├── .gitignore ├── LICENSE ├── README.md ├── README_CN.md ├── clip ├── __init__.py ├── bert_tokenizer.py ├── configuration_bert.py ├── model.py ├── model_configs │ ├── RBT3-chinese.json │ ├── RN50.json │ ├── RoBERTa-wwm-ext-base-chinese.json │ ├── RoBERTa-wwm-ext-large-chinese.json │ ├── ViT-B-16.json │ ├── ViT-B-32.json │ ├── ViT-H-14.json │ ├── ViT-L-14-336.json │ └── ViT-L-14.json ├── modeling_bert.py ├── utils.py └── vocab.txt ├── eval ├── cvinw_zeroshot_templates.py ├── data.py ├── evaluation.py ├── evaluation_tr.py ├── extract_features.py ├── make_topk_predictions.py ├── make_topk_predictions_tr.py ├── transform_ir_annotation_to_tr.py └── zeroshot_evaluation.py ├── examples ├── chatgpt.png ├── pokemon.jpeg └── tx_video.png ├── requirements.txt └── scripts └── zeroshot_eval.sh /.gitignore: -------------------------------------------------------------------------------- 1 | __pycache__/ 2 | *.pyc 3 | .idea/ 4 | .DS_Staore -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/README.md -------------------------------------------------------------------------------- /README_CN.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/README_CN.md -------------------------------------------------------------------------------- /clip/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/__init__.py -------------------------------------------------------------------------------- /clip/bert_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/bert_tokenizer.py -------------------------------------------------------------------------------- /clip/configuration_bert.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/configuration_bert.py -------------------------------------------------------------------------------- /clip/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/model.py -------------------------------------------------------------------------------- /clip/model_configs/RBT3-chinese.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/model_configs/RBT3-chinese.json -------------------------------------------------------------------------------- /clip/model_configs/RN50.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/model_configs/RN50.json -------------------------------------------------------------------------------- /clip/model_configs/RoBERTa-wwm-ext-base-chinese.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/model_configs/RoBERTa-wwm-ext-base-chinese.json -------------------------------------------------------------------------------- /clip/model_configs/RoBERTa-wwm-ext-large-chinese.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/model_configs/RoBERTa-wwm-ext-large-chinese.json -------------------------------------------------------------------------------- /clip/model_configs/ViT-B-16.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/model_configs/ViT-B-16.json -------------------------------------------------------------------------------- /clip/model_configs/ViT-B-32.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/model_configs/ViT-B-32.json -------------------------------------------------------------------------------- /clip/model_configs/ViT-H-14.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/model_configs/ViT-H-14.json -------------------------------------------------------------------------------- /clip/model_configs/ViT-L-14-336.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/model_configs/ViT-L-14-336.json -------------------------------------------------------------------------------- /clip/model_configs/ViT-L-14.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/model_configs/ViT-L-14.json -------------------------------------------------------------------------------- /clip/modeling_bert.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/modeling_bert.py -------------------------------------------------------------------------------- /clip/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/utils.py -------------------------------------------------------------------------------- /clip/vocab.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/clip/vocab.txt -------------------------------------------------------------------------------- /eval/cvinw_zeroshot_templates.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/eval/cvinw_zeroshot_templates.py -------------------------------------------------------------------------------- /eval/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/eval/data.py -------------------------------------------------------------------------------- /eval/evaluation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/eval/evaluation.py -------------------------------------------------------------------------------- /eval/evaluation_tr.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/eval/evaluation_tr.py -------------------------------------------------------------------------------- /eval/extract_features.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/eval/extract_features.py -------------------------------------------------------------------------------- /eval/make_topk_predictions.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/eval/make_topk_predictions.py -------------------------------------------------------------------------------- /eval/make_topk_predictions_tr.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/eval/make_topk_predictions_tr.py -------------------------------------------------------------------------------- /eval/transform_ir_annotation_to_tr.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/eval/transform_ir_annotation_to_tr.py -------------------------------------------------------------------------------- /eval/zeroshot_evaluation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/eval/zeroshot_evaluation.py -------------------------------------------------------------------------------- /examples/chatgpt.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/examples/chatgpt.png -------------------------------------------------------------------------------- /examples/pokemon.jpeg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/examples/pokemon.jpeg -------------------------------------------------------------------------------- /examples/tx_video.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/examples/tx_video.png -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/requirements.txt -------------------------------------------------------------------------------- /scripts/zeroshot_eval.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TencentARC-QQ/QA-CLIP/HEAD/scripts/zeroshot_eval.sh --------------------------------------------------------------------------------