1. Downloading data from https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz
170500096/170498071 [==============================] - 19s 0us/step
X_train original shape (50000, 32, 32, 3)
y_train original shape (50000, 1)
X_test original shape (10000, 32, 32, 3)
y_test original shape (10000, 1)
Text(0.5, 1.0, '[9]')
In [1]: import pandas as pd
import numpy as np
from keras.datasets import cifar10
from keras.utils import to_categorical
import matplotlib.pyplot as plt
In [2]: (x_train, y_train), (x_test, y_test) = cifar10.load_data()
In [5]: print("X_train original shape", x_train.shape)
print("y_train original shape", y_train.shape)
print("X_test original shape", x_test.shape)
print("y_test original shape", y_test.shape)
In [9]: plt.imshow(x_train[2], cmap='gray')
plt.title(y_train[2])
Out[9]:
In [10]: # Preprocess the data (these are NumPy arrays)
x_train = x_train.astype("float32") / 255
x_test = x_test.astype("float32") / 255
y_train = to_categorical(y_train)
y_test = to_categorical(y_test)
# allocating 10,000 samples for validation
x_val = x_train[-10000:]
y_val = y_train[-10000:]
x_train = x_train[:-10000]
y_train = y_train[:-10000]
In [13]: #instantiate the model
from keras import models
from keras import layers
model = models.Sequential()
model.add(layers.Conv2D(32, (3,3), activation='relu', input_shape=(32,32,3)))

2. Model: "sequential_2"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
conv2d_6 (Conv2D) (None, 30, 30, 32) 896
_________________________________________________________________
max_pooling2d_6 (MaxPooling2 (None, 15, 15, 32) 0
_________________________________________________________________
conv2d_7 (Conv2D) (None, 13, 13, 64) 18496
_________________________________________________________________
max_pooling2d_7 (MaxPooling2 (None, 6, 6, 64) 0
_________________________________________________________________
conv2d_8 (Conv2D) (None, 4, 4, 128) 73856
_________________________________________________________________
max_pooling2d_8 (MaxPooling2 (None, 2, 2, 128) 0
_________________________________________________________________
flatten_2 (Flatten) (None, 512) 0
_________________________________________________________________
dense_4 (Dense) (None, 128) 65664
_________________________________________________________________
dense_5 (Dense) (None, 10) 1290
=================================================================
Total params: 160,202
Trainable params: 160,202
Non-trainable params: 0
_________________________________________________________________
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(128, (3,3), activation='relu'))
model.add(layers.MaxPooling2D(2,2))
model.add(layers.Flatten())
model.add(layers.Dense(128, activation='relu'))
model.add(layers.Dense(10, activation='softmax'))
model.summary()
In [14]: model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'
In [17]: history = model.fit(x_train, y_train, epochs=100, batch_size=128,
validation_data=(x_val, y_val),verbose=0)
In [18]: train_loss = history.history['loss']
val_loss = history.history['val_loss']
epochs = range(1, len(history.history['loss']) + 1)
In [19]: plt.plot(epochs, train_loss, 'bo', label='Loss_Training')
plt.plot(epochs, val_loss, 'b', label='Loss_Validation')
plt.title('Training and Validation Losses')
plt.xlabel('Epochs')
plt.ylabel('Loss')
plt.legend()
plt.show()
plt.savefig('Results/6_2a_lossplot.png')

3. <Figure size 432x288 with 0 Axes>
<Figure size 432x288 with 0 Axes>
313/313 [==============================] - 1s 4ms/step - loss: 3.9177 - accuracy: 0.
6841
Test accuracy: 0.6840999722480774
In [20]: train_loss = history.history['accuracy']
val_loss = history.history['val_accuracy']
epochs = range(1, len(history.history['accuracy']) + 1)
In [21]: plt.plot(epochs, train_loss, 'bo', label='Training accuracy')
plt.plot(epochs, val_loss, 'b', label='Validation accuracy')
plt.title('Training and Validation Accuracy')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
plt.legend()
plt.show()
plt.savefig('Results/6_2a_accplot.png')
In [22]: model.save('Results/6_2a_model.h5')
In [23]: score = model.evaluate(x_test, y_test)
print('Test accuracy: ', score[1])
In [26]: predictions = np.argmax(model.predict(x_test), axis=1)

4. Actual Predictions
0 [0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... 3
1 [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, ... 8
2 [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, ... 8
3 [1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... 0
4 [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, ... 6
... ... ...
9995 [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, ... 3
9996 [0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... 5
9997 [0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, ... 5
9998 [0.0, 1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ... 4
9999 [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, ... 7
[10000 rows x 2 columns]
predictions = list(predictions)
actuals = list(y_test)
pred_res = pd.DataFrame({'Actual': actuals, 'Predictions': predictions})
pred_res.to_csv('Results/6_2a_predictions.csv', index=False)
print (pred_res)
In [27]: #Metrics output
with open('Results/6_2a_metrics.txt', 'w') as f:
f.write('Training Loss: {}'.format(str(history.history['loss'])))
f.write('nTraining Accuracy: {}'.format(str(history.history['accuracy'])))
f.write('nTest Loss: {}'.format(score[0]))
f.write('nTest Accuracy: {}'.format(score[1]))