tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False)
有些函数经过softmax激活,输出的是概率分布,有些函数则是原始输出;
from_logits=False 是在询问受否是原始输出,即没有经过概率分布的输出
如果神经网络预测结果输出前经过了概率分布,则为False
如果神经网络预测结果输出前没有经过概率分布,则为True
import tensorflow as tf
from sklearn import datasets
import numpy as np
x_train = datasets.load_iris().data
y_train = datasets.load_iris().target
np.random.seed(116)
np.random.shuffle(x_train)
np.random.seed(116)
np.random.shuffle(y_train)
tf.random.set_seed(116)
model = tf.keras.models.Sequential([
tf.keras.layers.Dense(3,activation='softmax',kernel_regularizer = tf.keras.regularizers.l2())
])
model.compile(optimizer = tf.keras.optimizers.SGD(lr = 0.1),
loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False),
metrics = ['sparse_categorical_accuracy'])
model.fit(x_train,y_train,batch_size=32,epochs=500,validation_split=0.2,validation_freq=20)
model.summary()
"""
val_sparse_categorical_accuracy: 1.0000
Model: "sequential_1"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
dense_1 (Dense) multiple 15
=================================================================
Total params: 15
Trainable params: 15
Non-trainable params: 0
_________________________________________________________________
"""
validation_freq=20
每迭代20次训练集,要在测试集中验证一次准确率
model.summary() 打印出网络结构和参数统计