從 CNN model 獲取權重和偏差並將它們保存到 csv 文件中

Question

我最近構建了一個 CNN model 用於 cifar 10 的圖像分類。model 有效，但我無法為每個可訓練層（卷積層和密集層）生成權重和偏差。 下面是在構建 model 然后加載它的代碼部分

  from __future__ import print_function

import keras
from keras.datasets import cifar10
from keras.preprocessing.image import ImageDataGenerator
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation, Flatten
from keras.layers import Conv2D, MaxPooling2D, AveragePooling2D
from keras.utils.data_utils import Sequence
import os
from keras import backend as K
import numpy as np



def square(x):
 return x * x;

batch_size = 64
num_classes = 10
epochs = 500
num_predictions = 20
save_dir = os.path.join(os.getcwd(), 'saved_models')
model_name = 'cifar_model2.h5'

 # The data, split between train and test sets:
 (x_train, y_train), (x_test, y_test) = cifar10.load_data()

  x_train = x_train.astype('float32')
  x_test = x_test.astype('float32')
  x_train = np.round(x_train/32)/8
  x_test = np.round(x_test/32)/8
  print('x_train shape:', x_train.shape)
  print(x_train.shape[0], 'train samples')
  print(x_test.shape[0], 'test samples')

  y_train = keras.utils.to_categorical(y_train, num_classes)
  y_test = keras.utils.to_categorical(y_test, num_classes)

  model = Sequential()
  model.add(Conv2D(128, (3, 3), padding='same',
             input_shape=x_train.shape[1:]))
  model.add(AveragePooling2D(pool_size=(2, 2)))
  model.add(Conv2D(83, (3, 3)))
  model.add(Dropout(0.25))
  model.add(Activation(square))

  model.add(AveragePooling2D(pool_size=(2, 2)))
  model.add(Conv2D(130, (5, 5), padding='same'))
  model.add(Activation(square))
  model.add(AveragePooling2D(pool_size=(2, 2)))
  model.add(Dropout(0.25))

  model.add(Flatten())
  model.add(Dense(512))
  model.add(Dense(num_classes))
  model.add(Activation('softmax'))

  opt = keras.optimizers.Adam(amsgrad=True, decay=0.0001, lr = 0.001)

  model.compile(loss='categorical_crossentropy', optimizer=opt, metrics=['accuracy'])


  print('Using real-time data augmentation.')
  datagen = ImageDataGenerator(
      featurewise_center=False,  
      samplewise_center=False,  
      featurewise_std_normalization=False,
      samplewise_std_normalization=False, 
      zca_whitening=False,  
      zca_epsilon=1e-06,  
      rotation_range=0,  
      width_shift_range=0.1,
      height_shift_range=0.1,
      shear_range=0.,
      zoom_range=0., 
      channel_shift_range=0., 
      fill_mode='nearest',
      cval=0.,
      horizontal_flip=True,
      vertical_flip=False, 
      rescale=None,
      preprocessing_function=None,
      data_format=None,
      validation_split=0.0)
 if not os.path.isdir(save_dir):
     os.makedirs(save_dir)
 model_path = os.path.join(save_dir, model_name)
 checkpoint = keras.callbacks.ModelCheckpoint(model_path, monitor='val_loss', verbose=0,    
 save_best_only=False, save_weights_only=False, mode='auto', period=10)
 callback_list=[checkpoint]
 datagen.fit(x_train)
 model.fit_generator(datagen.flow(x_train, y_train, batch_size=batch_size), 
 callbacks=callback_list,steps_per_epoch=len(x_train)/batch_size,
                    epochs=epochs, validation_data=(x_test, y_test), workers=4)

# Save 
 model.save(model_path)
 print('Saved trained model at %s ' % model_path)

 scores = model.evaluate(x_test, y_test, verbose=1)
 print('Test loss:', scores[0])
 print('Test accuracy:', scores[1])


from numpy import loadtxt
from keras.models import load_model
save_dir = os.path.join(os.getcwd(), 'saved_models')
model_name = 'cifar_model2.h5'
model_path = os.path.join(save_dir, model_name)
def square(x):
   return x * x;
# load model
model = load_model(model_path, custom_objects={'square': square})]

