from __future__ import absolute_import, division, print_function
from tensorflow.keras.datasets import mnist
from tensorflow.keras import Model, layers
import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt

num_classes = 10
num_features = 784
learning_rate = 0.5
training_steps = 4000
batch_size = 256
n_hidden_1 = 256
n_hidden_2 = 512

(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train, x_test = np.array(x_train, np.float32), np.array(x_test, np.float32)
x_train, x_test = x_train.reshape([-1, num_features]), x_test.reshape([-1, num_features])
x_train, x_test = x_train / 255., x_test / 255.
train_data = tf.data.Dataset.from_tensor_slices((x_train, y_train))
train_data = train_data.repeat().shuffle(5000).batch(batch_size).prefetch(1)

class NeuralNet(Model):
    def __init__(self):
        super(NeuralNet, self).__init__()
        self.fc1 = layers.Dense(n_hidden_1, activation=tf.nn.relu)
        self.fc2 = layers.Dense(n_hidden_2, activation=tf.nn.relu)
        self.out = layers.Dense(num_classes, activation=tf.nn.softmax)

    def call(self, x, is_training=False):
        x = self.fc1(x)
        x = self.out(x)
        if not is_training:
            x = tf.nn.softmax(x)
        return x

neural_net = NeuralNet()
def cross_entropy_loss(x, y):
    y = tf.cast(y, tf.int64)
    loss = tf.nn.sparse_softmax_cross_entropy_with_logits(labels=y, logits=x)
    return tf.reduce_mean(loss)

def accuracy(y_pred, y_true):
    correct_prediction = tf.equal(tf.argmax(y_pred, 1), tf.cast(y_true, tf.int64))
    return tf.reduce_mean(tf.cast(correct_prediction, tf.float32), axis=-1)

optimizer = tf.optimizers.SGD(learning_rate)
def run_optimization(x, y):
    with tf.GradientTape() as g:
        pred = neural_net(x, is_training=True)
        loss = cross_entropy_loss(pred, y)
    trainable_variables = neural_net.trainable_variables
    gradients = g.gradient(loss, trainable_variables)
    optimizer.apply_gradients(zip(gradients, trainable_variables))
for step, (batch_x, batch_y) in enumerate(train_data.take(training_steps), 1):
    run_optimization(batch_x, batch_y)

    if step % display_step == 0:
        pred = neural_net(batch_x, is_training=True)
        loss = cross_entropy_loss(pred, batch_y)
        acc = accuracy(pred, batch_y)
        print("step: %i, loss: %f, accuracy: %f" % (step, loss, acc))
pred = neural_net(x_test, is_training=False)
print("Test Accuracy: %f" % accuracy(pred, y_test))
n_images = 10
test_images = x_test[:n_images]
predictions = neural_net(test_images)
for i in range(n_images):
    plt.imshow(np.reshape(test_images[i], [28, 28]), cmap='gray')
    plt.show()
    print("Model prediction: %i" % np.argmax(predictions.numpy()[i]))