model.fit()再现性

Question

import tensorflow as tf

RANDOM_SEED_CONSTANT = 42  # FOR_REPRODUCIBILITY
tf.random.set_seed(RANDOM_SEED_CONSTANT)

# Prevent NHWC errors https://www.nuomiphp.com/eplan/en/50125.html
from tensorflow.keras import backend as K
K.set_image_data_format("channels_last")

from tensorflow import keras
from tensorflow.keras import datasets, layers, models

(train_images, train_labels), (test_images, test_labels) = datasets.cifar10.load_data()
train_images, test_images = train_images / 255.0, test_images / 255.0 # Normalize pixel values to be between 0 and 1

# Create a simple CNN
model = models.Sequential()
model.add(layers.Conv2D(32, (3, 3), activation='relu', input_shape=(32, 32, 3)))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(64, (3, 3), activation='relu'))
model.add(layers.MaxPooling2D((2, 2)))
model.add(layers.Conv2D(64, (3, 3), activation='relu'))
model.add(layers.Flatten())
model.add(layers.Dense(64, 
                       activation='relu', 
                       kernel_initializer=tf.keras.initializers.HeNormal(seed=RANDOM_SEED_CONSTANT)))
model.add(layers.Dense(10, 
                       kernel_initializer=tf.keras.initializers.HeNormal(seed=RANDOM_SEED_CONSTANT)))

print(model.summary())

model.compile(optimizer='adam',
              loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
              metrics=['accuracy'])

model.save_weights('myweights.h5')

# Run1
history = model.fit(train_images, train_labels, epochs=1, 
                    shuffle=False,
                    validation_data=(test_images, test_labels))

# Run2
model.load_weights('myweights.h5')
history = model.fit(train_images, train_labels, epochs=1, 
                    shuffle=False,
                    validation_data=(test_images, test_labels))

# Run3
model.load_weights('myweights.h5')
history = model.fit(train_images, train_labels, epochs=1, 
                    shuffle=False,
                    validation_data=(test_images, test_labels))

上面的3个model.fit()调用给出了以下结果：

1563/1563 [==============================] - 7s 4ms/step - loss: 1.4939 - accuracy: 0.4543 - val_loss: 1.2516 - val_accuracy: 0.5567

1563/1563 [==============================] - 6s 4ms/step - loss: 1.6071 - accuracy: 0.4092 - val_loss: 1.3857 - val_accuracy: 0.4951

1563/1563 [==============================] - 7s 4ms/step - loss: 1.5538 - accuracy: 0.4325 - val_loss: 1.3187 - val_accuracy: 0.5294

造成这种差异的原因是什么？我试图了解可能阻碍模型复制结果的来源。除了随机的种子，密集的层初始化，我还遗漏了什么？

Answer 1

你测试重复性的方式是不正确的。您需要关闭程序并重新运行它，以查看结果是否相同。否则，运行2取决于运行1期间发生的事件，而run 3取决于运行1和2期间发生的事件。

原因是，Tensorflow维护了一个随机生成的内部计数器，正如tf.random.set_seed文档( counter 是属于我的)中所述：

打印(tf.random.uniform(1，seed=1)) #生成“A1”打印(tf.random.uniform(1，seed=1)) #生成“A2”

我们之所以在上面的第二次调用中使用“A2”而不是“A1”，是因为TensorFlow使用相同的tf.random.uniform内核(即内部表示)来处理所有具有相同参数的调用，而内核维护一个内部计数器，每次执行时都会增加，从而产生不同的结果。

如果我只运行您程序的第一次运行两次，并在每次运行之间关闭该程序(在这种情况下是IPython )，我将得到：

In [1]: run program.py
1563/1563 [==============================] - 13s 8ms/step - loss: 1.4997 - accuracy: 0.4540 - val_loss: 1.2528 - val_accuracy: 0.5494
{'loss': [1.4996991157531738], 'accuracy': [0.4540199935436249], 'val_loss': [1.2527965307235718], 'val_accuracy': [0.5493999719619751]}
In [2]: run program.py
1563/1563 [==============================] - 12s 8ms/step - loss: 1.4997 - accuracy: 0.4540 - val_loss: 1.2528 - val_accuracy: 0.5494
{'loss': [1.4996991157531738], 'accuracy': [0.4540199935436249], 'val_loss': [1.2527965307235718], 'val_accuracy': [0.5493999719619751]}

减去执行计算所需的时间，这可能会因机器上的负载而略有变化，结果完全相同。