Tyler Yep
Tyler Yep
Functional layers cannot be detected by torchsummary. If this is a heavily requested feature, you can post it as an issue on my fork - see #124
Should all be fixed in #124
Available in my fork - See #124 It's available via the ModelStatistics object, which can be printed via str(model_stats)
See #124
I still had some issues even after trying `set -Ua fish_user_paths $HOME/.cargo/bin`, so instead I added the line `source $HOME/.cargo/env` to my `config.fish` file, which worked!
The error message is saying to call your model, you pass in an input tensor and an img_meta param. you should pass that to summary() as a kwarg. I would...
Closing due to inactivity. Feel free to reopen if you have followup questions.
Hi, I'd also like to +1 this suggestion. This warning is a false positive because a class attribute can never conflict with the builtin. For example, my custom LinkedList class...
I have the same issue. Does anyone know how to get this working?
This issue should be resolved in #181, but I'll wait for the release and a verification before closing