Instructions to use UCSC-VLAA/openvision2-vit-large-patch14-336-vision-only with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- OpenCLIP
How to use UCSC-VLAA/openvision2-vit-large-patch14-336-vision-only with OpenCLIP:
import open_clip model, preprocess_train, preprocess_val = open_clip.create_model_and_transforms('hf-hub:UCSC-VLAA/openvision2-vit-large-patch14-336-vision-only') tokenizer = open_clip.get_tokenizer('hf-hub:UCSC-VLAA/openvision2-vit-large-patch14-336-vision-only') - Notebooks
- Google Colab
- Kaggle
File size: 464 Bytes
ef01023 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 | {
"architecture": "OpenVision2TextDecoder",
"fusion_style": "concat",
"width": 768,
"depth": 12,
"num_heads": 12,
"mlp_dim": 3072,
"vocab_size": 32000,
"vision_width": 1024,
"layer_norm_eps": 1e-06,
"pad_id": 0,
"bos_id": 1,
"eos_id": 2,
"weights_file": "caption_decoder.safetensors",
"modeling_file": "modeling_openvision2_decoder.py",
"tokenizer": "bert wordpiece (bert_base_vocab_bos_eos.txt), lowercase, [PAD]=0 [bos]=1 [eos]=2"
} |