Yang Li

> Awesome! Thanks for finding that. Looks quite complete, only thing I notice is that `layers.txt` and `latent_stack.txt` are the same content, so there might be a mistake on copy/paste...

I would like to know how to use this model for spatial-temporal state forecast, such as nowcasting using radar echo, like ConvLSTM.

@NarineK Thanks a lot! I tested it following the tutorial [668](https://github.com/pytorch/captum/pull/668). I need to use `nn.Upsample` or `nn.ConvTanspose2d` to upsample output to the original size, because I want to use...

Hey, @NarineK I'm new to captum and XAI. I don't clear that how to add w/Epsilon rule to `nn.ConvTranspose2d`? Is there a similar example in captum? I'd like to learn...

Hey @NarineK, I have tried add `nn.ConvTranspose2d : EpsilonRule` to `SUPPORTED_LAYERS_WITH_RULES` or `SUPPORTED_NON_LINEAR_LAYERS` into `lrp.py`, but I get `AssertionError` due to target dimension. The error information is below, ```python ---------------------------------------------------------------------------...

@NarineK I tried to instead of tensor using the codes below, but raised a new error, ```python lrp = LRP(model) attribution = lrp.attribute(vifea[0:1].cuda(), target=[(0, 0, 0)], verbose=True) attribution = attribution.squeeze().permute(1,...

@NarineK see in [colab notebook](https://colab.research.google.com/drive/1KK0bR0Q4ctYv_eIFDSZEVZrdAlUkSEfk?usp=sharing), the pretrained weights and test data is [here](https://drive.google.com/drive/folders/1g_Rltqm9UMjOyJ7EW95JvoFPcU0fUI7C?usp=sharing). Please let me know if you need more infomation.

It's a very nice feature! I can hardly wait!

@chr5tphr Thanks sooo much! I will try it!

@maxdreyer It's great! Could you have a relevant notebook or example blog? Is it possible to share with me?