feat: add WinML hub catalog and per-model precision configs#431
Closed
ssss141414 wants to merge 1 commit into
Closed
feat: add WinML hub catalog and per-model precision configs#431ssss141414 wants to merge 1 commit into
ssss141414 wants to merge 1 commit into
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
This PR introduces a set of WinML-oriented model configuration files intended to support exporting and (optionally) quantizing a curated set of Hugging Face models with per-model precision variants (e.g., FP16, W8A8, W8A16).
Changes:
- Add per-model WinML export configurations with fixed input/output tensor specs for a variety of NLP and CV pipeline tasks.
- Add quantization configurations for W8A8 and W8A16 variants where applicable (QDQ mode, calibration settings, and model/task metadata).
- Add loader metadata (task, model class, model type) to drive model loading for each configuration.
Reviewed changes
Copilot reviewed 193 out of 193 changed files in this pull request and generated no comments.
Show a summary per file
| File | Description |
|---|---|
| .aitk/winml/w11wo_indonesian-roberta-base-posp-tagger/token-classification/w8a8_config.json | Add WinML export+W8A8 quant config for token classification. |
| .aitk/winml/w11wo_indonesian-roberta-base-posp-tagger/token-classification/w8a16_config.json | Add WinML export+W8A16 quant config for token classification. |
| .aitk/winml/w11wo_indonesian-roberta-base-posp-tagger/token-classification/fp16_config.json | Add WinML export FP16 config for token classification. |
| .aitk/winml/StanfordAIMI_dinov2-base-xray-224/image-feature-extraction/w8a8_config.json | Add WinML export+W8A8 quant config for image feature extraction. |
| .aitk/winml/StanfordAIMI_dinov2-base-xray-224/image-feature-extraction/w8a16_config.json | Add WinML export+W8A16 quant config for image feature extraction. |
| .aitk/winml/StanfordAIMI_dinov2-base-xray-224/image-feature-extraction/fp16_config.json | Add WinML export FP16 config for image feature extraction. |
| .aitk/winml/sentence-transformers_paraphrase-multilingual-mpnet-base-v2/sentence-similarity/w8a8_config.json | Add WinML export+W8A8 quant config for sentence similarity. |
| .aitk/winml/sentence-transformers_paraphrase-multilingual-mpnet-base-v2/sentence-similarity/w8a16_config.json | Add WinML export+W8A16 quant config for sentence similarity. |
| .aitk/winml/sentence-transformers_paraphrase-multilingual-mpnet-base-v2/sentence-similarity/fp16_config.json | Add WinML export FP16 config for sentence similarity. |
| .aitk/winml/sentence-transformers_paraphrase-multilingual-MiniLM-L12-v2/sentence-similarity/fp16_config.json | Add WinML export FP16 config for sentence similarity. |
| .aitk/winml/sentence-transformers_paraphrase-multilingual-MiniLM-L12-v2/feature-extraction/fp16_config.json | Add WinML export FP16 config for feature extraction. |
| .aitk/winml/sentence-transformers_all-MiniLM-L6-v2/sentence-similarity/fp16_config.json | Add WinML export FP16 config for sentence similarity. |
| .aitk/winml/sentence-transformers_all-MiniLM-L6-v2/feature-extraction/fp16_config.json | Add WinML export FP16 config for feature extraction. |
| .aitk/winml/rizvandwiki_gender-classification/image-classification/w8a8_config.json | Add WinML export+W8A8 quant config for image classification. |
| .aitk/winml/rizvandwiki_gender-classification/image-classification/w8a16_config.json | Add WinML export+W8A16 quant config for image classification. |
| .aitk/winml/rizvandwiki_gender-classification/image-classification/fp16_config.json | Add WinML export FP16 config for image classification. |
| .aitk/winml/ProsusAI_finbert/text-classification/fp16_config.json | Add WinML export FP16 config for text classification. |
| .aitk/winml/patrickjohncyh_fashion-clip/zero-shot-image-classification/w8a8_config_image-encoder.json | Add WinML export+W8A8 quant config for CLIP image encoder. |
| .aitk/winml/patrickjohncyh_fashion-clip/zero-shot-image-classification/w8a16_config_image-encoder.json | Add WinML export+W8A16 quant config for CLIP image encoder. |
