search

raghakot / keras-vis

Neural network visualization toolkit for keras
https://raghakot.github.io/keras-vis
MIT License
2.97k stars 664 forks source link

Fix #144 #149

Closed keisen closed 5 years ago

keisen commented 5 years ago

Problem

In full-convolutional-network model such as InceptionV3 case, it might not be understood input/output data shape of each layers before starting train/predict. That is model definition is construction like (None, None, None, channels) . So it might not be able to correctly work visualize_cam .

For details see #144 .

Improvement

It get tensor's shape from actual input/output data, not from model definition.

keisen commented 5 years ago

@raghakot , please review this PR. Thanks!

keisen commented 5 years ago

Although I'll merge this PR because this PR have passed about a half year, please revert it if you find any problem.

Thanks!