metadata
license: mit
library_name: peft
tags:
- generated_from_trainer
base_model: microsoft/phi-2
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: phi-finetuned-spam
results: []
phi-finetuned-spam
This model is a fine-tuned version of microsoft/phi-2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0822
- Accuracy: 0.989
- F1: 0.9890
- Precision: 0.9880
- Recall: 0.99
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: 3.628060796399553e-05
- train_batch_size: 8
- eval_batch_size: 2
- seed: 31
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.1312 | 1.0 | 1125 | 0.1122 | 0.974 | 0.9735 | 0.9917 | 0.956 |
| 0.0224 | 2.0 | 2250 | 0.0822 | 0.989 | 0.9890 | 0.9880 | 0.99 |
| 0.0654 | 3.0 | 3375 | 0.0806 | 0.988 | 0.988 | 0.988 | 0.988 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1