| .aitk/winml/patrickjohncyh_fashion-clip/zero-shot-image-classification/fp16_config_text-encoder.json | Add WinML export FP16 config for CLIP text encoder. |
| .aitk/winml/patrickjohncyh_fashion-clip/zero-shot-image-classification/fp16_config_image-encoder.json | Add WinML export FP16 config for CLIP image encoder. |
| .aitk/winml/openai_clip-vit-large-patch14/zero-shot-image-classification/w8a8_config_image-encoder.json | Add WinML export+W8A8 quant config for CLIP image encoder. |
| .aitk/winml/openai_clip-vit-large-patch14/zero-shot-image-classification/w8a16_config_image-encoder.json | Add WinML export+W8A16 quant config for CLIP image encoder. |
| .aitk/winml/openai_clip-vit-large-patch14/zero-shot-image-classification/fp16_config_text-encoder.json | Add WinML export FP16 config for CLIP text encoder. |
| .aitk/winml/openai_clip-vit-large-patch14/zero-shot-image-classification/fp16_config_image-encoder.json | Add WinML export FP16 config for CLIP image encoder. |
| .aitk/winml/openai_clip-vit-large-patch14-336/zero-shot-image-classification/w8a8_config_image-encoder.json | Add WinML export+W8A8 quant config for CLIP image encoder (336). |
| .aitk/winml/openai_clip-vit-large-patch14-336/zero-shot-image-classification/w8a16_config_image-encoder.json | Add WinML export+W8A16 quant config for CLIP image encoder (336). |
| .aitk/winml/openai_clip-vit-large-patch14-336/zero-shot-image-classification/fp16_config_text-encoder.json | Add WinML export FP16 config for CLIP text encoder (336). |
| .aitk/winml/openai_clip-vit-large-patch14-336/zero-shot-image-classification/fp16_config_image-encoder.json | Add WinML export FP16 config for CLIP image encoder (336). |
| .aitk/winml/openai_clip-vit-base-patch32/zero-shot-image-classification/w8a8_config_image-encoder.json | Add WinML export+W8A8 quant config for CLIP image encoder. |
| .aitk/winml/openai_clip-vit-base-patch32/zero-shot-image-classification/w8a16_config_image-encoder.json | Add WinML export+W8A16 quant config for CLIP image encoder. |
| .aitk/winml/openai_clip-vit-base-patch32/zero-shot-image-classification/fp16_config_text-encoder.json | Add WinML export FP16 config for CLIP text encoder. |
| .aitk/winml/openai_clip-vit-base-patch32/zero-shot-image-classification/fp16_config_image-encoder.json | Add WinML export FP16 config for CLIP image encoder. |
| .aitk/winml/openai_clip-vit-base-patch32/feature-extraction/w8a8_config.json | Add WinML export+W8A8 quant config for CLIP text feature extraction. |
| .aitk/winml/openai_clip-vit-base-patch32/feature-extraction/w8a16_config.json | Add WinML export+W8A16 quant config for CLIP text feature extraction. |
| .aitk/winml/openai_clip-vit-base-patch32/feature-extraction/fp16_config.json | Add WinML export FP16 config for CLIP text feature extraction. |
| .aitk/winml/openai_clip-vit-base-patch16/zero-shot-image-classification/w8a8_config_image-encoder.json | Add WinML export+W8A8 quant config for CLIP image encoder. |
| .aitk/winml/openai_clip-vit-base-patch16/zero-shot-image-classification/w8a16_config_image-encoder.json | Add WinML export+W8A16 quant config for CLIP image encoder. |
| .aitk/winml/openai_clip-vit-base-patch16/zero-shot-image-classification/fp16_config_text-encoder.json | Add WinML export FP16 config for CLIP text encoder. |
| .aitk/winml/openai_clip-vit-base-patch16/zero-shot-image-classification/fp16_config_image-encoder.json | Add WinML export FP16 config for CLIP image encoder. |
| .aitk/winml/openai_clip-vit-base-patch16/feature-extraction/w8a8_config.json | Add WinML export+W8A8 quant config for CLIP text feature extraction. |
| .aitk/winml/openai_clip-vit-base-patch16/feature-extraction/fp16_config.json | Add WinML export FP16 config for CLIP text feature extraction. |
| .aitk/winml/nvidia_segformer-b5-finetuned-ade-640-640/image-segmentation/w8a8_config.json | Add WinML export+W8A8 quant config for image segmentation. |
| .aitk/winml/nvidia_segformer-b5-finetuned-ade-640-640/image-segmentation/w8a16_config.json | Add WinML export+W8A16 quant config for image segmentation. |
| .aitk/winml/nvidia_segformer-b5-finetuned-ade-640-640/image-segmentation/fp16_config.json | Add WinML export FP16 config for image segmentation. |
