Instructions to use Mardiyyah/with_vague_label_all_false with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Mardiyyah/with_vague_label_all_false with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="Mardiyyah/with_vague_label_all_false")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("Mardiyyah/with_vague_label_all_false") model = AutoModelForTokenClassification.from_pretrained("Mardiyyah/with_vague_label_all_false", device_map="auto") - Notebooks
- Google Colab
- Kaggle
with_vague_label_all_false
This model is a fine-tuned version of microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext on the OTAR3088/CeLLaTe_with_vague1.0_IOB dataset. It achieves the following results on the evaluation set:
- Loss: 0.0983
- Precision: 0.6577
- Recall: 0.6520
- Micro F1: 0.6548
- Weighted F1: 0.6551
- Macro F1: 0.6531
- Accuracy: 0.9777
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 3407
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.01
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | Micro F1 | Weighted F1 | Macro F1 | Accuracy |
|---|---|---|---|---|---|---|---|---|---|
| 0.3528 | 1.0 | 263 | 0.0894 | 0.5972 | 0.5962 | 0.5967 | 0.5978 | 0.6007 | 0.9745 |
| 0.0582 | 2.0 | 526 | 0.0806 | 0.6162 | 0.6394 | 0.6276 | 0.6266 | 0.6266 | 0.9763 |
| 0.0398 | 3.0 | 789 | 0.0832 | 0.6312 | 0.6520 | 0.6414 | 0.6427 | 0.6424 | 0.9768 |
| 0.028 | 4.0 | 1052 | 0.0959 | 0.6734 | 0.6057 | 0.6377 | 0.6353 | 0.6344 | 0.9759 |
| 0.0207 | 5.0 | 1315 | 0.0975 | 0.6577 | 0.6520 | 0.6548 | 0.6551 | 0.6531 | 0.9777 |
| 0.0154 | 6.0 | 1578 | 0.1089 | 0.6550 | 0.6457 | 0.6503 | 0.6503 | 0.6499 | 0.9765 |
| 0.0121 | 7.0 | 1841 | 0.1136 | 0.6616 | 0.6512 | 0.6564 | 0.6551 | 0.6514 | 0.9770 |
| 0.0092 | 8.0 | 2104 | 0.1337 | 0.6857 | 0.6135 | 0.6476 | 0.6457 | 0.6454 | 0.9764 |
Framework versions
- Transformers 4.48.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.2
- Tokenizers 0.21.0
- Downloads last month
- 24