metadata
license: mit
library_name: peft
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
base_model: roberta-large
model-index:
- name: roberta-large-finetuned-ner
results: []
roberta-large-finetuned-ner
This model is a fine-tuned version of roberta-large on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0828
- Precision: 0.9043
- Recall: 0.9245
- F1: 0.9143
- Accuracy: 0.9793
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: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.8259 | 1.0 | 878 | 0.2398 | 0.6827 | 0.7083 | 0.6953 | 0.9371 |
| 0.2115 | 2.0 | 1756 | 0.1560 | 0.8021 | 0.8172 | 0.8096 | 0.9600 |
| 0.1612 | 3.0 | 2634 | 0.1274 | 0.8589 | 0.8506 | 0.8547 | 0.9672 |
| 0.124 | 4.0 | 3512 | 0.1081 | 0.8832 | 0.8793 | 0.8813 | 0.9722 |
| 0.1183 | 5.0 | 4390 | 0.0993 | 0.8910 | 0.9036 | 0.8973 | 0.9754 |
| 0.1074 | 6.0 | 5268 | 0.0921 | 0.8974 | 0.9119 | 0.9046 | 0.9773 |
| 0.1004 | 7.0 | 6146 | 0.0874 | 0.8983 | 0.9156 | 0.9068 | 0.9780 |
| 0.0967 | 8.0 | 7024 | 0.0846 | 0.9028 | 0.9227 | 0.9127 | 0.9792 |
| 0.0923 | 9.0 | 7902 | 0.0829 | 0.9039 | 0.9239 | 0.9138 | 0.9795 |
| 0.0884 | 10.0 | 8780 | 0.0828 | 0.9043 | 0.9245 | 0.9143 | 0.9793 |
Framework versions
- PEFT 0.8.2
- Transformers 4.37.2
- Pytorch 2.2.0
- Datasets 2.17.0
- Tokenizers 0.15.2