| .aitk/winml/nvidia_segformer-b2-finetuned-ade-512-512/image-segmentation/w8a8_config.json | Add WinML export+W8A8 quant config for image segmentation. |
| .aitk/winml/nvidia_segformer-b2-finetuned-ade-512-512/image-segmentation/w8a16_config.json | Add WinML export+W8A16 quant config for image segmentation. |
| .aitk/winml/nvidia_segformer-b2-finetuned-ade-512-512/image-segmentation/fp16_config.json | Add WinML export FP16 config for image segmentation. |
| .aitk/winml/nvidia_segformer-b1-finetuned-ade-512-512/image-segmentation/w8a8_config.json | Add WinML export+W8A8 quant config for image segmentation. |
| .aitk/winml/nvidia_segformer-b1-finetuned-ade-512-512/image-segmentation/w8a16_config.json | Add WinML export+W8A16 quant config for image segmentation. |
| .aitk/winml/nvidia_segformer-b1-finetuned-ade-512-512/image-segmentation/fp16_config.json | Add WinML export FP16 config for image segmentation. |
| .aitk/winml/microsoft_table-transformer-detection/object-detection/w8a8_config.json | Add WinML export+W8A8 quant config for object detection. |
| .aitk/winml/microsoft_table-transformer-detection/object-detection/w8a16_config.json | Add WinML export+W8A16 quant config for object detection. |
| .aitk/winml/microsoft_table-transformer-detection/object-detection/fp16_config.json | Add WinML export FP16 config for object detection. |
| .aitk/winml/microsoft_swin-large-patch4-window7-224/image-classification/w8a8_config.json | Add WinML export+W8A8 quant config for image classification. |
| .aitk/winml/microsoft_swin-large-patch4-window7-224/image-classification/w8a16_config.json | Add WinML export+W8A16 quant config for image classification. |
| .aitk/winml/microsoft_swin-large-patch4-window7-224/image-classification/fp16_config.json | Add WinML export FP16 config for image classification. |
| .aitk/winml/microsoft_resnet-50/image-classification/w8a8_config.json | Add WinML export+W8A8 quant config for image classification. |
| .aitk/winml/microsoft_resnet-50/image-classification/w8a16_config.json | Add WinML export+W8A16 quant config for image classification. |
| .aitk/winml/microsoft_resnet-50/image-classification/fp16_config.json | Add WinML export FP16 config for image classification. |
| .aitk/winml/microsoft_rad-dino/image-feature-extraction/w8a8_config.json | Add WinML export+W8A8 quant config for image feature extraction. |
| .aitk/winml/microsoft_rad-dino/image-feature-extraction/w8a16_config.json | Add WinML export+W8A16 quant config for image feature extraction. |
| .aitk/winml/microsoft_rad-dino/image-feature-extraction/fp16_config.json | Add WinML export FP16 config for image feature extraction. |
| .aitk/winml/mattmdjaga_segformer_b2_clothes/image-segmentation/w8a8_config.json | Add WinML export+W8A8 quant config for image segmentation. |
| .aitk/winml/mattmdjaga_segformer_b2_clothes/image-segmentation/w8a16_config.json | Add WinML export+W8A16 quant config for image segmentation. |
| .aitk/winml/mattmdjaga_segformer_b2_clothes/image-segmentation/fp16_config.json | Add WinML export FP16 config for image segmentation. |
| .aitk/winml/laion_CLIP-ViT-H-14-laion2B-s32B-b79K/zero-shot-image-classification/fp16_config_text-encoder.json | Add WinML export FP16 config for CLIP text encoder. |
| .aitk/winml/laion_CLIP-ViT-H-14-laion2B-s32B-b79K/zero-shot-image-classification/fp16_config_image-encoder.json | Add WinML export FP16 config for CLIP image encoder. |
| .aitk/winml/laion_CLIP-ViT-B-32-laion2B-s34B-b79K/zero-shot-image-classification/fp16_config_text-encoder.json | Add WinML export FP16 config for CLIP text encoder. |
| .aitk/winml/laion_CLIP-ViT-B-32-laion2B-s34B-b79K/zero-shot-image-classification/fp16_config_image-encoder.json | Add WinML export FP16 config for CLIP image encoder. |
| .aitk/winml/laion_CLIP-ViT-B-32-laion2B-s34B-b79K/feature-extraction/fp16_config.json | Add WinML export FP16 config for CLIP text feature extraction. |
| .aitk/winml/Intel_bert-base-uncased-mrpc/text-classification/fp16_config.json | Add WinML export FP16 config for text classification. |
