distilbert-base-uncased_fold_9_binary_v1
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.6965
- F1: 0.8090
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: 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: 25
Training results
Training Loss | Epoch | Step | Validation Loss | F1 |
---|---|---|---|---|
No log | 1.0 | 291 | 0.4193 | 0.7989 |
0.3993 | 2.0 | 582 | 0.4039 | 0.8026 |
0.3993 | 3.0 | 873 | 0.5227 | 0.7995 |
0.2044 | 4.0 | 1164 | 0.7264 | 0.8011 |
0.2044 | 5.0 | 1455 | 0.8497 | 0.8007 |
0.0882 | 6.0 | 1746 | 0.9543 | 0.8055 |
0.0374 | 7.0 | 2037 | 1.1349 | 0.7997 |
0.0374 | 8.0 | 2328 | 1.3175 | 0.8009 |
0.0151 | 9.0 | 2619 | 1.3585 | 0.8030 |
0.0151 | 10.0 | 2910 | 1.4202 | 0.8067 |
0.0068 | 11.0 | 3201 | 1.4364 | 0.8108 |
0.0068 | 12.0 | 3492 | 1.4443 | 0.8088 |
0.0096 | 13.0 | 3783 | 1.5308 | 0.8075 |
0.0031 | 14.0 | 4074 | 1.5061 | 0.8020 |
0.0031 | 15.0 | 4365 | 1.5769 | 0.7980 |
0.0048 | 16.0 | 4656 | 1.5962 | 0.8038 |
0.0048 | 17.0 | 4947 | 1.5383 | 0.8085 |
0.0067 | 18.0 | 5238 | 1.5456 | 0.8158 |
0.0062 | 19.0 | 5529 | 1.6325 | 0.8044 |
0.0062 | 20.0 | 5820 | 1.5430 | 0.8141 |
0.0029 | 21.0 | 6111 | 1.6590 | 0.8117 |
0.0029 | 22.0 | 6402 | 1.6650 | 0.8112 |
0.0017 | 23.0 | 6693 | 1.7016 | 0.8053 |
0.0017 | 24.0 | 6984 | 1.6998 | 0.8090 |
0.0011 | 25.0 | 7275 | 1.6965 | 0.8090 |
Framework versions
- Transformers 4.21.0
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1
- Downloads last month
- 11
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.