tensorflow模型怎么查看參數(shù)

要查看TensorFlow模型的參數(shù)，可以使用model.summary()方法來(lái)打印出模型的結(jié)構(gòu)和參數(shù)數(shù)量。示例代碼如下：

import tensorflow as tf

# 創(chuàng)建模型
model = tf.keras.Sequential([
    tf.keras.layers.Dense(64, activation='relu', input_shape=(784,)),
    tf.keras.layers.Dense(64, activation='relu'),
    tf.keras.layers.Dense(10)
])

# 打印模型的參數(shù)
model.summary()

運(yùn)行以上代碼，會(huì)輸出類似如下的結(jié)果：

Model: "sequential"
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense (Dense)                (None, 64)                50240     
_________________________________________________________________
dense_1 (Dense)              (None, 64)                4160      
_________________________________________________________________
dense_2 (Dense)              (None, 10)                650       
=================================================================
Total params: 55,050
Trainable params: 55,050
Non-trainable params: 0
_________________________________________________________________

以上結(jié)果中，Param #列顯示了每個(gè)層的參數(shù)數(shù)量，Total params列顯示了總的參數(shù)數(shù)量。