| .aitk/winml/Intel_bert-base-uncased-mrpc/feature-extraction/fp16_config.json | Add WinML export FP16 config for feature extraction. |
| .aitk/winml/google-bert_bert-large-uncased-whole-word-masking-finetuned-squad/question-answering/fp16_config.json | Add WinML export FP16 config for question answering. |
| .aitk/winml/google-bert_bert-base-uncased/fill-mask/fp16_config.json | Add WinML export FP16 config for fill-mask. |
| .aitk/winml/google-bert_bert-base-multilingual-uncased/fill-mask/fp16_config.json | Add WinML export FP16 config for fill-mask. |
| .aitk/winml/google-bert_bert-base-multilingual-cased/feature-extraction/fp16_config.json | Add WinML export FP16 config for feature extraction. |
| .aitk/winml/google_vit-base-patch16-224/image-classification/w8a8_config.json | Add WinML export+W8A8 quant config for image classification. |
| .aitk/winml/google_vit-base-patch16-224/image-classification/w8a16_config.json | Add WinML export+W8A16 quant config for image classification. |
| .aitk/winml/google_vit-base-patch16-224/image-classification/fp16_config.json | Add WinML export FP16 config for image classification. |
| .aitk/winml/google_vit-base-patch16-224-in21k/image-feature-extraction/w8a8_config.json | Add WinML export+W8A8 quant config for image feature extraction. |
| .aitk/winml/google_vit-base-patch16-224-in21k/image-feature-extraction/w8a16_config.json | Add WinML export+W8A16 quant config for image feature extraction. |
| .aitk/winml/google_vit-base-patch16-224-in21k/image-feature-extraction/fp16_config.json | Add WinML export FP16 config for image feature extraction. |
| .aitk/winml/FacebookAI_xlm-roberta-large/fill-mask/w8a8_config.json | Add WinML export+W8A8 quant config for fill-mask. |
| .aitk/winml/FacebookAI_xlm-roberta-large/fill-mask/w8a16_config.json | Add WinML export+W8A16 quant config for fill-mask. |
| .aitk/winml/FacebookAI_xlm-roberta-large/fill-mask/fp16_config.json | Add WinML export FP16 config for fill-mask. |
| .aitk/winml/FacebookAI_xlm-roberta-base/fill-mask/w8a8_config.json | Add WinML export+W8A8 quant config for fill-mask. |
| .aitk/winml/FacebookAI_xlm-roberta-base/fill-mask/w8a16_config.json | Add WinML export+W8A16 quant config for fill-mask. |
| .aitk/winml/FacebookAI_xlm-roberta-base/fill-mask/fp16_config.json | Add WinML export FP16 config for fill-mask. |
| .aitk/winml/FacebookAI_roberta-large/fill-mask/w8a8_config.json | Add WinML export+W8A8 quant config for fill-mask. |
| .aitk/winml/FacebookAI_roberta-large/fill-mask/w8a16_config.json | Add WinML export+W8A16 quant config for fill-mask. |
| .aitk/winml/FacebookAI_roberta-large/fill-mask/fp16_config.json | Add WinML export FP16 config for fill-mask. |
| .aitk/winml/FacebookAI_roberta-base/fill-mask/w8a8_config.json | Add WinML export+W8A8 quant config for fill-mask. |
| .aitk/winml/FacebookAI_roberta-base/fill-mask/w8a16_config.json | Add WinML export+W8A16 quant config for fill-mask. |
| .aitk/winml/FacebookAI_roberta-base/fill-mask/fp16_config.json | Add WinML export FP16 config for fill-mask. |
| .aitk/winml/facebook_dinov2-small/image-feature-extraction/w8a8_config.json | Add WinML export+W8A8 quant config for image feature extraction. |
| .aitk/winml/facebook_dinov2-small/image-feature-extraction/w8a16_config.json | Add WinML export+W8A16 quant config for image feature extraction. |
| .aitk/winml/facebook_dinov2-small/image-feature-extraction/fp16_config.json | Add WinML export FP16 config for image feature extraction. |
| .aitk/winml/facebook_dinov2-large/image-feature-extraction/w8a8_config.json | Add WinML export+W8A8 quant config for image feature extraction. |
| .aitk/winml/facebook_dinov2-large/image-feature-extraction/w8a16_config.json | Add WinML export+W8A16 quant config for image feature extraction. |
| .aitk/winml/facebook_dinov2-large/image-feature-extraction/fp16_config.json | Add WinML export FP16 config for image feature extraction. |
| .aitk/winml/facebook_dinov2-base/image-feature-extraction/w8a8_config.json | Add WinML export+W8A8 quant config for image feature extraction. |
| .aitk/winml/facebook_dinov2-base/image-feature-extraction/w8a16_config.json | Add WinML export+W8A16 quant config for image feature extraction. |
