tf实现一个简单的fashionmnist识别

import tensorflow as tf
from  tensorflow import keras
from tensorflow.keras import datasets, layers, optimizers,Sequential,metrics

(x, y), (x_test, y_test) =  datasets.fashion_mnist.load_data()
print(x.shape, y.shape)

def proprocess(x, y):
    x = tf.cast(x, dtype=tf.float32) / 255.
    y = tf.cast(y, dtype=tf.int32)
    return x,y

batch_size = 128
db = tf.data.Dataset.from_tensor_slices((x, y))
db = db.map(proprocess).shuffle(10000).batch(batch_size)
db_test = tf.data.Dataset.from_tensor_slices((x_test, y_test))
db_test = db_test.map(proprocess).batch(batch_size)

#db_iter = iter(db)
#sample = next(db_iter)
#print(sample[0].shape, sample[1].shape)

model = Sequential([
    layers.Dense(256, activation=tf.nn.relu),#[b, 784] => [b, 256]
    layers.Dense(128, activation=tf.nn.relu),#[b, 256] => [b, 128]
    layers.Dense(64, activation=tf.nn.relu),#[b, 128] => [b, 64]
    layers.Dense(32, activation=tf.nn.relu),#[b, 64] => [b, 32]
    layers.Dense(10),#[b, 32] => [b, 10]
])
model.build(input_shape=[None, 28*28])
model.summary()

optimizer = optimizers.Adam(lr=1e-3)
def main():
    for epoch in range(10):
        for step, (x, y) in enumerate(db):
            # x = [b, 28, 28] => [b, 784]
            # y = [b]
            x = tf.reshape(x, [-1, 28*28])
            with tf.GradientTape() as tape:
                logits = model(x)
                y_onehot = tf.one_hot(y, depth=10)
                loss = tf.reduce_mean(tf.losses.MSE(y_onehot, logits))
                loss2 = tf.losses.categorical_crossentropy(y_onehot, logits, from_logits=True)
                loss2 = tf.reduce_mean(loss2)
            grads = tape.gradient(loss2, model.trainable_variables)
            optimizer.apply_gradients(zip(grads, model.trainable_variables))

            if step % 100 == 0:
                print(epoch, step, float(loss), float(loss2))
        # test

        total_correct = 0
        total_number = 0
        for (x_t, y_t) in db_test:
            x_t = tf.reshape(x_t, [-1, 28*28])
            logits_test = model(x_t)
            prob_test = tf.nn.softmax(logits_test, axis=1)
            pred = tf.argmax(prob_test, axis=1)
            correct = tf.equal(tf.cast(pred, dtype=tf.int32), y_t)
            correct = tf.reduce_sum(tf.cast(correct, dtype=tf.int32))
            total_correct += int(correct)
            total_number += x_t.shape[0]
        print("epoch, total_correct, total_number, acc", epoch, total_correct, total_number, float(total_correct/total_number))

if __name__ == '__main__':
    main()

这个代码很简单.

这个代码还可使用network.complie和net.fit来进行简化。