aa02dfb603b2dc87ed9bd26e0198a0d0
This model is a fine-tuned version of distilbert/distilbert-base-uncased-distilled-squad on the ccdv/patent-classification [abstract] dataset. It achieves the following results on the evaluation set:
- Loss: 1.4480
- Data Size: 1.0
- Epoch Runtime: 22.4982
- Accuracy: 0.6356
- F1 Macro: 0.5903
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 32
- total_eval_batch_size: 32
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: constant
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Data Size | Epoch Runtime | Accuracy | F1 Macro |
|---|---|---|---|---|---|---|---|
| No log | 0 | 0 | 2.1711 | 0 | 2.0357 | 0.1941 | 0.0615 |
| No log | 1 | 781 | 1.9999 | 0.0078 | 2.3529 | 0.2218 | 0.0406 |
| No log | 2 | 1562 | 1.9046 | 0.0156 | 2.4318 | 0.2698 | 0.1046 |
| No log | 3 | 2343 | 1.6173 | 0.0312 | 2.8650 | 0.3702 | 0.2051 |
| 0.0413 | 4 | 3124 | 1.3973 | 0.0625 | 3.5350 | 0.5182 | 0.3476 |
| 1.3611 | 5 | 3905 | 1.2288 | 0.125 | 4.9177 | 0.5661 | 0.4046 |
| 1.1813 | 6 | 4686 | 1.1124 | 0.25 | 7.3893 | 0.6102 | 0.4789 |
| 1.01 | 7 | 5467 | 1.0400 | 0.5 | 12.6409 | 0.6426 | 0.5724 |
| 0.9023 | 8.0 | 6248 | 1.0476 | 1.0 | 23.3686 | 0.6585 | 0.5940 |
| 0.7212 | 9.0 | 7029 | 1.0730 | 1.0 | 22.7716 | 0.6581 | 0.6001 |
| 0.5386 | 10.0 | 7810 | 1.2044 | 1.0 | 22.5663 | 0.6412 | 0.5913 |
| 0.3815 | 11.0 | 8591 | 1.4480 | 1.0 | 22.4982 | 0.6356 | 0.5903 |
Framework versions
- Transformers 4.57.0
- Pytorch 2.8.0+cu128
- Datasets 4.3.0
- Tokenizers 0.22.1
- Downloads last month
- -