Question

我正在使用Keras构建神经网络模型：

输出如下所示。我想知道是否有可能输掉损失，比如说每10个时代而不是每个时代？谢谢！

model_keras = Sequential()
model_keras.add(Dense(4, input_dim=input_num, activation='relu',kernel_regularizer=regularizers.l2(0.01)))
model_keras.add(Dense(1, activation='linear',kernel_regularizer=regularizers.l2(0.01)))
sgd = optimizers.SGD(lr=0.01, clipnorm=0.5)
model_keras.compile(loss='mean_squared_error',  optimizer=sgd)
model_keras.fit(X_norm_train, y_norm_train, batch_size=20, epochs=100)

Answer 1

无法降低日志记录到stdout的频率，但是，将verbose=0参数传递给fit()方法会完全关闭日志记录。

由于在Keras＆＃39;中没有暴露出超过时代的循环。顺序模型，一种使用自定义频率收集标量变量摘要的方法是使用Keras callbacks。特别是，您可以使用TensorBoard（假设您正在运行tensorflow后端）或CSVLogger（任何后端）回调来收集任何标量变量摘要（在您的情况下，训练损失）：

from keras.callbacks import TensorBoard

model_keras = Sequential()
model_keras.add(Dense(4, input_dim=input_num, activation='relu',kernel_regularizer=regularizers.l2(0.01)))
model_keras.add(Dense(1, activation='linear',kernel_regularizer=regularizers.l2(0.01)))
sgd = optimizers.SGD(lr=0.01, clipnorm=0.5)
model_keras.compile(loss='mean_squared_error',  optimizer=sgd)

TB = TensorBoard(histogram_freq=10, batch_size=20)

model_keras.fit(X_norm_train, y_norm_train, batch_size=20, epochs=100, callbacks=[TB])

设置histogram_freq=10将每10个时期节省一次损失。

编辑：将validation_data=(...)传递给fit方法也可以检查验证级别指标。

Answer 2

创建Keras回调以减少日志行数。默认情况下，Keras每个时期都会打印日志。以下代码仅打印10条日志行，而不考虑时期数。

class callback(tf.keras.callbacks.Callback):
  def on_epoch_end(this,Epoch,Logs):
    L = Logs["loss"];

    if Epoch%Lafte==Lafte-1: #Log after a number of epochs
      print(f"Average batch loss: {L:.9f}");
    if Epoch==Epochs-1:
      print(f"Fin-avg batch loss: {L:.9f}"); #Final average

Model = model();
Model.compile(...);

Dsize  = ...   #Number of samples in training data
Bsize  = ...   #Number of samples to process in 1 batch
Steps  = 1000; #Number of batches to use to train
Epochs = round(Steps/(Dsize/Bsize));
Lafte  = round(Epochs/10); #Log 10 times only, regardless of num of Epochs
if Lafte==0: Lafte=1;      #Avoid modulus by zero in on_epoch_end

Model.fit(Data, epochs=Epochs, steps_per_epoch=round(Dsize/Bsize),
          callbacks=[callback()], verbose=0);

Keras模型输出信息/日志级别

2 个答案: