27843137]
[0.34117647 0.35294118 0.27843137]
[0.30980392 0.31764706 0.2745098 ]]

[[0.54901961 0.62745098 0.6627451 ]
[0.56862745 0.6 0.60392157]
[0.49019608 0.49019608 0.4627451 ]
...
[0.37647059 0.38823529 0.30588235]
[0.30196078 0.31372549 0.24313725]
[0.27843137 0.28627451 0.23921569]]

[[0.54901961 0.60784314 0.64313725]
[0.54509804 0.57254902 0.58431373]
[0.45098039 0.45098039 0.43921569]
...
[0.30980392 0.32156863 0.25098039]
[0.26666667 0.2745098 0.21568627]
[0.2627451 0.27058824 0.21568627]]

...

[[0.68627451 0.65490196 0.65098039]
[0.61176471 0.60392157 0.62745098]
[0.60392157 0.62745098 0.66666667]
...
[0.16470588 0.13333333 0.14117647]
[0.23921569 0.20784314 0.22352941]
[0.36470588 0.3254902 0.35686275]]

[[0.64705882 0.60392157 0.50196078]
[0.61176471 0.59607843 0.50980392]
[0.62352941 0.63137255 0.55686275]
...
[0.40392157 0.36470588 0.37647059]
[0.48235294 0.44705882 0.47058824]
[0.51372549 0.4745098 0.51372549]]

[[0.63921569 0.58039216 0.47058824]
[0.61960784 0.58039216 0.47843137]
[0.63921569 0.61176471 0.52156863]
...
[0.56078431 0.52156863 0.54509804]
[0.56078431 0.5254902 0.55686275]
[0.56078431 0.52156863 0.56470588]]]


[[[1. 1. 1. ]
[0.99215686 0.99215686 0.99215686]
[0.99215686 0.99215686 0.99215686]
...
[0.99215686 0.99215686 0.99215686]
[0.99215686 0.99215686 0.99215686]
[0.99215686 0.99215686 0.99215686]]

[[1. 1. 1. ]
[1. 1. 1. ]
[1. 1. 1. ]
...
[1. 1. 1. ]
[1. 1. 1. ]
[1. 1. 1. ]]

[[1. 1. 1. ]
[0.99607843 0.99607843 0.99607843]
[0.99607843 0.99607843 0.99607843]
...
[0.99607843 0.99607843 0.99607843]
[0.99607843 0.99607843 0.99607843]
[0.99607843 0.99607843 0.99607843]]

...

[[0.44313725 0.47058824 0.43921569]
[0.43529412 0.4627451 0.43529412]
[0.41176471 0.43921569 0.41568627]
...
[0.28235294 0.31764706 0.31372549]
[0.28235294 0.31372549 0.30980392]
[0.28235294 0.31372549 0.30980392]]

[[0.43529412 0.4627451 0.43137255]
[0.40784314 0.43529412 0.40784314]
[0.38823529 0.41568627 0.38431373]
...
[0.26666667 0.29411765 0.28627451]
[0.2745098 0.29803922 0.29411765]
[0.30588235 0.32941176 0.32156863]]

[[0.41568627 0.44313725 0.41176471]
[0.38823529 0.41568627 0.38431373]
[0.37254902 0.4 0.36862745]
...
[0.30588235 0.33333333 0.3254902 ]
[0.30980392 0.33333333 0.3254902 ]
[0.31372549 0.3372549 0.32941176]]]


...


[[[0.1372549 0.69803922 0.92156863]
[0.15686275 0.69019608 0.9372549 ]
[0.16470588 0.69019608 0.94509804]
...
[0.38823529 0.69411765 0.85882353]
[0.30980392 0.57647059 0.77254902]
[0.34901961 0.58039216 0.74117647]]

[[0.22352941 0.71372549 0.91764706]
[0.17254902 0.72156863 0.98039216]
[0.19607843 0.71764706 0.94117647]
...
[0.61176471 0.71372549 0.78431373]
[0.55294118 0.69411765 0.80784314]
[0.45490196 0.58431373 0.68627451]]

[[0.38431373 0.77254902 0.92941176]
[0.25098039 0.74117647 0.98823529]
[0.27058824 0.75294118 0.96078431]
...
[0.7372549 0.76470588 0.80784314]
[0.46666667 0.52941176 0.57647059]
[0.23921569 0.30980392 0.35294118]]

...

