Question

我使用Keras中的Dense Neural Networks组合了一个VAE。在model.fit期间，我遇到了尺寸不匹配的问题，但不确定是什么使代码丢了。下面是我的代码的样子

from keras.layers import Lambda, Input, Dense
from keras.models import Model
from keras.datasets import mnist
from keras.losses import mse, binary_crossentropy
from keras.utils import plot_model
from keras import backend as K
import keras

import numpy as np
import matplotlib.pyplot as plt
import argparse
import os

(x_train, y_train), (x_test, y_test) = mnist.load_data()

image_size = x_train.shape[1]
original_dim = image_size * image_size
x_train = np.reshape(x_train, [-1, original_dim])
x_test = np.reshape(x_test, [-1, original_dim])
x_train = x_train.astype('float32') / 255
x_test = x_test.astype('float32') / 255

# network parameters
input_shape = (original_dim, )
intermediate_dim = 512
batch_size = 128
latent_dim = 2
epochs = 50


x = Input(batch_shape=(batch_size, original_dim))
h = Dense(intermediate_dim, activation='relu')(x)
z_mean = Dense(latent_dim)(h)
z_log_sigma = Dense(latent_dim)(h)

def sampling(args):
    z_mean, z_log_sigma = args
    #epsilon = K.random_normal(shape=(batch, dim))
    epsilon = K.random_normal(shape=(batch_size, latent_dim))
    return z_mean + K.exp(z_log_sigma) * epsilon

# note that "output_shape" isn't necessary with the TensorFlow backend
# so you could write `Lambda(sampling)([z_mean, z_log_sigma])`
z = Lambda(sampling, output_shape=(latent_dim,))([z_mean, z_log_sigma])

decoder_h = Dense(intermediate_dim, activation='relu')
decoder_mean = Dense(original_dim, activation='sigmoid')
h_decoded = decoder_h(z)
x_decoded_mean = decoder_mean(h_decoded)

print('X Decoded Mean shape: ', x_decoded_mean.shape)

# end-to-end autoencoder
vae = Model(x, x_decoded_mean)

# encoder, from inputs to latent space
encoder = Model(x, z_mean)

# generator, from latent space to reconstructed inputs
decoder_input = Input(shape=(latent_dim,))
_h_decoded = decoder_h(decoder_input)
_x_decoded_mean = decoder_mean(_h_decoded)
generator = Model(decoder_input, _x_decoded_mean)

def vae_loss(x, x_decoded_mean):
    xent_loss = keras.metrics.binary_crossentropy(x, x_decoded_mean)
    kl_loss = - 0.5 * K.mean(1 + z_log_sigma - K.square(z_mean) - K.exp(z_log_sigma), axis=-1)
    return xent_loss + kl_loss

vae.compile(optimizer='rmsprop', loss=vae_loss)


print('X train shape: ', x_train.shape)
print('X test shape: ', x_test.shape)

vae.fit(x_train, x_train,
        shuffle=True,
        epochs=epochs,
        batch_size=batch_size,
        validation_data=(x_test, x_test))

这是调用model.fit时看到的堆栈跟踪。

File "/home/asattar/workspace/projects/keras-examples/blogautoencoder/VariationalAutoEncoder.py", line 81, in <module>
    validation_data=(x_test, x_test))
  File "/usr/local/lib/python2.7/dist-packages/Keras-2.2.4-py2.7.egg/keras/engine/training.py", line 1047, in fit
    validation_steps=validation_steps)
  File "/usr/local/lib/python2.7/dist-packages/Keras-2.2.4-py2.7.egg/keras/engine/training_arrays.py", line 195, in fit_loop
    outs = fit_function(ins_batch)
  File "/usr/local/lib/python2.7/dist-packages/Keras-2.2.4-py2.7.egg/keras/backend/tensorflow_backend.py", line 2897, in __call__
    return self._call(inputs)
  File "/usr/local/lib/python2.7/dist-packages/Keras-2.2.4-py2.7.egg/keras/backend/tensorflow_backend.py", line 2855, in _call
    fetched = self._callable_fn(*array_vals)
  File "/home/asattar/.local/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1439, in __call__
    run_metadata_ptr)
  File "/home/asattar/.local/lib/python2.7/site-packages/tensorflow/python/framework/errors_impl.py", line 528, in __exit__
    c_api.TF_GetCode(self.status.status))
tensorflow.python.framework.errors_impl.InvalidArgumentError: Incompatible shapes: [128,784] vs. [96,784]
     [[{{node training/RMSprop/gradients/loss/dense_5_loss/logistic_loss/mul_grad/BroadcastGradientArgs}} = BroadcastGradientArgs[T=DT_INT32, _class=["loc:@train...ad/Reshape"], _device="/job:localhost/replica:0/task:0/device:CPU:0"](training/RMSprop/gradients/loss/dense_5_loss/logistic_loss/mul_grad/Shape, training/RMSprop/gradients/loss/dense_5_loss/logistic_loss/mul_grad/Shape_1)]]

请注意跟踪结尾处的“堆栈跟踪中的不兼容形状：[128,784]与[96,784]”。

Answer 1

根据Keras: What if the size of data is not divisible by batch_size?，此处最好使用model.fit_generator而不是model.fit。

要使用model.fit_generator，应定义自己的生成器对象。以下是一个示例：

from keras.utils import Sequence
import math

class Generator(Sequence):
    # Class is a dataset wrapper for better training performance
    def __init__(self, x_set, y_set, batch_size=256):
        self.x, self.y = x_set, y_set
        self.batch_size = batch_size
        self.indices = np.arange(self.x.shape[0])

    def __len__(self):
        return math.floor(self.x.shape[0] / self.batch_size)

    def __getitem__(self, idx):
        inds = self.indices[idx * self.batch_size:(idx + 1) * self.batch_size]
        batch_x = self.x[inds]
        batch_y = self.y[inds]
        return batch_x, batch_y

    def on_epoch_end(self):
        np.random.shuffle(self.indices)

train_datagen = Generator(x_train, x_train, batch_size)
test_datagen = Generator(x_test, x_test, batch_size)

vae.fit_generator(train_datagen,
    steps_per_epoch=len(x_train)//batch_size,
    validation_data=test_datagen,
    validation_steps=len(x_test)//batch_size,
    epochs=epochs)

从How to shuffle after each epoch using a custom generator?开始采用的代码。

Answer 2

只是尝试复制并发现在您定义时

x = Input(batch_shape=(batch_size, original_dim))

您正在设置批次大小，并且在开始验证时会导致不匹配。更改为

x = Input(shape=input_shape)

您应该已经准备就绪。

model.fit

2 个答案: