Instructions to use HopitAI/moda-pro-lite-plus with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- OpenCLIP
How to use HopitAI/moda-pro-lite-plus with OpenCLIP:
import open_clip model, preprocess_train, preprocess_val = open_clip.create_model_and_transforms('hf-hub:HopitAI/moda-pro-lite-plus') tokenizer = open_clip.get_tokenizer('hf-hub:HopitAI/moda-pro-lite-plus') - Notebooks
- Google Colab
- Kaggle
Download config.json from HopitAI/moda-pro-lite-plus: direct link, hf CLI and curl.
- Browser
- Download file 950 Bytes
-
https://huggingface.co/HopitAI/moda-pro-lite-plus/resolve/main/config.json
- Command line
-
hf download hf://HopitAI/moda-pro-lite-plus/config.json
-
curl -L -o config.json https://huggingface.co/HopitAI/moda-pro-lite-plus/resolve/main/config.json
950 Bytes
| { | |
| "name": "MODA Pro Lite+", | |
| "artifact_type": "retrieval_system", | |
| "base_model": "HopitAI/moda-pro-lite", | |
| "base_model_parameters": 213159938, | |
| "additional_learned_parameters": 0, | |
| "embedding_dimension": 768, | |
| "gallery": { | |
| "views": { | |
| "official": 1.0, | |
| "pad": 0.25, | |
| "foreground_pad": 0.25 | |
| }, | |
| "normalize": true, | |
| "stored_vectors_per_item": 1 | |
| }, | |
| "query": { | |
| "prompts": { | |
| "raw": 1.0, | |
| "photo": 0.25 | |
| }, | |
| "normalize": true | |
| }, | |
| "prompt_templates": { | |
| "raw": "{query}", | |
| "photo": "a photo of {query}" | |
| }, | |
| "serving": { | |
| "stored_vectors_per_item": 1, | |
| "ann_queries_per_search": 1, | |
| "image_forwards_per_item_at_index_time": 3, | |
| "text_forwards_per_query": 2, | |
| "rerank": false | |
| }, | |
| "evaluation": { | |
| "protocol": "full corpus, all ground-truth queries, MAP@10, pytrec_eval map_cut.10, paired bootstrap 10000", | |
| "receipt": "results/multiview_target_eval_fp32/" | |
| } | |
| } |