softmax 独热编码

import tensorflow as tf
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

# fashion_mnist = tf.keras.datasets.fashion_mnist.load_data()
(train_image, train_lable), (test_image, test_label) = tf.keras.datasets.fashion_mnist.load_data()
# print(train_image.shape)  # (60000, 28, 28)
# print(train_lable.shape)  # (60000,)
# print(test_image.shape)  # (10000, 28, 28)
# print(test_label.shape)  # (10000,)
# plt.imshow(train_image[0])
# plt.show()
# print(np.max(train_image[0])) # 255
# print(train_lable)
# 归一化
train_image = train_image / 255
test_image = test_image / 255

# print(train_image.shape)

# 建模
model = tf.keras.Sequential()
model.add(tf.keras.layers.Flatten(input_shape=(28, 28)))  # 变成 28*28的向量
model.add(tf.keras.layers.Dense(128, activation='relu'))
model.add(tf.keras.layers.Dense(10, activation='softmax'))
#
# model.compile(optimizer='adam'
#               , loss='sparse_categorical_crossentropy'
#               , metrics=['acc'])
# model.fit(train_image,train_lable,epochs=5)
#
# print(model.evaluate(test_image, test_label))




# 独热编码
train_lable_onehot = tf.keras.utils.to_categorical(train_lable)
# print(train_lable_onehot[-1])

test_label_onehot= tf.keras.utils.to_categorical(test_label)
# print(test_label_onehot)

model.compile(optimizer='adam'
              , loss='categorical_crossentropy'
              , metrics=['acc'])

model.fit(train_image,train_lable_onehot,epochs=5)   # loss: 0.2938 - acc: 0.8928
predict = model.predict(test_image)
print(predict.shape)
print(predict[0])
print(np.argmax(predict[0]))  #获取最大概率的位置
print(test_label[0])