KERAS_WITH_TRAINING_MNIST.py

import keras
from keras.datasets import mnist
from keras.layers import Input, Dense
from keras.models import load_model
from skimage.util import invert

import numpy as np
import matplotlib.pyplot as plt

(X_train, y_train), (X_test, y_test) = mnist.load_data()

X_train = invert(X_train.astype('float32') / 255)
X_test = invert(X_test.astype('float32') / 255)

print('Size of test images 1 : ', X_test.shape)

X_train = X_train.reshape(len(X_train), np.prod(X_train.shape[1:]))
X_test = X_test.reshape(len(X_test), np.prod(X_test.shape[1:]))

print('Size of test images 2 : ', X_test.shape)

# convert class vectors to binary class matrices
# y_train = keras.utils.to_categorical(y_train, 10)
# y_test = keras.utils.to_categorical(y_test, 10)

inputs = Input(shape=(784,))
x = Dense(units=64, activation='relu')(inputs)
x = Dense(units=64, activation='relu')(x)
outputs = Dense(units=10, activation='softmax')(x)

model = keras.Model(inputs=inputs, outputs=outputs)

model.compile(optimizer=keras.optimizers.Adam(),  # Optimizer
              # Loss function to minimize
              loss=keras.losses.sparse_categorical_crossentropy,
              # List of metrics to monitor
              metrics=['accuracy'])
history = model.fit(X_train, y_train, batch_size=64, epochs=1, validation_data=(X_test, y_test))

print('\nhistory dict:', history.history)

# Evaluate the model on the test data using `evaluate`
print('\n# Evaluate on test data')
results = model.evaluate(X_test, y_test, batch_size=64)

print('test loss, test acc:', results)

print('\n Save and Load Model **************---------------****************** ')
model.save('MNIST.h5')

test_model = load_model('MNIST.h5')
test_model.summary()

print('Size of test images 3 : ', X_test.shape)
k = X_test[2533]
#k = invert(k)
test_img = k.reshape(28, 28)
plt.imshow(test_img, cmap='gray')
plt.show()
print(test_img.shape)

k = np.array(k)
print(k.shape)
k = k.reshape(1, 784)
#print(k)
prediction = test_model.predict(k)

print(np.argmax(prediction))

from keras.datasets import mnist
from keras.models import load_model
import numpy as np
import sys
from skimage.io import imread, imsave
from skimage.util import invert
import matplotlib.image as mpimg
import matplotlib.pyplot as plt

np.set_printoptions(threshold=sys.maxsize)


img_width, img_height = 28, 28

(X_train, y_train), (X_test, y_test) = mnist.load_data()
img = X_train[5589]
testing_img = imsave('img.png', invert(img))

im = mpimg.imread('img.png')
plt.imshow(im, cmap='gray')
plt.show()

model = load_model('MNIST.h5')
im = np.array(im).reshape(1, 784)

predict = model.predict(im)

print(predict)
print(np.argmax(predict))