[[0.28627451 0.30980392 0.30196078]
[0.20784314 0.24705882 0.26666667]
[0.21176471 0.26666667 0.31372549]
...
[0.06666667 0.15686275 0.25098039]
[0.08235294 0.14117647 0.2 ]
[0.12941176 0.18823529 0.19215686]]

[[0.23921569 0.26666667 0.29411765]
[0.21568627 0.2745098 0.3372549 ]
[0.22352941 0.30980392 0.40392157]
...
[0.09411765 0.18823529 0.28235294]
[0.06666667 0.1372549 0.20784314]
[0.02745098 0.09019608 0.1254902 ]]

[[0.17254902 0.21960784 0.28627451]
[0.18039216 0.25882353 0.34509804]
[0.19215686 0.30196078 0.41176471]
...
[0.10588235 0.20392157 0.30196078]
[0.08235294 0.16862745 0.25882353]
[0.04705882 0.12156863 0.19607843]]]


[[[0.74117647 0.82745098 0.94117647]
[0.72941176 0.81568627 0.9254902 ]
[0.7254902 0.81176471 0.92156863]
...
[0.68627451 0.76470588 0.87843137]
[0.6745098 0.76078431 0.87058824]
[0.6627451 0.76078431 0.8627451 ]]

[[0.76078431 0.82352941 0.9372549 ]
[0.74901961 0.81176471 0.9254902 ]
[0.74509804 0.80784314 0.92156863]
...
[0.67843137 0.75294118 0.8627451 ]
[0.67058824 0.74901961 0.85490196]
[0.65490196 0.74509804 0.84705882]]

[[0.81568627 0.85882353 0.95686275]
[0.80392157 0.84705882 0.94117647]
[0.8 0.84313725 0.9372549 ]
...
[0.68627451 0.74901961 0.85098039]
[0.6745098 0.74509804 0.84705882]
[0.6627451 0.74901961 0.84313725]]

...

[[0.81176471 0.78039216 0.70980392]
[0.79607843 0.76470588 0.68627451]
[0.79607843 0.76862745 0.67843137]
...
[0.52941176 0.51764706 0.49803922]
[0.63529412 0.61960784 0.58823529]
[0.65882353 0.63921569 0.59215686]]

[[0.77647059 0.74509804 0.66666667]
[0.74117647 0.70980392 0.62352941]
[0.70588235 0.6745098 0.57647059]
...
[0.69803922 0.67058824 0.62745098]
[0.68627451 0.6627451 0.61176471]
[0.68627451 0.6627451 0.60392157]]

[[0.77647059 0.74117647 0.67843137]
[0.74117647 0.70980392 0.63529412]
[0.69803922 0.66666667 0.58431373]
...
[0.76470588 0.72156863 0.6627451 ]
[0.76862745 0.74117647 0.67058824]
[0.76470588 0.74509804 0.67058824]]]


[[[0.89803922 0.89803922 0.9372549 ]
[0.9254902 0.92941176 0.96862745]
[0.91764706 0.9254902 0.96862745]
...
[0.85098039 0.85882353 0.91372549]
[0.86666667 0.8745098 0.91764706]
[0.87058824 0.8745098 0.91372549]]

[[0.87058824 0.86666667 0.89803922]
[0.9372549 0.9372549 0.97647059]
[0.91372549 0.91764706 0.96470588]
...
[0.8745098 0.8745098 0.9254902 ]
[0.89019608 0.89411765 0.93333333]
[0.82352941 0.82745098 0.8627451 ]]

[[0.83529412 0.80784314 0.82745098]
[0.91764706 0.90980392 0.9372549 ]
[0.90588235 0.91372549 0.95686275]
...
[0.8627451 0.8627451 0.90980392]
[0.8627451 0.85882353 0.90980392]
[0.79215686 0.79607843 0.84313725]]

...

[[0.58823529 0.56078431 0.52941176]
[0.54901961 0.52941176 0.49803922]
[0.51764706 0.49803922 0.47058824]
...
[0.87843137 0.87058824 0.85490196]
[0.90196078 0.89411765 0.88235294]
[0.94509804 0.94509804 0.93333333]]

[[0.5372549 0.51764706 0.49411765]
[0.50980392 0.49803922 0.47058824]
[0.49019608 0.4745098 0.45098039]
...
[0.70980392 0.70588235 0.69803922]
[0.79215686 0.78823529 0.77647059]
[0.83137255 0.82745098 0.81176471]]

