FashionMNNST数据集
使用tensorflow2.x,实现FashionMNNST数据集的简单网络结构的搭建(5层全连接层),以及测试。
# -*- coding: utf-8 -*-
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import datasets,layers,optimizers,Sequential,metrics
def preprocess(x,y):#数据预处理
x = tf.cast(x, dtype=tf.float32)/255.
y = tf.cast(y, dtype=tf.int32)
return x,y
#加载数据集
(x, y), (x_test, y_test) = datasets.fashion_mnist.load_data()
print(x.shape, y.shape)#(60000, 28, 28) (60000,) (10000, 28, 28) (10000,)
batchsz = 128
db = tf.data.Dataset.from_tensor_slices((x,y))
db = db.map(preprocess).shuffle(10000).batch(batchsz) #调用预处理函数,对每一个x,y进行处理,打散数据,截断
db_test = tf.data.Dataset.from_tensor_slices((x_test,y_test))
db_test = db_test.map(preprocess).batch(batchsz)
db_iter = iter(db)
sample = next(db_iter)
print('batch:',sample[0].shape,sample[1].shape)
#Squential是个容器,将创建的list传给Sequential,它就是一个网络层
model = Sequential([
layers.Dense(256, activation = tf.nn.relu), #[b,784]==>[b,256] 全连接层,线性激活函数
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])#喂一个输入,构建一个权值
model.summary()#调试,可以将网络结构打印出来
#优化器 : w = w - lr*grad
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])
#求梯度
with tf.GradientTape() as tape:
#[b,784] ==> [b,10]
logits = model(x)#调用网络层,得到没有归一化的输出值logits
y_onehot = tf.one_hot(y, depth = 10) #将y转成one_hot
loss_mse = tf.reduce_mean(tf.losses.MSE(y_onehot, logits)) #均方差损失函数MSE
loss_ce = tf.losses.categorical_crossentropy(y_onehot, logits, from_logits=True) #交叉熵损失函数
loss_ce = tf.reduce_mean(loss_ce)
grads = tape.gradient(loss_ce, model.trainable_variables) #model.trainable_variables返回【w,b】参数列表
optimizer.apply_gradients(zip(grads, model.trainable_variables)) #原地更新参数
if step%100 == 0:
print(epoch,step,'loss:',float(loss_ce),float(loss_mse))
total_correct, total_num = 0, 0
#test
for x,y in enumerate(db_test):
# x:[b,28,28] ==> [b,784]
# y:[b]
x = tf.reshape(x, [-1, 28 * 28])
# [b,784] ==> [b,10]
logits = model(x) # 调用网络层,使用当前状态的【w,b】,得出测试图片的输出值
#logits ==> prob 将输出值转换成概率(归一化)
prob = tf.nn.softmax(logits, axis = 1)
#[b,10] ==> [b] pred: int64 ==> int32
pred = tf.argmax(prob, axis=1) #获取prob的最大得分的索引位置【0-9】
pred = tf.cast(pred, dtype=tf.int32)
#pred:[b]
#y:[b] 存储的是类别的值【0-9】
correct = tf.equal(pred,y) #比较pred与y,相等返回True,不等返回False
correct = tf.reduce_sum(tf.cast(correct, dtype=tf.int32)) #将bool型转成int,求出b张图片中正确的个数
total_correct += int(correct) #求出测试数据集中总的正确个数
total_num += x.shape[0] #计算测试集的图片个数
acc = total_correct / total_num #计算正确率
print(epoch, 'test acc:', acc)
if __name__ == '__main__':
main()