keras1.2.2报错 "Flatten" is not fully defined (got (0, 7, 512). "input_shape" or "batch_input_shape"

keras1.2.2以及theano里面报错如下：
File “demo/demo_train_miml_vgg.py”, line 28, in
base_model = VGG_16(vgg_model_path)
File “lib/cocodemo/vgg_16.py”, line 49, in VGG_16
model.add(Flatten())
File “/home/muli/.conda/envs/python2/lib/python2.7/site-packages/keras/models.py”, line 332, in add
output_tensor = layer(self.outputs[0])
output_shapes = to_list(outbound_layer.get_output_shape_for(input_shapes[0]))
File “/home/muli/.conda/envs/python2/lib/python2.7/site-packages/keras/layers/core.py”, line 476, in get_output_shape_for
'(got ’ + str(input_shape[1:]) + '. ’
ValueError: The shape of the input to “Flatten” is not fully defined (got (0, 7, 512). Make sure to pass a complete “input_shape” or “batch_input_shape” argument to the first layer in your model.

解决办法： ~/.keras/keras.json 文件，注意keras为隐藏文件，修改如下：

{
    "floatx": "float32",
    "image_data_format": "channels_first",
    "epsilon": 1e-07,
    "image_dim_ordering": "th",
    "backend": "theano"
}

参考链接：https://github.com/keras-team/keras/issues/3850