<!-- 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. -->
M6_MLM_cross
This model is a fine-tuned version of S2312dal/M6_MLM on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0197
- Pearson: 0.9680
- Spearmanr: 0.9098
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: 16
- eval_batch_size: 16
- seed: 25
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 8.0
- num_epochs: 3
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Pearson | Spearmanr |
|---|---|---|---|---|---|
| 0.0723 | 1.0 | 131 | 0.0646 | 0.8674 | 0.8449 |
| 0.0433 | 2.0 | 262 | 0.0322 | 0.9475 | 0.9020 |
| 0.0015 | 3.0 | 393 | 0.0197 | 0.9680 | 0.9098 |
Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1
