<!-- 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. -->
distilbert-helpdesk-sentiment
This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3008
- Accuracy: 0.9417
Model description
Model was trained on Dataset of abstractive summaries of customer and helpdesk which include mixed sentiment[negative and positive] in single sample(sentence) but data labeled carefully to customer's sentiment.
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: 2
- 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 | Accuracy |
|---|---|---|---|---|
| No log | 0.68 | 100 | 0.3234 | 0.8932 |
| No log | 1.35 | 200 | 0.3211 | 0.9029 |
| No log | 2.03 | 300 | 0.3265 | 0.9126 |
| No log | 2.7 | 400 | 0.3553 | 0.9126 |
| No log | 3.38 | 500 | 0.3894 | 0.9223 |
| 0.2663 | 4.05 | 600 | 0.3379 | 0.9223 |
| 0.2663 | 4.73 | 700 | 0.2877 | 0.9320 |
| 0.2663 | 5.41 | 800 | 0.3072 | 0.9417 |
| 0.2663 | 6.08 | 900 | 0.2992 | 0.9320 |
| 0.2663 | 6.76 | 1000 | 0.3008 | 0.9417 |
Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3
