gemma-2b-heart
This model is a fine-tuned version of
google/gemma-2b on the None dataset.
It achieves the following results on the evaluation set:
-
Loss: 0.4600
-
Accuracy: 0.7944
-
Report: precision recall f1-score support
absence 0.80 0.79 0.80 92
presence 0.79 0.80 0.79 88
accuracy 0.79 180
macro avg 0.79 0.79 0.79 180
weighted avg 0.79 0.79 0.79 180
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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Report |
|---|
| No log | 1.0 | 105 | 0.6255 | 0.7833 | precision recall f1-score support |
absence 0.77 0.82 0.79 92
presence 0.80 0.75 0.77 88
accuracy 0.78 180
macro avg 0.78 0.78 0.78 180
weighted avg 0.78 0.78 0.78 180
|
| No log | 2.0 | 210 | 0.5303 | 0.7778 | precision recall f1-score support
absence 0.78 0.79 0.78 92
presence 0.78 0.76 0.77 88
accuracy 0.78 180
macro avg 0.78 0.78 0.78 180
weighted avg 0.78 0.78 0.78 180
|
| No log | 3.0 | 315 | 0.5015 | 0.7722 | precision recall f1-score support
absence 0.74 0.86 0.79 92
presence 0.82 0.68 0.75 88
accuracy 0.77 180
macro avg 0.78 0.77 0.77 180
weighted avg 0.78 0.77 0.77 180
|
| No log | 4.0 | 420 | 0.4662 | 0.7833 | precision recall f1-score support
absence 0.78 0.79 0.79 92
presence 0.78 0.77 0.78 88
accuracy 0.78 180
macro avg 0.78 0.78 0.78 180
weighted avg 0.78 0.78 0.78 180
|
| 0.566 | 5.0 | 525 | 0.4600 | 0.7944 | precision recall f1-score support
absence 0.80 0.79 0.80 92
presence 0.79 0.80 0.79 88
accuracy 0.79 180
macro avg 0.79 0.79 0.79 180
weighted avg 0.79 0.79 0.79 180
|
Framework versions
- PEFT 0.15.2
- Transformers 4.51.3
- Pytorch 2.6.0+cu124
- Datasets 2.14.4
- Tokenizers 0.21.1
