Instructions to use merit/Model_Durian with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use merit/Model_Durian with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("merit/Model_Durian") model = AutoModelForSeq2SeqLM.from_pretrained("merit/Model_Durian", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download trainer_state.json from merit/Model_Durian: direct link, hf CLI and curl.
- Browser
- Download file 2.13 kB
-
https://huggingface.co/merit/Model_Durian/resolve/main/trainer_state.json
- Command line
-
hf download hf://merit/Model_Durian/trainer_state.json
-
curl -L -o trainer_state.json https://huggingface.co/merit/Model_Durian/resolve/main/trainer_state.json
2.13 kB
| { | |
| "best_metric": 1.2918661832809448, | |
| "epoch": 5.0, | |
| "global_step": 6230, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 9.866238630283575e-06, | |
| "loss": 0.8765, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 9.73247726056715e-06, | |
| "loss": 0.8865, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 9.598715890850723e-06, | |
| "loss": 0.8513, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 9.464954521134298e-06, | |
| "loss": 0.8626, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 9.331193151417872e-06, | |
| "loss": 0.8445, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 9.197431781701446e-06, | |
| "loss": 0.8685, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 9.06367041198502e-06, | |
| "loss": 0.8664, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 8.929909042268593e-06, | |
| "loss": 0.8514, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 8.796147672552167e-06, | |
| "loss": 0.8476, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 8.662386302835741e-06, | |
| "loss": 0.8252, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 8.528624933119315e-06, | |
| "loss": 0.8658, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 8.39486356340289e-06, | |
| "loss": 0.8453, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 18.796, | |
| "eval_loss": 1.2918661832809448, | |
| "eval_rouge1": 39.8436, | |
| "eval_rouge2": 15.5632, | |
| "eval_rougeL": 33.522, | |
| "eval_rougeLsum": 35.2901, | |
| "eval_runtime": 722.0768, | |
| "eval_samples_per_second": 2.077, | |
| "eval_steps_per_second": 1.039, | |
| "step": 6230 | |
| } | |
| ], | |
| "max_steps": 37380, | |
| "num_train_epochs": 6, | |
| "total_flos": 1.48195611648e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |