Issue about AI explainability for TLT classification models

Hello.

I would like to realize the how TLT classification models make prediction by CAM or Grad-CAM

However. It seems that there is no infomation about AI explainability for TLT classification models.

How could I do to generate heatmap for TLT classification models by combining CAM or Grad-CAM ?

Thank you for your help in advance.

1 Like

TAO did not provide info for CAM or Grad-CAM. User can leverage other visualization instead. Refer to Visualizing Training - NVIDIA Docs

Excuse me @Morganh

Is it possible to obtain a class activation heatmap such as the picture below from a TAO Toolkit image classification model ?

image

Excuse me @Morganh

Is it possible to obtain a class activation heatmap such as the picture below from a TAO Toolkit image classification model ?

image

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks

TAO did not support this. User can leverage other visualization instead. Refer to Visualizing Training - NVIDIA Docs

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.