Tengo que decir que tensorflow2.0 es mejor que 1.x
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import datasets
import os
#os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
# x: [60k, 28, 28], [10, 28, 28]
# y: [60k], [10k]
# 加载数据
(x, y), (x_test, y_test) = datasets.mnist.load_data()
# x: [0~255] => [0~1.]
x = tf.convert_to_tensor(x, dtype=tf.float32) / 255.
y = tf.convert_to_tensor(y, dtype=tf.int32)
x_test = tf.convert_to_tensor(x_test, dtype=tf.float32) / 255.
y_test = tf.convert_to_tensor(y_test, dtype=tf.int32)
print(x.shape, y.shape, x.dtype, y.dtype)
print(tf.reduce_min(x), tf.reduce_max(x))
print(tf.reduce_min(y), tf.reduce_max(y))
train_db = tf.data.Dataset.from_tensor_slices((x, y)).batch(128)
test_db = tf.data.Dataset.from_tensor_slices((x_test, y_test)).batch(128)
train_iter = iter(train_db)
sample = next(train_iter)
#print('batch:', sample[0].shape, sample[1].shape)
# [b, 784] => [b, 256] => [b, 128] => [b, 10]
# [dim_in, dim_out], [dim_out]
w1 = tf.Variable(tf.random.truncated_normal([784, 256], stddev=0.1))
b1 = tf.Variable(tf.zeros([256]))
w2 = tf.Variable(tf.random.truncated_normal([256, 128], stddev=0.1))
b2 = tf.Variable(tf.zeros([128]))
w3 = tf.Variable(tf.random.truncated_normal([128, 10], stddev=0.1))
b3 = tf.Variable(tf.zeros([10]))
lr = 1e-3
with tf.device("cpu"):
for epoch in range(10): # iterate db for 10
for step, (x, y) in enumerate(train_db): # for every batch
# x:[128, 28, 28]
# y: [128]
# [b, 28, 28] => [b, 28*28]
x = tf.reshape(x, [-1, 28 * 28]) # -1：根据剩余维度计算
with tf.GradientTape() as tape: # tf.Variable计算梯度
# x: [b, 28*28]
# h1 = x@w1 + b1
# [b, 784]@[784, 256] + [256] => [b, 256] + [256] => [b, 256] + [b, 256]
h1 = x @ w1 + tf.broadcast_to(b1, [x.shape[0], 256])
h1 = tf.nn.relu(h1)
# [b, 256] => [b, 128]
h2 = h1 @ w2 + b2
h2 = tf.nn.relu(h2)
# [b, 128] => [b, 10]
out = h2 @ w3 + b3
# compute loss
# out: [b, 10]
# y: [b] => [b, 10]
y_onehot = tf.one_hot(y, depth=10)
# mse = mean(sum(y-out)^2)
# [b, 10]
loss = tf.square(y_onehot - out)
# mean: scalar
loss = tf.reduce_mean(loss)
# compute gradients
grads = tape.gradient(loss, [w1, b1, w2, b2, w3, b3])
# print(grads)
# w1 = w1 - lr * w1_grad
w1.assign_sub(lr * grads[0])
b1.assign_sub(lr * grads[1])
w2.assign_sub(lr * grads[2])
b2.assign_sub(lr * grads[3])
w3.assign_sub(lr * grads[4])
b3.assign_sub(lr * grads[5])
# if step % 100 == 0:
# print(epoch, step, 'loss:', float(loss))
# test/evluation
# [w1, b1, w2, b2, w3, b3]
total_correct, total_num = 0, 0
for step, (x, y) in enumerate(test_db):
# [b, 28, 28] => [b, 28*28]
x = tf.reshape(x, [-1, 28 * 28])
# [b, 784] => [b, 256] => [b, 128] => [b, 10]
h1 = tf.nn.relu(x @ w1 + b1)
h2 = tf.nn.relu(h1 @ w2 + b2)
out = h2 @ w3 + b3
# out: [b, 10] ~ R
# prob: [b, 10] ~ [0, 1]
prob = tf.nn.softmax(out, axis=1)
# [b, 10] => [b]
# int64!!!
pred = tf.argmax(prob, axis=1)
pred = tf.cast(pred, dtype=tf.int32)
# y: [b]
# [b], int32
# print(pred.dtype, y.dtype)
# 0=预测正确；1=预测错误
correct = tf.cast(tf.equal(pred, y), dtype=tf.int32)
correct = tf.reduce_sum(correct)
total_correct += int(correct)
total_num += x.shape[0]
acc = total_correct / total_num
print('test acc:', acc)
mejoramiento
El código usa api de alto nivel y luego profundiza el número de capas de red. Encontrará que la precisión del 95% se alcanza en la primera ronda. Esto es sorprendente. La red superficial anterior solo se ha entrenado para varias rondas con una precisión de aproximadamente 65%. .
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import datasets, layers, optimizers, Sequential, metrics
import os
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'
def preprocess(x, y):
x = tf.cast(x, dtype=tf.float32) / 255.
y = tf.cast(y, dtype=tf.int32)
return x, y
(x, y), (x_test, y_test) = datasets.mnist.load_data()
print(x.shape, y.shape)
batchsz = 128
# 让每一个x都对应一个y
db = tf.data.Dataset.from_tensor_slices((x, y))
# 混合顺序
db = db.map(preprocess).shuffle(10000).batch(batchsz)
db_test = tf.data.Dataset.from_tensor_slices((x_test, y_test))
# 指定函数形成映射关系
db_test = db_test.map(preprocess).batch(batchsz)
db_iter = iter(db)
sample = next(db_iter)
print('batch:', 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], 330 = 32*10 + 10
])
model.build(input_shape=[None, 28 * 28])
model.summary()
# w = w - lr*grad
optimizer = optimizers.Adam(lr=1e-3)
def main():
for epoch in range(30):
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:
# [b, 784] => [b, 10]
logits = model(x)
y_onehot = tf.one_hot(y, depth=10)
# [b]
loss_mse = tf.reduce_mean(tf.losses.MSE(y_onehot, logits))
loss_ce = tf.losses.categorical_crossentropy(y_onehot, logits, from_logits=True)
loss_ce = tf.reduce_mean(loss_ce)
grads = tape.gradient(loss_ce, model.trainable_variables)
optimizer.apply_gradients(zip(grads, model.trainable_variables))
if step % 100 == 0:
print(epoch, step, 'loss:', float(loss_ce), float(loss_mse))
# test
total_correct = 0
total_num = 0
for x, y in db_test:
# x: [b, 28, 28] => [b, 784]
# y: [b]
x = tf.reshape(x, [-1, 28 * 28])
# [b, 10]
logits = model(x)
# logits => prob, [b, 10]
prob = tf.nn.softmax(logits, axis=1)
# [b, 10] => [b], int64
pred = tf.argmax(prob, axis=1)
pred = tf.cast(pred, dtype=tf.int32)
# pred:[b]
# y: [b]
# correct: [b], True: equal, False: not equal
correct = tf.equal(pred, y)
correct = tf.reduce_sum(tf.cast(correct, dtype=tf.int32))
total_correct += int(correct)
total_num += x.shape[0]
acc = total_correct / total_num
print(epoch, 'test acc:', acc)
if __name__ == '__main__':
main()