timm 0.9.16

PyTorch Image Models

06-18-2023 219 words 2 minutes 0 views

Contents

PyTorch Image Models

Stars: 29537, Watchers: 29537, Forks: 4561, Open Issues: 105

The huggingface/pytorch-image-models repo was created 5 years ago and the last code push was 15 hours ago.
The project is extremely popular with a mindblowing 29537 github stars!

How to Install timm

You can install timm using pip

pip install timm

or add it to a project with poetry

poetry add timm

Package Details

Author
License: Apache-2.0
Homepage: https://github.com/huggingface/pytorch-image-models
PyPi:: https://pypi.org/project/timm/
Documentation:: https://huggingface.co/docs/timm/en/index
GitHub Repo:: https://github.com/huggingface/pytorch-image-models

Classifiers

Scientific/Engineering
Scientific/Engineering/Artificial Intelligence
Software Development
Software Development/Libraries
Software Development/Libraries/Python Modules

No timm pypi packages just yet.

Errors

A list of common timm errors.

Code Examples

Here are some timm code examples and snippets.

GitHub Issues

The timm package has 105 open issues on GitHub

Omit Normalization Layer from MLP Class
[BUG] Normalization Layer Not Used In MLP Class
Support other features only modes for EfficientNet
[BUG] Some models do not drop unnecessary stages when specifying out_indices
[BUG] Inconsistent behavior with negative out_indices
Use of Nvidia Transformer Engine
[FEATURE] Support FasterViT
[FEATURE] Effective drop path.
Add drop path schedule
[FEATURE] Update weights for wide_resnet101_2
Test Python 3.11 support
[FEATURE] Support Tied-Augment
[FEATURE] Torchserve documentation/examples
[FEATURE] timm.list_models(features_only=True)
[FEATURE] Save dataset_train.reader.class_to_idx inside the model and load it by default during inference and validation