torchinfo
torchinfo copied to clipboard
View model summaries in PyTorch!
Option for column representing Percentage FLOPS or Multiply Adds. (similar to #199) I think this option would be really useful to see which part of model should be optimized if...
**Describe the bug** estimate model size is different with nvidia-smi usage **To Reproduce** 1. used code, and command line 2. The code will run on the cuda:2 device ``` import...
updates: - [github.com/astral-sh/ruff-pre-commit: v0.4.4 → v0.6.2](https://github.com/astral-sh/ruff-pre-commit/compare/v0.4.4...v0.6.2)
**Describe the bug** my code: ``` import torch from PIL import Image from thop import profile, clever_format from torchinfo import summary import cn_clip.clip as clip from cn_clip.clip import load_from_name, available_models...
**Version** torchinfo==1.8.0 **Above all** Empty-Vectors mean: ```python ev = torch.rand([]) ``` which: ``` ev.shape=torch.Size([]) type(ev)=torch.Tensor ``` **To Reproduce** ```python from diffusers import StableDiffusionPipeline import torch from torchinfo import summary pipe...
Hi, I came across the notion of [static/dynamic flow](https://pytorch.org/docs/stable/fx.html#limitations-of-symbolic-tracing) on PyTorch's doc. I realize that dynamic flow (that is when module calls may not be consistent between different inputs) poses...
Hello, I managed without problem to run `torchinfo.summary` on `YOLOv5n` that I loaded from the torch hub. ```pycon >>> torchinfo.summary(torch.hub.load('ultralytics/yolov5', 'yolov5n'), input_size=(1, 3, 640, 640), depth=4) Using cache found in...
Fixes #319.
**Is your feature request related to a problem? Please describe.** torch.nn.Conv2d has a "groups" attribute set to a value greater than 1 for EfficientNet, ResNeXt, ConvNeXt, and other current vision...