MORPH
SIGN IN SIGN UP
huggingface / pytorch-image-models UNCLAIMED

The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

0 0 1 Python

COMMITS

main
March 23, 2026
R
Release 1.0.26
Ross Wightman committed
8d0f79e
March 13, 2026
R
Implement PRR as a pooling module. Alternative to #2678
Ross Wightman committed
6e3fdda
R
Improve 2d and latent attention pool dimension handling. Fix #2682
Ross Wightman committed
3e8def8
March 17, 2026
R
Add comments for DinoV3 re global pool (class token). Fix #2681
Ross Wightman committed
8b4239c
March 16, 2026
R
Change avg_checkpoints.py to use more secure load helper
Ross Wightman committed
52e6d19
R
Fix FX tracing on resolve_self_attn_mask
Ross Wightman committed
7a2f49b
R
Improve attention mask handling for vision_transformer and eva and related blocks
Ross Wightman committed
61a26c7
March 10, 2026
R
Update version.py
Ross Wightman committed
a94c10f
March 9, 2026
Y
fix: branch Hiera MaskUnitAttention into 4D global path for FlashAttention dispatch
Your Name committed
0c90043
R
Further refine weights_only=True, add safe globals for argparse Namespace to avoid failures on timm train checkpoints
Ross Wightman committed
a346c76
R
default weights_only=True for load fns
Ross Wightman committed
7b439f4
March 5, 2026
S
Fix CLS and Reg tokens usage when pos_embed is disabled
Sina Hajimiri committed
fa0d31e
J
Enhance SGDP optimizer with caution parameter (#2675)
Jinghui Yuan committed
85bb330
February 26, 2026
Y
Add timmx model export tool to README
Youssef Boulaoaune committed
f4f4cfc
February 25, 2026
H
fix: replace bare except clauses with except Exception
haosenwang1018 committed
2845814
February 23, 2026
R
Release 1.0.25
Ross Wightman committed
9326ff2
February 18, 2026
R
Add DTensor compatible NS impl for Muon
Ross Wightman committed
75684fa
February 15, 2026
R
Change clamp_min_ to clamp_(min=) as former doesn't work with DTensor / FSDP2 without custom registration
Ross Wightman committed
feed46c
February 11, 2026
L
fix(optim): replace bare except with Exception in Lion optimizer
Luka Aladashvili committed
56467ad
February 6, 2026
R
Fix #2661 ... don't skip reset_parameters/init when meta device detected as it breaks use of accelerate and similar dispatch override context managers
Ross Wightman committed
41754f2
R
Remove torch.jit calls except for .ignore and .is_scripting to avoid 2.11 deprecation. Fix #2663
Ross Wightman committed
bdab30b
January 29, 2026
J
Enhance the numerical stability of the Cautious Optimizer
Jinghui Yuan committed
9171d82
January 28, 2026
J
Add the test of cadamp
Jinghui Yuan committed
17b9764
J
Add cadamp optimizer to optimization factory
Jinghui Yuan committed
ba85080
J
Adding the cautious optimizer and the spherical cautious optimizer
Jinghui Yuan committed
a3643d2
January 21, 2026
R
Update README.md
Ross Wightman committed
370df82
January 20, 2026
R
Fix #2653, no models with weights impacted so just a clean fix, remove buffer as benchmarks w/ pt 2.9 show no difference
Ross Wightman committed
247c545
January 18, 2026
O
Fix distilled head dropout using wrong token in PiT forward_head
Ofer Hasson committed
836dd99
January 9, 2026
R
Remove mistaken .in_chans attr set in classifier class
Ross Wightman committed
3c6f79d
R
Refactor distillation tasks to allow creation of DistillationTecher within Task. Add TokenDistillationTask.
Ross Wightman committed
7945d3a