|
--- |
|
license: mit |
|
library_name: peft |
|
tags: |
|
- generated_from_trainer |
|
base_model: openai-community/gpt2 |
|
datasets: |
|
- emotion |
|
metrics: |
|
- accuracy |
|
- f1 |
|
- precision |
|
- recall |
|
model-index: |
|
- name: emotion-gpt2-lora |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# emotion-gpt2-lora |
|
|
|
This model is a fine-tuned version of [openai-community/gpt2](https://huggingface.co/openai-community/gpt2) on the emotion dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1521 |
|
- Accuracy: 0.933 |
|
- F1: 0.9334 |
|
- Precision: 0.9347 |
|
- Recall: 0.933 |
|
|
|
## 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: 64 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 4 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:| |
|
| No log | 1.0 | 250 | 0.3191 | 0.8895 | 0.8902 | 0.8933 | 0.8895 | |
|
| 0.6939 | 2.0 | 500 | 0.1939 | 0.935 | 0.9349 | 0.9352 | 0.935 | |
|
| 0.6939 | 3.0 | 750 | 0.1689 | 0.931 | 0.9315 | 0.9329 | 0.931 | |
|
| 0.1897 | 4.0 | 1000 | 0.1521 | 0.933 | 0.9334 | 0.9347 | 0.933 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.10.0 |
|
- Transformers 4.40.2 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
