ceykmc / pytorch_model_summary Goto Github PK

View Code? Open in Web Editor NEW

94.0 94.0 14.0 33 KB

pytorch model summary, statistic parameters number, memory usage, FLOPs and so on

Python 100.00%

pytorch_model_summary's People

Contributors

Stargazers

Watchers

Forkers

wjgaas shlpu h-jia lijiunderstand frizy-up pplntech guker aidasliaudanskas chenxi-yang jianweilin baiti01 nguyenduykhanh240595 ahangchen

pytorch_model_summary's Issues

'tuple' object has no attribute 'size' for RNN module

Does model summary support RNN analysis?

Thanks for sharing~

How to compute the madd of layers with two inputs (such as Bilinear)

Your tool is so useful!
However, I found that your tool doesn't support the madd of layers with two inputs (such as Bilinear). I tried to add a compute_Bilinear_madd func, but I am confused about the input. It seems you use input[0] as the input of a layer (madd = compute_module_madd(module, input[0], output)). How can I achieve this function with min effort？

What does MAdd mean exactly?

1 Madd is 2 Flops? and 1 Madd is 1 MAc?

Is this right?

Thanks.

不支持inception v3吗

closed

RT 谢谢分享

Definition of Inference Memory

Hi,
Thanks for this repository.
I wanted to understand what you mean by inference memory : is it the memory size of the model, or does it refer to amount of memory the model consumes (RAM) during inference.