dipanjanS

EPOCHS = 10
train_images_scaled = train_images_gr / 255.
model.fit(train_images_scaled, train_labels, validation_split=0.1, epochs=EPOCHS)

# Output
Train on 54000 samples, validate on 6000 samples
Epoch 1/10
54000/54000 [====] - 7s 122us/sample - loss: 0.4614 - acc: 0.8323 - val_loss: 0.3462 - val_acc: 0.8725
Epoch 2/10
54000/54000 [====] - 5s 86us/sample - loss: 0.3073 - acc: 0.8892 - val_loss: 0.2825 - val_acc: 0.8950

dipanjanS

INPUT_SHAPE = (28, 28, 1)

def create_cnn_architecture_model1(input_shape):
    inp = keras.layers.Input(shape=input_shape)

    conv1 = keras.layers.Conv2D(filters=16, kernel_size=(3, 3), strides=(1, 1), 
                                activation='relu', padding='same')(inp)
    pool1 = keras.layers.MaxPooling2D(pool_size=(2, 2))(conv1)
    conv2 = keras.layers.Conv2D(filters=32, kernel_size=(3, 3), strides=(1, 1), 
                                   activation='relu', padding='same')(pool1)

dipanjanS

# reshape for feeding into the model
train_images_gr = train_images.reshape(train_images.shape[0], 28, 28, 1)
test_images_gr = test_images.reshape(test_images.shape[0], 28, 28, 1)

print('\nTrain_images.shape: {}, of {}'.format(train_images_gr.shape, train_images_gr.dtype))
print('Test_images.shape: {}, of {}'.format(test_images_gr.shape, test_images_gr.dtype))

dipanjanS

fashion_mnist = keras.datasets.fashion_mnist
(train_images, train_labels), (test_images, test_labels) = fashion_mnist.load_data()

class_names = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat',
               'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle boot']

print('\nTrain_images.shape: {}, of {}'.format(train_images.shape, train_images.dtype))
print('Test_images.shape: {}, of {}'.format(test_images.shape, test_images.dtype))

dipanjanS

# to prevent unnecessary warnings
import warnings
warnings.simplefilter(action='ignore', category=FutureWarning)

# TensorFlow and tf.keras
import tensorflow as tf
from tensorflow import keras

# Helper libraries
import numpy as np

dipanjanS

fig, ax = plt.subplots(2, 4, figsize=(12, 6))

for idx, img_idx in enumerate([15, 123, 230, 340, 450, 560, 670]):
    id1 = 1 if idx > 3 else 0
    id2 = idx % 4
    predicted_label = class_label_mapping[
                            np.argmax(
                                resnet_ft_model4.predict(
                                    np.array([test_data_X[img_idx]])
                                ),axis=1

dipanjanS

with tf.device('/cpu:0'):
    resnet_ft_model4 = keras.models.load_model('resnet_finetune_full_models/resnet50_finetune_full_seti_epoch_497.h5')
evaluate_model_results(resnet_ft_model4, test_data_X, test_data_y, 
                       class_label_mapping, class_labels)

dipanjanS

vgg19_full_ft_model1 = keras.models.load_model('vgg19_finetune_full_models/vgg19_finetune_full_seti.h5')
evaluate_model_results(vgg19_full_ft_model1, test_data_X, test_data_y, 
                       class_label_mapping, class_labels)

dipanjanS

vgg19_partial_ft_model3 = keras.models.load_model('vgg19_finetune_partial_models/vgg19_finetune_partial_seti_epoch_99.h5')
evaluate_model_results(vgg19_partial_ft_model3, test_data_X, test_data_y, 
                       class_label_mapping, class_labels)

dipanjanS

import pandas as pd
from sklearn.metrics import classification_report, confusion_matrix

def evaluate_model_results(model, test_data, test_labels, 
                           class_label_mapping, class_labels):
    predictions = model.predict(test_data, verbose=1)
    prediction_labels = [class_label_mapping[idx] for idx in predictions.argmax(axis=1)]
    print(classification_report(y_true=test_labels, y_pred=prediction_labels))
    return pd.DataFrame(confusion_matrix(y_true=test_labels, y_pred=prediction_labels, labels=class_labels),
                        index=class_labels, columns=class_labels)