|
--- |
|
license: mit |
|
library_name: peft |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
base_model: microsoft/deberta-v3-xsmall |
|
model-index: |
|
- name: STS-Lora-Fine-Tuning-Capstone-Deberta-old-model-pipe-test |
|
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. --> |
|
|
|
# STS-Lora-Fine-Tuning-Capstone-Deberta-old-model-pipe-test |
|
|
|
This model is a fine-tuned version of [microsoft/deberta-v3-xsmall](https://huggingface.co/microsoft/deberta-v3-xsmall) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.4820 |
|
- Accuracy: 0.3771 |
|
|
|
## 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: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 10 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| No log | 1.0 | 360 | 1.7474 | 0.2429 | |
|
| 1.7416 | 2.0 | 720 | 1.7279 | 0.2429 | |
|
| 1.6866 | 3.0 | 1080 | 1.6799 | 0.2883 | |
|
| 1.6866 | 4.0 | 1440 | 1.6220 | 0.3372 | |
|
| 1.6241 | 5.0 | 1800 | 1.5787 | 0.3466 | |
|
| 1.5474 | 6.0 | 2160 | 1.5306 | 0.3604 | |
|
| 1.484 | 7.0 | 2520 | 1.5180 | 0.3626 | |
|
| 1.484 | 8.0 | 2880 | 1.5028 | 0.3706 | |
|
| 1.4452 | 9.0 | 3240 | 1.4871 | 0.3753 | |
|
| 1.429 | 10.0 | 3600 | 1.4820 | 0.3771 | |
|
|
|
|
|
### Framework versions |
|
|
|
- PEFT 0.9.0 |
|
- Transformers 4.38.2 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |