metadata
license: mit
tags:
- generated_from_trainer
model-index:
- name: farsi_lastname_classifier
results: []
farsi_lastname_classifier
This model is a fine-tuned version of microsoft/deberta-v3-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0501
- Pearson: 0.9174
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: 128
- eval_batch_size: 256
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Pearson |
|---|---|---|---|---|
| No log | 1.0 | 12 | 0.3069 | 0.3652 |
| No log | 2.0 | 24 | 0.2113 | 0.6984 |
| No log | 3.0 | 36 | 0.1256 | 0.7083 |
| No log | 4.0 | 48 | 0.1069 | 0.7647 |
| No log | 5.0 | 60 | 0.0523 | 0.8959 |
| No log | 6.0 | 72 | 0.0620 | 0.8976 |
| No log | 7.0 | 84 | 0.0581 | 0.8984 |
| No log | 8.0 | 96 | 0.0458 | 0.9189 |
| No log | 9.0 | 108 | 0.0520 | 0.9146 |
| No log | 10.0 | 120 | 0.0490 | 0.9212 |
| No log | 11.0 | 132 | 0.0466 | 0.9230 |
| No log | 12.0 | 144 | 0.0491 | 0.9198 |
| No log | 13.0 | 156 | 0.0495 | 0.9196 |
| No log | 14.0 | 168 | 0.0499 | 0.9176 |
| No log | 15.0 | 180 | 0.0501 | 0.9174 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.2