Instructions to use hellonlp/promcse-bert-base-zh with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use hellonlp/promcse-bert-base-zh with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, BertForCL tokenizer = AutoTokenizer.from_pretrained("hellonlp/promcse-bert-base-zh") model = BertForCL.from_pretrained("hellonlp/promcse-bert-base-zh", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download trainer_state.json from hellonlp/promcse-bert-base-zh: direct link, hf CLI and curl.
- Browser
- Download file 845 Bytes
-
https://huggingface.co/hellonlp/promcse-bert-base-zh/resolve/main/trainer_state.json
- Command line
-
hf download hf://hellonlp/promcse-bert-base-zh/trainer_state.json
-
curl -L -o trainer_state.json https://huggingface.co/hellonlp/promcse-bert-base-zh/resolve/main/trainer_state.json
845 Bytes
| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.984819734345351, | |
| "global_step": 2100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.715370018975332e-05, | |
| "loss": 1.2474, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 2.4307400379506644e-05, | |
| "loss": 0.9239, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 2.146110056925996e-05, | |
| "loss": 0.8101, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 1.861480075901328e-05, | |
| "loss": 0.731, | |
| "step": 2000 | |
| } | |
| ], | |
| "max_steps": 5270, | |
| "num_train_epochs": 10, | |
| "total_flos": 126741142877995008, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |