metadata
base_model: microsoft/mdeberta-v3-base
library_name: peft
license: mit
metrics:
- precision
- recall
- f1
- accuracy
tags:
- generated_from_trainer
model-index:
- name: mdeberta-v3-base-finetuned-ner
results: []
mdeberta-v3-base-finetuned-ner
This model is a fine-tuned version of microsoft/mdeberta-v3-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1094
- Precision: 0.9101
- Recall: 0.9567
- F1: 0.9328
- Accuracy: 0.9757
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: 0.0002
- train_batch_size: 8
- eval_batch_size: 8
- 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 |
|---|---|---|---|---|---|---|---|
| 0.2289 | 1.0 | 1167 | 0.1496 | 0.8186 | 0.8964 | 0.8557 | 0.9524 |
| 0.1466 | 2.0 | 2334 | 0.1193 | 0.8771 | 0.9312 | 0.9033 | 0.9677 |
| 0.1064 | 3.0 | 3501 | 0.1143 | 0.8768 | 0.9451 | 0.9097 | 0.9673 |
| 0.0808 | 4.0 | 4668 | 0.0978 | 0.8968 | 0.9491 | 0.9222 | 0.9724 |
| 0.0668 | 5.0 | 5835 | 0.1111 | 0.8889 | 0.9533 | 0.9200 | 0.9713 |
| 0.0559 | 6.0 | 7002 | 0.1168 | 0.9054 | 0.9561 | 0.9301 | 0.9746 |
| 0.0459 | 7.0 | 8169 | 0.1085 | 0.9174 | 0.9457 | 0.9313 | 0.9749 |
| 0.0421 | 8.0 | 9336 | 0.1077 | 0.9141 | 0.9516 | 0.9325 | 0.9759 |
| 0.0324 | 9.0 | 10503 | 0.1084 | 0.9148 | 0.9575 | 0.9357 | 0.9766 |
| 0.03 | 10.0 | 11670 | 0.1094 | 0.9101 | 0.9567 | 0.9328 | 0.9757 |
Framework versions
- PEFT 0.11.1
- Transformers 4.42.3
- Pytorch 2.1.2
- Datasets 2.20.0
- Tokenizers 0.19.1