|
--- |
|
license: mit |
|
base_model: microsoft/phi-2 |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: phi_2_patent |
|
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. --> |
|
|
|
# phi_2_patent |
|
|
|
This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.9403 |
|
- Accuracy: 0.6678 |
|
- F1 Macro: 0.6213 |
|
- F1 Micro: 0.6678 |
|
|
|
## 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: 5e-06 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 2 |
|
- total_train_batch_size: 64 |
|
- total_eval_batch_size: 64 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:| |
|
| 1.9105 | 0.13 | 50 | 1.8338 | 0.3248 | 0.2207 | 0.3248 | |
|
| 1.6023 | 0.26 | 100 | 1.6011 | 0.438 | 0.3087 | 0.438 | |
|
| 1.4113 | 0.38 | 150 | 1.4239 | 0.4912 | 0.3599 | 0.4912 | |
|
| 1.3062 | 0.51 | 200 | 1.2828 | 0.5498 | 0.4109 | 0.5498 | |
|
| 1.2574 | 0.64 | 250 | 1.1801 | 0.5822 | 0.4702 | 0.5822 | |
|
| 1.1687 | 0.77 | 300 | 1.1401 | 0.6032 | 0.4825 | 0.6032 | |
|
| 1.1396 | 0.9 | 350 | 1.0853 | 0.613 | 0.5246 | 0.613 | |
|
| 1.0529 | 1.02 | 400 | 1.0639 | 0.6226 | 0.5368 | 0.6226 | |
|
| 1.0261 | 1.15 | 450 | 1.0742 | 0.6304 | 0.5449 | 0.6304 | |
|
| 1.0068 | 1.28 | 500 | 1.0340 | 0.6444 | 0.5825 | 0.6444 | |
|
| 0.975 | 1.41 | 550 | 1.0151 | 0.65 | 0.5777 | 0.65 | |
|
| 0.966 | 1.53 | 600 | 1.0022 | 0.6498 | 0.5923 | 0.6498 | |
|
| 1.0201 | 1.66 | 650 | 0.9899 | 0.6562 | 0.5854 | 0.6562 | |
|
| 0.9346 | 1.79 | 700 | 0.9807 | 0.6598 | 0.5735 | 0.6598 | |
|
| 0.9807 | 1.92 | 750 | 0.9694 | 0.6586 | 0.6004 | 0.6586 | |
|
| 0.917 | 2.05 | 800 | 0.9664 | 0.6608 | 0.6086 | 0.6608 | |
|
| 0.9268 | 2.17 | 850 | 0.9619 | 0.6626 | 0.6107 | 0.6626 | |
|
| 1.0107 | 2.3 | 900 | 0.9548 | 0.6648 | 0.6156 | 0.6648 | |
|
| 0.9378 | 2.43 | 950 | 0.9559 | 0.6656 | 0.6109 | 0.6656 | |
|
| 0.9199 | 2.56 | 1000 | 0.9514 | 0.6658 | 0.6165 | 0.6658 | |
|
| 0.8467 | 2.69 | 1050 | 0.9454 | 0.6714 | 0.6203 | 0.6714 | |
|
| 0.8923 | 2.81 | 1100 | 0.9413 | 0.67 | 0.6206 | 0.67 | |
|
| 0.9545 | 2.94 | 1150 | 0.9403 | 0.6678 | 0.6213 | 0.6678 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.39.0.dev0 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
