umberto-wikipedia-uncased-v1
This model is a fine-tuned version of Musixmatch/umberto-wikipedia-uncased-v1 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4643
- F1: 0.6160
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: 32
- seed: 0
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- lr_scheduler_warmup_ratio: 0.05
- num_epochs: 5.0
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| 0.5923 | 0.33 | 20 | 0.6075 | 0.2974 |
| 0.6694 | 0.67 | 40 | 0.5938 | 0.2974 |
| 0.5225 | 1.0 | 60 | 0.5346 | 0.3024 |
| 0.5458 | 1.33 | 80 | 0.5172 | 0.4682 |
| 0.5052 | 1.67 | 100 | 0.5550 | 0.4361 |
| 0.3689 | 2.0 | 120 | 0.5188 | 0.4395 |
| 0.4097 | 2.33 | 140 | 0.4918 | 0.4676 |
| 0.4904 | 2.67 | 160 | 0.4756 | 0.4640 |
| 0.3696 | 3.0 | 180 | 0.4788 | 0.4688 |
| 0.3631 | 3.33 | 200 | 0.4751 | 0.4735 |
| 0.4345 | 3.67 | 220 | 0.4609 | 0.4797 |
| 0.3398 | 4.0 | 240 | 0.4637 | 0.5673 |
| 0.2985 | 4.33 | 260 | 0.4850 | 0.5092 |
| 0.3248 | 4.67 | 280 | 0.5089 | 0.6057 |
| 0.3323 | 5.0 | 300 | 0.4643 | 0.6160 |
Framework versions
- Transformers 4.34.0.dev0
- Pytorch 2.1.0.dev20230628+cu121
- Datasets 2.14.5
- Tokenizers 0.14.0
- Downloads last month
- 1