Instructions to use dendimaki/multilabel_classification_v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use dendimaki/multilabel_classification_v2 with PEFT:
from peft import PeftModel from transformers import AutoModelForSequenceClassification base_model = AutoModelForSequenceClassification.from_pretrained("mistralai/Mistral-7B-v0.1") model = PeftModel.from_pretrained(base_model, "dendimaki/multilabel_classification_v2") - Notebooks
- Google Colab
- Kaggle
Download checkpoint-478/trainer_state.json from dendimaki/multilabel_classification_v2: direct link, hf CLI and curl.
- Browser
- Download file 859 Bytes
-
https://huggingface.co/dendimaki/multilabel_classification_v2/resolve/main/checkpoint-478/trainer_state.json
- Command line
-
hf download hf://dendimaki/multilabel_classification_v2/checkpoint-478/trainer_state.json
-
curl -L -o trainer_state.json https://huggingface.co/dendimaki/multilabel_classification_v2/resolve/main/checkpoint-478/trainer_state.json
859 Bytes
| { | |
| "best_metric": 0.09056513756513596, | |
| "best_model_checkpoint": "multilabel_classification/checkpoint-478", | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 478, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_f1_macro": 0.13708533383969368, | |
| "eval_f1_micro": 0.391941391941392, | |
| "eval_f1_weighted": 0.34667186055746785, | |
| "eval_loss": 0.09056513756513596, | |
| "eval_runtime": 160.5811, | |
| "eval_samples_per_second": 2.653, | |
| "eval_steps_per_second": 0.336, | |
| "step": 478 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 478, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 1.2446242298855424e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |