finetuned-customer-intent-distilbert
This model is a fine-tuned version of distilbert-base-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2456
- Accuracy: 0.8247
- F1: 0.8247
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: 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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|---|---|---|---|---|---|
| No log | 1.0 | 25 | 3.1005 | 0.2835 | 0.2666 |
| No log | 2.0 | 50 | 2.5885 | 0.6598 | 0.6428 |
| No log | 3.0 | 75 | 2.0839 | 0.6959 | 0.6772 |
| No log | 4.0 | 100 | 1.6845 | 0.7371 | 0.7289 |
| No log | 5.0 | 125 | 1.4019 | 0.7835 | 0.7799 |
| No log | 6.0 | 150 | 1.2387 | 0.8093 | 0.8090 |
| No log | 7.0 | 175 | 1.1484 | 0.8144 | 0.8143 |
| No log | 8.0 | 200 | 1.1057 | 0.8247 | 0.8247 |
| No log | 9.0 | 225 | 1.1020 | 0.8247 | 0.8247 |
| No log | 10.0 | 250 | 1.1103 | 0.8247 | 0.8247 |
| No log | 11.0 | 275 | 1.1397 | 0.8247 | 0.8247 |
| No log | 12.0 | 300 | 1.1622 | 0.8247 | 0.8247 |
| No log | 13.0 | 325 | 1.1783 | 0.8247 | 0.8247 |
| No log | 14.0 | 350 | 1.1990 | 0.8247 | 0.8247 |
| No log | 15.0 | 375 | 1.2142 | 0.8247 | 0.8247 |
| No log | 16.0 | 400 | 1.2248 | 0.8247 | 0.8247 |
| No log | 17.0 | 425 | 1.2333 | 0.8247 | 0.8247 |
| No log | 18.0 | 450 | 1.2397 | 0.8247 | 0.8247 |
| No log | 19.0 | 475 | 1.2447 | 0.8247 | 0.8247 |
| No log | 20.0 | 500 | 1.2456 | 0.8247 | 0.8247 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 78
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.