<!-- 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. -->
disaster-tweet-distilbert-3
This model is a fine-tuned version of distilbert-base-uncased-finetuned-sst-2-english on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4337
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: 4e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
2.2875 | 0.12 | 12 | 2.3044 |
2.1117 | 0.25 | 24 | 2.2753 |
2.2477 | 0.38 | 36 | 2.2243 |
2.2616 | 0.5 | 48 | 2.1505 |
2.1278 | 0.62 | 60 | 2.0550 |
2.0568 | 0.75 | 72 | 1.9456 |
1.8358 | 0.88 | 84 | 1.8217 |
1.7638 | 1.0 | 96 | 1.6916 |
1.531 | 1.12 | 108 | 1.5616 |
1.5567 | 1.25 | 120 | 1.4224 |
1.3747 | 1.38 | 132 | 1.2834 |
1.2675 | 1.5 | 144 | 1.1505 |
1.0291 | 1.62 | 156 | 1.0330 |
0.9212 | 1.75 | 168 | 0.9307 |
0.91 | 1.88 | 180 | 0.8426 |
0.7726 | 2.0 | 192 | 0.7676 |
0.671 | 2.12 | 204 | 0.7126 |
0.6759 | 2.25 | 216 | 0.6655 |
0.6012 | 2.38 | 228 | 0.6287 |
0.6228 | 2.5 | 240 | 0.5989 |
0.5432 | 2.62 | 252 | 0.5753 |
0.5475 | 2.75 | 264 | 0.5555 |
0.5788 | 2.88 | 276 | 0.5381 |
0.4944 | 3.0 | 288 | 0.5245 |
0.4692 | 3.12 | 300 | 0.5158 |
0.4743 | 3.25 | 312 | 0.4995 |
0.4333 | 3.38 | 324 | 0.4912 |
0.4768 | 3.5 | 336 | 0.4813 |
0.4653 | 3.62 | 348 | 0.4730 |
0.4249 | 3.75 | 360 | 0.4701 |
0.4815 | 3.88 | 372 | 0.4613 |
0.4349 | 4.0 | 384 | 0.4552 |
0.3723 | 4.12 | 396 | 0.4509 |
0.456 | 4.25 | 408 | 0.4469 |
0.3988 | 4.38 | 420 | 0.4458 |
0.4142 | 4.5 | 432 | 0.4456 |
0.4008 | 4.62 | 444 | 0.4385 |
0.3943 | 4.75 | 456 | 0.4376 |
0.3862 | 4.88 | 468 | 0.4348 |
0.3778 | 5.0 | 480 | 0.4337 |
Framework versions
- Transformers 4.28.1
- Pytorch 1.13.0
- Datasets 2.1.0
- Tokenizers 0.13.2