How to extract saliency map from the ImageNet pretrained with non-RBG input?

qubvel / segmentation_models

Segmentation models with pretrained backbones. Keras and TensorFlow Keras.

MIT License

4.66k stars 1.03k forks source link

How to extract saliency map from the ImageNet pretrained with non-RBG input? #564

Open N-Jaro opened 1 year ago

N-Jaro commented 1 year ago

I created a model with the non-RGB input, trained it, and saved it in h5 format. I want to get the activation and saliency maps when I predict an input image.

Could anyone please advise how I should proceed with this?

Thanks