search

fullfanta / multimodal_transfer

tensorflow implementation of 'Multimodal Transfer: A Hierarchical Deep Convolutional Neural Network for Fast Artistic Style Transfer'
34 stars 9 forks source link

Exception loading graph #1

Open alain40 opened 7 years ago

alain40 commented 7 years ago

Hi wanted to try this algorithm to see results on large images.

Running into the following exception when loading the graph:

ValueError: Tensor("Placeholder:0", shape=(1, 1200, 630, 3), dtype=float32) must be from the same graph as name: "style_subnet/Shape_1"

suke27 commented 6 years ago

i also face this problem, do you resolve it

alain40 commented 6 years ago

No I gave up.

AR

On Sun, Feb 11, 2018 at 7:26 AM, suke27 notifications@github.com wrote:

i also face this problem, do you resolve it

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/fullfanta/multimodal_transfer/issues/1#issuecomment-364726186, or mute the thread https://github.com/notifications/unsubscribe-auth/AAxd4NIljG3QWsSI23pCpx3ubXOyKUVOks5tTogcgaJpZM4OJcdu .

LouieYang commented 6 years ago

Fix it by changing tensorflow version from 1.4.1 to 1.0.1

leeqiaogithub commented 6 years ago

+1 same problem

leeqiaogithub commented 6 years ago

@alain40 I found the solution. reference: https://stackoverflow.com/questions/42616625/valueerror-tensor-must-be-from-the-same-graph-as-tensor-with-bidirectinal-rnn-i

shubhambagwari commented 2 years ago

I am unable to load the model (from train.py) in stylize.py. Can anybody help me?