yashk2810/visualization-of-convolutional-layers

visualization-of-convolutional-layers's Issues

Visualization-of-Convolutional-Layers using images with 3 channels

I've tried your code from your jupyter notebook and would like to try it on a network I've trained for a different purpose. The difference from your set up is that I use Tensorflow, another network and color images (3-channels).

When i try this I get the following error:

Traceback (most recent call last): File "vis_NN.py", line 508, in <module> layer_to_visualize(layer_dict[layer_name]) File "vis_NN..py", line 485, in layer_to_visualize convolutions = convout1_f(img_to_visualize) File "vis_NN..py", line 483, in convout1_f return _convout1_f([0] + [X]) File "/home/chaagu/.local/lib/python3.5/site-packages/keras/backend/tensorflow_backend.py", line 2715, in __call__ return self._call(inputs) File "/home/chaagu/.local/lib/python3.5/site-packages/keras/backend/tensorflow_backend.py", line 2675, in _call fetched = self._callable_fn(*array_vals) File "/home/chaagu/.local/lib/python3.5/site-packages/tensorflow/python/client/session.py", line 1399, in __call__ run_metadata_ptr) File "/home/chaagu/.local/lib/python3.5/site-packages/tensorflow/python/framework/errors_impl.py", line 526, in __exit__ c_api.TF_GetCode(self.status.status)) tensorflow.python.framework.errors_impl.InvalidArgumentError: You must feed a value for placeholder tensor 'input_anchors' with dtype float and shape [?,?,4] [[{{node input_anchors}} = Placeholder[dtype=DT_FLOAT, shape=[?,?,4], _device="/job:localhost/replica:0/task:0/device:CPU:0"]()]]

Do you know what the reason to this error message is?
And do you know how this should be solved?