Instructions to use FinText/TimesFM_20M_2013_Global with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use FinText/TimesFM_20M_2013_Global with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, TimesFMForHF tokenizer = AutoTokenizer.from_pretrained("FinText/TimesFM_20M_2013_Global") model = TimesFMForHF.from_pretrained("FinText/TimesFM_20M_2013_Global", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download config.json from FinText/TimesFM_20M_2013_Global: direct link, hf CLI and curl.
- Browser
- Download file 326 Bytes
-
https://huggingface.co/FinText/TimesFM_20M_2013_Global/resolve/main/config.json
- Command line
-
hf download hf://FinText/TimesFM_20M_2013_Global/config.json
-
curl -L -o config.json https://huggingface.co/FinText/TimesFM_20M_2013_Global/resolve/main/config.json
326 Bytes
| { | |
| "architectures": [ | |
| "TimesFMForHF" | |
| ], | |
| "context_len": 512, | |
| "head_dim": 72, | |
| "hidden_size": 432, | |
| "horizon_len": 128, | |
| "intermediate_size": 1248, | |
| "model_type": "timesfm", | |
| "num_heads": 6, | |
| "num_kv_heads": 6, | |
| "num_layers": 9, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.47.1" | |
| } | |