| .aitk/winml/facebook_dinov2-base/image-feature-extraction/fp16_config.json | Add WinML export FP16 config for image feature extraction. |
| .aitk/winml/facebook_dino-vits16/image-feature-extraction/w8a8_config.json | Add WinML export+W8A8 quant config for image feature extraction. |
| .aitk/winml/facebook_dino-vits16/image-feature-extraction/w8a16_config.json | Add WinML export+W8A16 quant config for image feature extraction. |
| .aitk/winml/facebook_dino-vits16/image-feature-extraction/fp16_config.json | Add WinML export FP16 config for image feature extraction. |
| .aitk/winml/facebook_dino-vitb16/image-feature-extraction/w8a8_config.json | Add WinML export+W8A8 quant config for image feature extraction. |
| .aitk/winml/facebook_dino-vitb16/image-feature-extraction/w8a16_config.json | Add WinML export+W8A16 quant config for image feature extraction. |
| .aitk/winml/facebook_dino-vitb16/image-feature-extraction/fp16_config.json | Add WinML export FP16 config for image feature extraction. |
| .aitk/winml/facebook_convnext-tiny-224/image-classification/w8a8_config.json | Add WinML export+W8A8 quant config for image classification. |
| .aitk/winml/facebook_convnext-tiny-224/image-classification/w8a16_config.json | Add WinML export+W8A16 quant config for image classification. |
| .aitk/winml/facebook_convnext-tiny-224/image-classification/fp16_config.json | Add WinML export FP16 config for image classification. |
| .aitk/winml/dslim_bert-base-NER/token-classification/fp16_config.json | Add WinML export FP16 config for token classification. |
| .aitk/winml/deepset_tinyroberta-squad2/question-answering/w8a8_config.json | Add WinML export+W8A8 quant config for question answering. |
| .aitk/winml/deepset_tinyroberta-squad2/question-answering/w8a16_config.json | Add WinML export+W8A16 quant config for question answering. |
| .aitk/winml/deepset_tinyroberta-squad2/question-answering/fp16_config.json | Add WinML export FP16 config for question answering. |
| .aitk/winml/deepset_roberta-base-squad2/question-answering/w8a8_config.json | Add WinML export+W8A8 quant config for question answering. |
| .aitk/winml/deepset_roberta-base-squad2/question-answering/w8a16_config.json | Add WinML export+W8A16 quant config for question answering. |
| .aitk/winml/deepset_roberta-base-squad2/question-answering/fp16_config.json | Add WinML export FP16 config for question answering. |
| .aitk/winml/deepset_bert-large-uncased-whole-word-masking-squad2/question-answering/fp16_config.json | Add WinML export FP16 config for question answering. |
| .aitk/winml/dbmdz_bert-large-cased-finetuned-conll03-english/token-classification/fp16_config.json | Add WinML export FP16 config for token classification. |
| .aitk/winml/cardiffnlp_twitter-roberta-base-sentiment-latest/text-classification/w8a8_config.json | Add WinML export+W8A8 quant config for text classification. |
| .aitk/winml/cardiffnlp_twitter-roberta-base-sentiment-latest/text-classification/w8a16_config.json | Add WinML export+W8A16 quant config for text classification. |
| .aitk/winml/cardiffnlp_twitter-roberta-base-sentiment-latest/text-classification/fp16_config.json | Add WinML export FP16 config for text classification. |
| .aitk/winml/Babelscape_wikineural-multilingual-ner/token-classification/fp16_config.json | Add WinML export FP16 config for token classification. |
| .aitk/winml/BAAI_bge-small-en-v1.5/sentence-similarity/fp16_config.json | Add WinML export FP16 config for sentence similarity. |
| .aitk/winml/BAAI_bge-small-en-v1.5/feature-extraction/fp16_config.json | Add WinML export FP16 config for feature extraction. |
| .aitk/winml/BAAI_bge-large-en-v1.5/sentence-similarity/fp16_config.json | Add WinML export FP16 config for sentence similarity. |
| .aitk/winml/BAAI_bge-base-en-v1.5/sentence-similarity/fp16_config.json | Add WinML export FP16 config for sentence similarity. |
| .aitk/winml/BAAI_bge-base-en-v1.5/feature-extraction/fp16_config.json | Add WinML export FP16 config for feature extraction. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.