|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: roberta_large-chunking_0811_v7 |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# roberta_large-chunking_0811_v7 |
|
|
|
This model is a fine-tuned version of [roberta-large](https://huggingface.co./roberta-large) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3687 |
|
- Precision: 0.8237 |
|
- Recall: 0.8406 |
|
- F1: 0.8320 |
|
- Accuracy: 0.9134 |
|
|
|
## 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: 1e-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: 7 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| 0.1929 | 1.0 | 1249 | 0.4165 | 0.8034 | 0.8191 | 0.8112 | 0.9047 | |
|
| 0.0789 | 2.0 | 2498 | 0.4161 | 0.8262 | 0.8363 | 0.8312 | 0.9088 | |
|
| 0.0319 | 3.0 | 3747 | 0.5684 | 0.8104 | 0.8380 | 0.8240 | 0.9037 | |
|
| 0.0198 | 4.0 | 4996 | 0.6959 | 0.8237 | 0.8433 | 0.8334 | 0.9067 | |
|
| 0.0098 | 5.0 | 6245 | 0.7280 | 0.8234 | 0.8453 | 0.8342 | 0.9084 | |
|
| 0.0075 | 6.0 | 7494 | 0.7482 | 0.8259 | 0.8482 | 0.8369 | 0.9075 | |
|
| 0.0041 | 7.0 | 8743 | 0.7807 | 0.8396 | 0.8527 | 0.8461 | 0.9113 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.21.1 |
|
- Pytorch 1.12.0+cu113 |
|
- Datasets 2.4.0 |
|
- Tokenizers 0.12.1 |
|
|
