pytorch-image-models

pytorch-image-models copied to clipboard

[FEATURE] Save dataset_train.reader.class_to_idx inside the model and load it by default during inference and validation

2

**Is your feature request related to a problem? Please describe.** I'm not using imagenet, but during inference it loads the imagenet class_map by default. **Describe the solution you'd like** Instead,...

JustinMBrown

Resuming compiled model checkpoint

2

Sometimes im getting `Missing key(s) in state_dict: "stem.conv1.c.weight", [...] Unexpected key(s) in state_dict: "_orig_mod.stem.conv1.c.weight", [...]` while trying to resume training from checkpoint of compiled (`--torchcompile=inductor`) model. This PR solves the...

mjamroz

Add Uniformer model

1

Add Uniformer implementation from https://github.com/Sense-X/UniFormer paper: https://arxiv.org/pdf/2201.09450.pdf I converted the code to use the build_model_with_cfg( ) function

iliasprc

[FEATURE] Addition of missing typehints/docs

2

This is not exactly a feature request, I guess, but hopefully still something along those lines? **Is your feature request related to a problem? Please describe.** Multiple files such as...

a-r-r-o-w

This PR implements [Tied-Augment](https://proceedings.mlr.press/v202/kurtulus23a/kurtulus23a.pdf) for #1828.

ekurtulus

https://huggingface.co/microsoft/BiomedCLIP-PubMedBERT_256-vit_base_patch16_224 [BiomedCLIP](https://aka.ms/biomedclip-paper) is a biomedical vision-language foundation model that is pretrained on [PMC-15M](https://aka.ms/biomedclip-paper), a dataset of 15 million figure-caption pairs extracted from biomedical research articles in PubMed Central, using contrastive...

NightMachinery

**Is your feature request related to a problem? Please describe.** When I search for some model: https://huggingface.co/timm/convnext_large_mlp.clip_laion2b_augreg_ft_in1k_384, the model card said it was fine-tuned on ImageNet-1k in timm by Ross...

luckyhug

[FEATURE] Minimum required GPU RAM for different architectures

1

**Is your feature request related to a problem? Please describe.** Is the minimum required GPU memory for different architectures documented anywhere? E.g., I want to know what GPU(s) I need...

NightMachinery

[FEATURE] Support repeatability experiments

3

Thank you for your code and engineering. I use the following code to fix the seed, which can be tested to fully reproduce the comparison experiment.(Because I have found that...

wangat

[FEATURE] Weights of DINO fine-tuned on Imagenet?

5

Currently, only headless weights are available for DINO ViTs. It'd be good if the weights of a version fine-tuned for Imagenet was also added. I don't know if Facebook actually...

NightMachinery