torchview
torchview copied to clipboard
mert-kurttutan
torchview: visualize pytorch models
The [`torchstat`](https://github.com/Swall0w/torchstat) package has functions to compute the number of parameters and FLOPs for torch modules. I think it should be easy to use that to add an option to...
Update 'requirements-dev.txt' for correct syntax
**Is your feature request related to a problem? Please describe.** Deep Learning models usually are stacked layers, from shallow layers to deep layers. With Transform models being prevalent, it is...
**Bug** ``` ValueError: `.to` is not supported for `4-bit` or `8-bit` bitsandbytes models. Please use the model as it is, since the model has already been set to the correct...
**Is your feature request related to a problem? Please describe.** If saving to disk, the default output format is PNG which is inefficient and low-resolution. I couldn't find a way...
**Error Message** RuntimeError: Can not infer total number of classes from meta tensor. **Cause** `model_graph = draw_graph(model, input_data=input_tensor, device='meta')` And my model has a layer `c = nn.functional.one_hot(c, num_classes=self.n_classes).type(torch.float)` **Solution**...
how to use visualize custom vision models like unet using torchview, when I try to visualize my model, it throws the following error: . . . . . RuntimeError: Failed...
I have installed torchview on Windows using `conda install`. When diaplying a simple graph, I'm getting the following warning: (process:20416): Pango-WARNING **: 09:23:59.938: couldn't load font "Linux libertine Not-Rotated 10",...
Currently if the input to the model forward pass contains an ndarray it is treated as an iterable causing the error: ``` File ~/code/torchview/torchview/torchview.py:372, in traverse_data(data, action_fn, aggregate_fn) 368 return...
**Describe the bug** Currently if torchview detects gets a `list` or `dict` as the input to `forward` it assumes the user will have a forward function in their model that...