search

sicara / tf-explain

Interpretability Methods for tf.keras models with Tensorflow 2.x
https://tf-explain.readthedocs.io
MIT License
1.02k stars 112 forks source link

TF-Explain support for multi input model which take image and text as input for classification task #154

Closed harsha89 closed 3 years ago

harsha89 commented 3 years ago

Hi,

Thank you for the awesome work on the module.

I have a requirement of visualizing grad cam for multi input model which take image and text as input. This grad cam is specific to the image input of the multi modal tensorflow network architecure.

Since two models concatenate at the end, how should I select the last layer or how gradients can flow through concatenation layer? Any input is highly welcome.

Thanks, Harsha

harsha89 commented 3 years ago

I have found the solution. Hence closing the issue.