接下來我輸入代碼：

model.summary()

用於查看所有帶有參數的層（可訓練層）。 它顯示了圖層：

_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
conv2d_4 (Conv2D)            (None, 32, 32, 128)       3584      
_________________________________________________________________
average_pooling2d_4 (Average (None, 16, 16, 128)       0         
_________________________________________________________________
conv2d_5 (Conv2D)            (None, 14, 14, 83)        95699     
_________________________________________________________________
dropout_3 (Dropout)          (None, 14, 14, 83)        0         
_________________________________________________________________
activation_4 (Activation)    (None, 14, 14, 83)        0         
_________________________________________________________________
average_pooling2d_5 (Average (None, 7, 7, 83)          0         
_________________________________________________________________
conv2d_6 (Conv2D)            (None, 7, 7, 130)         269880    
_________________________________________________________________
activation_5 (Activation)    (None, 7, 7, 130)         0         
_________________________________________________________________
average_pooling2d_6 (Average (None, 3, 3, 130)         0         
_________________________________________________________________
dropout_4 (Dropout)          (None, 3, 3, 130)         0         
_________________________________________________________________
flatten_2 (Flatten)          (None, 1170)              0         
_________________________________________________________________
dense_3 (Dense)              (None, 512)               599552    
_________________________________________________________________
dense_4 (Dense)              (None, 10)                5130      
_________________________________________________________________
activation_6 (Activation)    (None, 10)                0         
=================================================================

總參數：973,845 可訓練參數：973,845 不可訓練參數：0

現在的問題是我如何獲得每一層的權重和偏差，然后將它們保存在“csv”文件中？我剛剛使用下面的代碼設法獲得了單層的權重和偏差：

 from numpy import loadtxt
 from keras.models import load_model
 from keras.engine import InputLayer

 output_center=2
 input_kernel_range = range(0,10)
 input_shape=[14,14,83]
 input_maps = input_shape[2]
 model_ = Sequential()
 model_.add(AveragePooling2D(pool_size=(2, 2),input_shape=[14,14,83]))
 model_.add(Conv2D(130, (5, 5), padding='same'))
 model_.layers[1].set_weights(model.layers[6].get_weights())
 test = np.zeros(input_shape)[np.newaxis,...]
 bias = model_.predict(test)[0,output_center,output_center,:]
 A = 0
for c in range(0, input_maps):
   for y in input_kernel_range:
      for x in input_kernel_range:
         test = np.zeros(input_shape)[np.newaxis,...]
         test[0,x,y,c]=1
         prediction = model_.predict(test)
         d = prediction[0,output_center,output_center,:] - bias
         if (isinstance(A, int)):
            A = d
        else:
            A = np.c_[A, d]
A.tofile("weights.csv", sep=',')
bias.tofile("bias.csv", sep=',')

這僅適用於一層。 就像我說的，我正在嘗試獲取權重和偏差，並將其保存在“csv”中，用於所有卷積層和密集層。有人可以幫忙嗎？

提前致謝

Answer 1

要獲得 Conv2D 和 Dense 層的權重和偏差，您可以執行以下操作：

wt = open('weights.csv', 'w')
bs = open('biases.csv', 'w')
for idx,i in enumerate(model.layers):
  if(isinstance(i, Conv2D) or isinstance(i, Dense)):
    weights = i.get_weights()[0]
    biases = i.get_weights()[1]
    weights = weights.flatten()
    biases = biases.flatten()
    print(weights.shape, biases.shape)
    wt.write(','.join(map(str,weights.tolist()))+"\n")
    bs.write(','.join(map(str,biases.tolist()))+"\n")
    #np.savetxt("weight" + str(idx)+".csv" , weights , fmt='%s', delimiter=',')
    #np.savetxt("bias" + str(idx) +".csv" , biases , fmt='%s', delimiter=',')

您還可以跟蹤圖層的原始形狀，因為它被展平以存儲在 CSV 中。 如果您以后想再次加載這些權重，這將對您有所幫助。