esm2-t6-8M-lora-256-remote-homology-filtered
This model is a fine-tuned version of facebook/esm2_t6_8M_UR50D on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5982
- Precision: 0.6901
- Recall: 0.6529
- F1-score: 0.6709
- Accuracy: 0.6788
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 16
- 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: 5
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1-score | Accuracy |
---|---|---|---|---|---|---|---|
0.6365 | 1.0 | 7969 | 0.6357 | 0.6218 | 0.7071 | 0.6617 | 0.6374 |
0.6046 | 2.0 | 15938 | 0.6102 | 0.6864 | 0.6149 | 0.6487 | 0.6660 |
0.6134 | 3.0 | 23907 | 0.6017 | 0.6887 | 0.6469 | 0.6672 | 0.6763 |
0.6108 | 4.0 | 31876 | 0.5986 | 0.6920 | 0.6468 | 0.6687 | 0.6785 |
0.5831 | 5.0 | 39845 | 0.5982 | 0.6901 | 0.6529 | 0.6709 | 0.6788 |
Framework versions
- PEFT 0.11.1
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 0
Model tree for sasuface/esm2-t6-8M-lora-256-remote-homology-filtered
Base model
facebook/esm2_t6_8M_UR50D