metadata
license: mit
base_model: roberta-base
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: bryan-NM
results: []
bryan-NM
This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.6055
- Accuracy: 0.5427
- F1: 0.5374
- Precision: 0.5365
- Recall: 0.5427
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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 2.5019 | 1.0 | 3538 | 2.4996 | 0.3528 | 0.2684 | 0.2489 | 0.3528 |
| 2.1176 | 2.0 | 7076 | 2.1051 | 0.4478 | 0.3790 | 0.3630 | 0.4478 |
| 1.865 | 3.0 | 10614 | 1.9322 | 0.4827 | 0.4286 | 0.4186 | 0.4827 |
| 1.6961 | 4.0 | 14152 | 1.8417 | 0.5057 | 0.4681 | 0.4666 | 0.5057 |
| 1.5709 | 5.0 | 17690 | 1.8285 | 0.5149 | 0.4812 | 0.4826 | 0.5149 |
| 1.3717 | 6.0 | 21228 | 1.8153 | 0.5219 | 0.4938 | 0.5049 | 0.5219 |
| 1.2889 | 7.0 | 24766 | 1.8005 | 0.5369 | 0.5089 | 0.5040 | 0.5369 |
| 1.1258 | 8.0 | 28304 | 1.8269 | 0.5311 | 0.5103 | 0.5090 | 0.5311 |
| 1.0029 | 9.0 | 31842 | 1.8697 | 0.5421 | 0.5274 | 0.5289 | 0.5421 |
| 0.9032 | 10.0 | 35380 | 1.9533 | 0.5393 | 0.5255 | 0.5271 | 0.5393 |
| 0.7787 | 11.0 | 38918 | 2.0320 | 0.5371 | 0.5242 | 0.5245 | 0.5371 |
| 0.7137 | 12.0 | 42456 | 2.0956 | 0.5425 | 0.5331 | 0.5335 | 0.5425 |
| 0.6612 | 13.0 | 45994 | 2.1384 | 0.5419 | 0.5294 | 0.5291 | 0.5419 |
| 0.5733 | 14.0 | 49532 | 2.2058 | 0.5385 | 0.5289 | 0.5273 | 0.5385 |
| 0.5251 | 15.0 | 53070 | 2.2882 | 0.5397 | 0.5304 | 0.5276 | 0.5397 |
| 0.4666 | 16.0 | 56608 | 2.3806 | 0.5393 | 0.5327 | 0.5337 | 0.5393 |
| 0.4345 | 17.0 | 60146 | 2.4534 | 0.5485 | 0.5379 | 0.5366 | 0.5485 |
| 0.3668 | 18.0 | 63684 | 2.5234 | 0.5433 | 0.5368 | 0.5370 | 0.5433 |
| 0.3695 | 19.0 | 67222 | 2.5849 | 0.5417 | 0.5377 | 0.5381 | 0.5417 |
| 0.3226 | 20.0 | 70760 | 2.6055 | 0.5427 | 0.5374 | 0.5365 | 0.5427 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2