ecoli-bsubtilis-finetuned-lora-NucleotideTransformer
This model is a fine-tuned version of InstaDeepAI/nucleotide-transformer-2.5b-multi-species on the jacquelinegrimm/ecoli-bsubtilis dataset. It achieves the following results on the evaluation set:
- Loss: 0.0882
- F1 Score: 0.9721
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: 0.0005
- train_batch_size: 8
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 1000
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 Score |
|---|---|---|---|---|
| 0.4278 | 0.05 | 100 | 0.9650 | 0.8596 |
| 0.3987 | 0.1 | 200 | 0.3144 | 0.9504 |
| 0.2956 | 0.14 | 300 | 0.1939 | 0.9504 |
| 0.3176 | 0.19 | 400 | 0.1423 | 0.9643 |
| 0.2747 | 0.24 | 500 | 0.1104 | 0.9705 |
| 0.1569 | 0.29 | 600 | 0.1003 | 0.9624 |
| 0.239 | 0.33 | 700 | 0.0628 | 0.9737 |
| 0.1457 | 0.38 | 800 | 0.1421 | 0.9611 |
| 0.1239 | 0.43 | 900 | 0.1064 | 0.975 |
| 0.1221 | 0.48 | 1000 | 0.0882 | 0.9721 |
Framework versions
- PEFT 0.10.0
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 0