[[0.47843137 0.46666667 0.44705882]
[0.4627451 0.45490196 0.43137255]
[0.47058824 0.45490196 0.43529412]
...
[0.70196078 0.69411765 0.67843137]
[0.64313725 0.64313725 0.63529412]
[0.63921569 0.63921569 0.63137255]]]]
Epoch 1/10
50000/50000 [==============================] - 9s 177us/sample - loss: 1.4020 - acc: 0.4998
Epoch 2/10
50000/50000 [==============================] - 9s 174us/sample - loss: 1.0189 - acc: 0.6459
Epoch 3/10
50000/50000 [==============================] - 9s 179us/sample - loss: 0.8563 - acc: 0.7043
Epoch 4/10
50000/50000 [==============================] - 9s 178us/sample - loss: 0.7378 - acc: 0.7441
Epoch 5/10
50000/50000 [==============================] - 9s 178us/sample - loss: 0.6414 - acc: 0.7827
Epoch 6/10
50000/50000 [==============================] - 9s 180us/sample - loss: 0.5610 - acc: 0.8091
Epoch 7/10
50000/50000 [==============================] - 9s 182us/sample - loss: 0.4938 - acc: 0.8328
Epoch 8/10
50000/50000 [==============================] - 9s 180us/sample - loss: 0.4329 - acc: 0.8536
Epoch 9/10
50000/50000 [==============================] - 9s 180us/sample - loss: 0.3846 - acc: 0.8702
Epoch 10/10
50000/50000 [==============================] - 9s 177us/sample - loss: 0.3494 - acc: 0.8830 7s - loss: 0.3079 - acc: 0.8981
10000/10000 [==============================] - 1s 98us/sample - loss: 1.3305 - acc: 0.6932
[8.3880089e-08 1.5042266e-05 1.8243658e-04 9.4401753e-01 3.9289816e-06
5.5519883e-02 2.3185095e-05 1.3974671e-04 7.1556511e-05 2.6683236e-05]
Index: 3
Predicted_values: [3, 8, 8, 0, 4, 6, 1, 6, 3, 1]
Actual_values: [3 8 8 0 6 6 1 6 3 1]
Epoch 1/10
50000/50000 [==============================] - 9s 187us/sample - loss: 1.5135 - acc: 0.4515
Epoch 2/10
50000/50000 [==============================] - 9s 185us/sample - loss: 1.1761 - acc: 0.5828
Epoch 3/10
50000/50000 [==============================] - 9s 190us/sample - loss: 1.0472 - acc: 0.6301
Epoch 4/10
50000/50000 [==============================] - 9s 187us/sample - loss: 0.9552 - acc: 0.6656
Epoch 5/10
50000/50000 [==============================] - 9s 185us/sample - loss: 0.8951 - acc: 0.6849
Epoch 6/10
50000/50000 [==============================] - 9s 182us/sample - loss: 0.8385 - acc: 0.7042
Epoch 7/10
50000/50000 [==============================] - 9s 179us/sample - loss: 0.7897 - acc: 0.7204
Epoch 8/10
50000/50000 [==============================] - 9s 183us/sample - loss: 0.7491 - acc: 0.7361
Epoch 9/10
50000/50000 [==============================] - 9s 180us/sample - loss: 0.7141 - acc: 0.7452
Epoch 10/10
50000/50000 [==============================] - 9s 181us/sample - loss: 0.6836 - acc: 0.7577
10000/10000 [==============================] - 1s 101us/sample - loss: 0.7977 - acc: 0.7260

Tested in Anaconda and Python 3.7

# -*- coding: utf-8 -*-
"""
Created on Thu Jun 16 20:59:14 2022
 
@author: Tensorflow : https://www.tensorflow.org/tutorials/images/cnn
"""
 
#Importer TensorFlow
import tensorflow as tf
 
from tensorflow.keras import datasets, layers, models
import matplotlib.pyplot as plt
 
#Télécharger et préparer le jeu de données CIFAR10
(train_images, train_labels), (test_images, test_labels) = datasets.cifar10.load_data()
 
# Normalize pixel values to be between 0 and 1
train_images, test_images = train_images / 255.0, test_images / 255.0
 
#Vérifier les données
class_names = ['airplane', 'automobile', 'bird', 'cat', 'deer',
               'dog', 'frog', 'horse', 'ship', 'truck']
 
plt.figure(figsize=(10,10))
for i in range(25):
    plt.subplot(5,5,i+1)
    plt.xticks([])
    plt.yticks([])
    plt.grid(False)
    plt.imshow(train_images[i])
    # The CIFAR labels happen to be arrays, 
    # which is why you need the extra index
    plt.xlabel(class_names[train_labels[i][0]])
plt.show()
 
