metadata
library_name: transformers
base_model: YituTech/conv-bert-base
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: convbert-finetuned-ner
results: []
convbert-finetuned-ner
This model is a fine-tuned version of YituTech/conv-bert-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1088
- Precision: 0.9589
- Recall: 0.9707
- F1: 0.9648
- Accuracy: 0.9805
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: 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: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| 0.074 | 1.0 | 5285 | 0.1325 | 0.9499 | 0.9621 | 0.9560 | 0.9744 |
| 0.0658 | 2.0 | 10570 | 0.1142 | 0.9567 | 0.9679 | 0.9623 | 0.9788 |
| 0.0908 | 3.0 | 15855 | 0.1088 | 0.9589 | 0.9707 | 0.9648 | 0.9805 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.19.1