如何在Keras中保存训练期间保存的模型列表（仅保存最佳）？

如果您为每个时期存储到保存的模型的所有路径，则可以使用Callback，因为Callback它只是一个python对象，可以收集数据。

例如，它可以在列表中存储模型的路径：

import datetime

class SaveEveryEpoch(Callback):
    def __init__(self, model_name, *args, **kwargs):
        self.model_checkpoint_paths = []
        self.model_name = model_name
        super().__init__(*args, **kwargs)

    def on_epoch_end(self, epoch, logs):
        # I suppose here it's a Functional model
        print(logs['acc'])
        path_to_checkpoint = (
            str(datetime.datetime.now()).split(' ')[0] 
            + f'_{self.model_name}'
            + f'_{epoch:02d}.hdf5'
        )
        self.model.save(path_to_checkpoint)
        self.model_checkpoint_paths.append(path_to_checkpoint)

• __init__ 初始化一个空列表并存储模型基本名称
• on_epoch_end在每个时期结束时保存模型；还将模型路径追加到模型路径列表中

使用例

from keras.datasets import mnist
from keras.models import Model
from keras.layers import Dense, Input
import numpy as np

(X_tr, y_tr), (X_te, y_te) = mnist.load_data()
X_tr = (X_tr / 255.).reshape((60000, 784))
X_te = (X_te / 255.).reshape((10000, 784))


def binarize_labels(y):
    y_bin = np.zeros((len(y), len(np.unique(y)))) 
    y_bin[range(len(y)), y] = 1
    return y_bin

y_train_bin, y_test_bin = binarize_labels(y_tr), binarize_labels(y_te)

model = Sequential()
model.add(InputLayer((784,)))
model.add(Dense(784, activation='relu'))
model.add(Dense(256, activation='relu'))
model.add(Dense(10, activation='softmax'))

model = Model(inp, out)
model.compile(loss='categorical_crossentropy', optimizer='adam')

使用检查点运行模型

checkpoints = SaveEveryEpoch('mnist_model')
history = model.fit(X_tr, y_train_bin, callbacks=[checkpoints], epochs=3)
# ... training progress ...
checkpoints.model_checkpoint_paths
Out:
['2019-02-06_mnist_model_00.hdf5',
 '2019-02-06_mnist_model_01.hdf5',
 '2019-02-06_mnist_model_02.hdf5']

ls 输出：

2019-02-06_mnist_model_00.hdf5
2019-02-06_mnist_model_01.hdf5
2019-02-06_mnist_model_02.hdf5

变化

修改on_epoch_end以创建一些集合，该集合可以通过进行排序（loss例如，logs参数包含一个字典，该字典包含，loss并且提供了一个度量，acc如果提供了某个度量，则该度量名为）。因此，您以后可以选择损耗/度量值最小的模型：

class SaveEveryEpoch(Callback):
    def __init__(self, model_name, *args, **kwargs):
        self.model_checkpoints_with_loss = []
        self.model_name = model_name
        super().__init__(*args, **kwargs)

    def on_epoch_end(self, epoch, logs):
        # I suppose here it's a Functional model
        print(logs['acc'])
        path_to_checkpoint = (
            str(datetime.datetime.now()).split(' ')[0] 
            + f'_{self.model_name}'
            + f'_{epoch:02d}.hdf5'
        )
        self.model.save(path_to_checkpoint)
        self.model_checkpoints_with_loss.append((logs['loss'], path_to_checkpoint))

另外，您可以重载默认回调，例如ModelCheckpoint，以保存所有路径，不仅是最佳模型，而且我想在这种情况下没有必要。