应该有人真的可以澄清这一点。
以下是Keras文档中的一些初始信息: Keras中的 fit 函数仅针对给定数量的纪元训练模型。 evaluate 函数返回测试模式下模型的损失值和指标值。
因此,两个函数都返回损失。为了举例说明,如果我有一个训练示例,则在每个训练步骤后,我从拟合函数中获得的损失应与从评估函数(在相同的训练步骤后)中获得的损失相同。 (这里的假设是,我在同一训练集上同时运行 fit 和 evaluate 函数(仅包含1个示例)。
我的网络定义如下:
def identity_loss(y_true, y_pred):
return K.mean(y_pred - 0 * y_true)
model = ResNet50(weights='imagenet')
model.layers.pop()
x = model.get_layer('flatten_1').output # layer 'flatten_1' is the last layer of the model
model_out = Dense(128, activation='relu', name='model_out')(x)
model_out = Lambda(lambda x: K.l2_normalize(x,axis=-1))(model_out)
new_model = Model(inputs=model.input, outputs=model_out)
anchor_input = Input(shape=(224, 224, 3), name='anchor_input')
pos_input = Input(shape=(224, 224, 3), name='pos_input')
neg_input = Input(shape=(224, 224, 3), name='neg_input')
encoding_anchor = new_model(anchor_input)
encoding_pos = new_model(pos_input)
encoding_neg = new_model(neg_input)
loss = Lambda(triplet_loss)([encoding_anchor, encoding_pos, encoding_neg])
siamese_network = Model(inputs = [anchor_input, pos_input, neg_input],
outputs = loss)
siamese_network.compile(loss=identity_loss, optimizer=Adam(lr=.00003))
后来,我使用拟合功能训练我的火车组(仅包含1个示例),历时10个纪元。为了检查拟合和评估函数之间的差异,我还在每个纪元的拟合函数之后立即运行评估函数,输出看起来像下面这样:
nr_epoch: 0
Epoch 1/1
1/1 [==============================] - 4s 4s/step - loss: 2.0035
1/1 [==============================] - 3s 3s/step
eval_score for train set: 2.0027356147766113
nr_epoch: 1
Epoch 1/1
1/1 [==============================] - 1s 1s/step - loss: 1.9816
1/1 [==============================] - 1s 1s/step
eval_score for train set: 2.001833915710449
nr_epoch: 2
Epoch 1/1
1/1 [==============================] - 1s 1s/step - loss: 1.9601
1/1 [==============================] - 1s 1s/step
eval_score for train set: 2.00126576423645
nr_epoch: 3
Epoch 1/1
1/1 [==============================] - 1s 1s/step - loss: 1.9388
1/1 [==============================] - 1s 1s/step
eval_score for train set: 2.0009117126464844
nr_epoch: 4
Epoch 1/1
1/1 [==============================] - 1s 1s/step - loss: 1.9176
1/1 [==============================] - 1s 1s/step
eval_score for train set: 2.000725746154785
nr_epoch: 5
Epoch 1/1
1/1 [==============================] - 1s 1s/step - loss: 1.8964
1/1 [==============================] - 1s 1s/step
eval_score for train set: 2.0006520748138428
nr_epoch: 6
Epoch 1/1
1/1 [==============================] - 1s 1s/step - loss: 1.8759
1/1 [==============================] - 1s 1s/step
eval_score for train set: 2.0006656646728516
nr_epoch: 7
Epoch 1/1
1/1 [==============================] - 1s 1s/step - loss: 1.8555
1/1 [==============================] - 1s 1s/step
eval_score for train set: 2.0007567405700684
nr_epoch: 8
Epoch 1/1
1/1 [==============================] - 1s 1s/step - loss: 1.8355
1/1 [==============================] - 1s 1s/step
eval_score for train set: 2.0009000301361084
nr_epoch: 9
Epoch 1/1
1/1 [==============================] - 2s 2s/step - loss: 1.8159
1/1 [==============================] - 2s 2s/step
eval_score for train set: 2.001085042953491
如图所示, fit 函数(在每个纪元末)报告的损失正在减少。来自评估函数的损失并没有减少。
因此,难题是:如果我在1个单个训练示例上运行模型,我是否应该从拟合和评估同一时期的函数中看到相同的损失(在每个时期之后)?如果我继续训练,火车的损失在减少,但是来自评估函数的损失却保持在同一水平,并且不会减少
最后,这是我如何称呼拟合和评估函数:
z = np.zeros(len(anchor_path))
siamese_network.fit(x=[anchor_imgs, pos_imgs, neg_imgs],
y=z,
batch_size=batch_size,
epochs=1,
verbose=1,
callbacks=None,
validation_split=0.0,
validation_data=None,
shuffle=True,
class_weight=None,
sample_weight=None,
initial_epoch=0,
steps_per_epoch=None,
validation_steps=None)
eval_score = siamese_network.evaluate(x=[anchor_imgs, pos_imgs, neg_imgs],
y=z,
batch_size = batch_size,
verbose = 1)
print('eval_score for train set: ', eval_score)
那么,为什么在fit函数执行过程中损失减少了,而评估函数却没有?我在哪里犯错了?
答案 0 :(得分:1)
ResNet使用批处理归一化,在训练和测试过程中表现不一样。您认为应该从model.fit
和model.evaluate
得到相同的训练损失的假设是错误的。
答案 1 :(得分:0)
通过进一步的研究(通过使用不同的关键字进行搜索),我发现了以下信息,这些信息也提供了解决方案。看来,许多人一直在遭受这个问题的困扰,尤其是在尝试利用迁移学习时。
以下是该问题的讨论和解决方案: Strange behaviour of the loss function in keras model, with pretrained convolutional base
这是有关此主题的博客文章: http://blog.datumbox.com/the-batch-normalization-layer-of-keras-is-broken/
很不幸,我认为Tensorflow和Keras都有非常糟糕的文档。