如何将fix超参数作为变量传递给Keras-Tuner？

Question

我想用Keras调谐器对Keras模型进行超参数调整。

import tensorflow as tf
from tensorflow import keras
import keras_tuner as kt

def model_builder(hp):

  model = keras.Sequential()
  model.add(keras.layers.Flatten(input_shape=(28, 28)))

  hp_units = hp.Int('units', min_value=32, max_value=512, step=32)
  model.add(keras.layers.Dense(units=hp_units, activation='relu'))
  model.add(keras.layers.Dense(10))

  hp_learning_rate = hp.Choice('learning_rate', values=[1e-2, 1e-3, 1e-4])

  model.compile(optimizer=keras.optimizers.Adam(learning_rate=hp_learning_rate),
                loss=keras.losses.SparseCategoricalCrossentropy(from_logits=True),
                metrics=['accuracy'])

  return model

tuner = kt.Hyperband(model_builder,
                     objective='val_accuracy',
                     max_epochs=10,
                     factor=3)

tuner.search(train_X, train_y, epochs=50)

到现在为止还好。但是，我还想定义一些模型参数(如输入图像尺寸)作为model_builder的输入参数，我不知道该怎么做：

def model_builder(hp, img_dim1, img_dim2):

  model = keras.Sequential()
  model.add(keras.layers.Flatten(input_shape=(img_dim1, img_dim2)))
...

和

tuner = kt.Hyperband(model_builder(img_dim1, img_dim2),
                     objective='val_accuracy',
                     max_epochs=10,
                     factor=3)

看起来好像不管用。如何在hp之外向模型提供img_dim1, img_dim2

Answer 1

一个简单的解决方案是在python中使用“部分函数”，如下所示：

from functools import partial

#...

model_builder_ready = partial(model_builder, img_dim1 = value1, img_dim2 = value2)

tuner = kt.Hyperband(model_builder_ready,
                     objective='val_accuracy',
                     max_epochs=10,
                     factor=3)