Instructions to use TheMrguiller/Flamingo-tiny-task_BilbaoQA-ScienceQA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use TheMrguiller/Flamingo-tiny-task_BilbaoQA-ScienceQA with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("TheMrguiller/Flamingo-tiny-task_BilbaoQA-ScienceQA", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download config.json from TheMrguiller/Flamingo-tiny-task_BilbaoQA-ScienceQA: direct link, hf CLI and curl.
- Browser
- Download file 728 Bytes
-
https://huggingface.co/TheMrguiller/Flamingo-tiny-task_BilbaoQA-ScienceQA/resolve/main/config.json
- Command line
-
hf download hf://TheMrguiller/Flamingo-tiny-task_BilbaoQA-ScienceQA/config.json
-
curl -L -o config.json https://huggingface.co/TheMrguiller/Flamingo-tiny-task_BilbaoQA-ScienceQA/resolve/main/config.json
728 Bytes
| { | |
| "_name_or_path": "/home/d4k/Documents/guillermo/MUCSI_Modal/flamingo-train_task/training/flamingo-Bilbao/checkpoint-9990", | |
| "architectures": [ | |
| "FlamingoModel" | |
| ], | |
| "clip_model_type": "openai/clip-vit-large-patch14", | |
| "dim": 768, | |
| "dim_visual": 1024, | |
| "freeze_language_model": true, | |
| "freeze_vision_model": true, | |
| "lm": "facebook/opt-125m", | |
| "resampler_act": "sqrelu", | |
| "resampler_depth": 6, | |
| "resampler_dim_head": 64, | |
| "resampler_ff_mult": 4, | |
| "resampler_heads": 8, | |
| "resampler_num_latents": 64, | |
| "resampler_num_time_embeds": 4, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.29.2", | |
| "xattn_act": "sqrelu", | |
| "xattn_dim_head": 64, | |
| "xattn_every": 1, | |
| "xattn_ff_mult": 4, | |
| "xattn_heads": 8 | |
| } | |