#Créer la base convolutive
model = models.Sequential()
model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(32, 32, 3)))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(64, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(64, (3, 3), activation='relu'))
 
#Affichons l'architecture de votre modèle jusqu'à présent :
model.summary()
 
#Ajouter des couches denses sur le dessus
model.add(layers.Flatten())
model.add(layers.Dense(64, activation='relu'))
model.add(layers.Dense(10))
 
#Voici l'architecture complète de votre modèle :
model.summary()
 
#Compiler et entraîner le modèle
model.compile(optimizer='adam',
              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
              metrics=['acc'])
 
history = model.fit(train_images, train_labels, epochs=10, 
                    validation_data=(test_images, test_labels))
 
#Évaluer le modèle
plt.plot(history.history['acc'], label='Accuracy')
plt.plot(history.history['val_acc'], label = 'val_accuracy')
plt.xlabel('Epoch')
plt.ylabel('Accuracy')
plt.ylim([0.5, 1])
plt.legend(loc='lower right')
 
test_loss, test_acc = model.evaluate(test_images,  test_labels, verbose=2)
 
print(test_acc)
 

Result

Model: "sequential"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
conv2d (Conv2D) (None, 30, 30, 32) 896
_________________________________________________________________
max_pooling2d (MaxPooling2D) (None, 15, 15, 32) 0
_________________________________________________________________
conv2d_1 (Conv2D) (None, 13, 13, 64) 18496
_________________________________________________________________
max_pooling2d_1 (MaxPooling2 (None, 6, 6, 64) 0
_________________________________________________________________
conv2d_2 (Conv2D) (None, 4, 4, 64) 36928
=================================================================
Total params: 56,320
Trainable params: 56,320
Non-trainable params: 0
_________________________________________________________________
Model: "sequential"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
conv2d (Conv2D) (None, 30, 30, 32) 896
_________________________________________________________________
max_pooling2d (MaxPooling2D) (None, 15, 15, 32) 0
_________________________________________________________________
conv2d_1 (Conv2D) (None, 13, 13, 64) 18496
_________________________________________________________________
max_pooling2d_1 (MaxPooling2 (None, 6, 6, 64) 0
_________________________________________________________________
conv2d_2 (Conv2D) (None, 4, 4, 64) 36928
_________________________________________________________________
flatten (Flatten) (None, 1024) 0
_________________________________________________________________
dense (Dense) (None, 64) 65600
_________________________________________________________________
dense_1 (Dense) (None, 10) 650
=================================================================
Total params: 122,570
Trainable params: 122,570
Non-trainable params: 0
_________________________________________________________________
Train on 50000 samples, validate on 10000 samples

50000/50000 [==============================] - 12s 238us/sample - loss: 1.5028 - acc: 0.4556 - val_loss: 1.4155 - val_acc: 0.5131
Epoch 2/10
50000/50000 [==============================] - 10s 196us/sample - loss: 1.1316 - acc: 0.6003 - val_loss: 1.0506 - val_acc: 0.6277
Epoch 3/10
50000/50000 [==============================] - 10s 196us/sample - loss: 0.9700 - acc: 0.6582 - val_loss: 0.9494 - val_acc: 0.6670
Epoch 4/10
50000/50000 [==============================] - 10s 201us/sample - loss: 0.8795 - acc: 0.6915 - val_loss: 0.9110 - val_acc: 0.6802
Epoch 5/10
50000/50000 [==============================] - 10s 196us/sample - loss: 0.8057 - acc: 0.7171 - val_loss: 0.8866 - val_acc: 0.6960
Epoch 6/10
50000/50000 [==============================] - 10s 193us/sample - loss: 0.7477 - acc: 0.7395 - val_loss: 0.8815 - val_acc: 0.6996
Epoch 7/10
50000/50000 [==============================] - 10s 197us/sample - loss: 0.7029 - acc: 0.7526 - val_loss: 0.8872 - val_acc: 0.6995
Epoch 8/10
50000/50000 [==============================] - 10s 207us/sample - loss: 0.6562 - acc: 0.7691 - val_loss: 0.8546 - val_acc: 0.7127
Epoch 9/10
50000/50000 [==============================] - 10s 192us/sample - loss: 0.6168 - acc: 0.7830 - val_loss: 0.8714 - val_acc: 0.7140
Epoch 10/10
50000/50000 [==============================] - 10s 198us/sample - loss: 0.5775 - acc: 0.7952 - val_loss: 0.8378 - val_acc: 0.7233
10000/10000 - 1s - loss: 0.8378 - acc: 0.7233
0.7233