|
|
--- |
|
|
license: mit |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: ec-biogpt-noised-pubmed-v2 |
|
|
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. --> |
|
|
|
|
|
# ec-biogpt-noised-pubmed-v2 |
|
|
|
|
|
This model is a fine-tuned version of [microsoft/biogpt](https://huggingface.co/microsoft/biogpt) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 1.2703 |
|
|
|
|
|
## 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-05 |
|
|
- train_batch_size: 16 |
|
|
- eval_batch_size: 16 |
|
|
- seed: 42 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- lr_scheduler_warmup_steps: 10 |
|
|
- num_epochs: 5 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
|
| 1.1503 | 0.11 | 500 | 1.3369 | |
|
|
| 1.3766 | 0.21 | 1000 | 1.2721 | |
|
|
| 1.3523 | 0.32 | 1500 | 1.2516 | |
|
|
| 1.3123 | 0.43 | 2000 | 1.2394 | |
|
|
| 1.1954 | 0.54 | 2500 | 1.2265 | |
|
|
| 1.226 | 0.64 | 3000 | 1.2182 | |
|
|
| 1.1269 | 0.75 | 3500 | 1.2118 | |
|
|
| 1.212 | 0.86 | 4000 | 1.2053 | |
|
|
| 1.3253 | 0.96 | 4500 | 1.1984 | |
|
|
| 1.0722 | 1.07 | 5000 | 1.2016 | |
|
|
| 1.1208 | 1.18 | 5500 | 1.2009 | |
|
|
| 1.132 | 1.28 | 6000 | 1.1992 | |
|
|
| 1.1228 | 1.39 | 6500 | 1.1967 | |
|
|
| 1.1529 | 1.5 | 7000 | 1.1918 | |
|
|
| 1.0342 | 1.61 | 7500 | 1.1916 | |
|
|
| 1.0881 | 1.71 | 8000 | 1.1889 | |
|
|
| 1.084 | 1.82 | 8500 | 1.1852 | |
|
|
| 1.1409 | 1.93 | 9000 | 1.1807 | |
|
|
| 0.9794 | 2.03 | 9500 | 1.2098 | |
|
|
| 0.9821 | 2.14 | 10000 | 1.2146 | |
|
|
| 0.9695 | 2.25 | 10500 | 1.2096 | |
|
|
| 0.9866 | 2.35 | 11000 | 1.2088 | |
|
|
| 1.0305 | 2.46 | 11500 | 1.2059 | |
|
|
| 0.9532 | 2.57 | 12000 | 1.2060 | |
|
|
| 0.9978 | 2.68 | 12500 | 1.2041 | |
|
|
| 1.0013 | 2.78 | 13000 | 1.2006 | |
|
|
| 1.0401 | 2.89 | 13500 | 1.2023 | |
|
|
| 1.0899 | 3.0 | 14000 | 1.1988 | |
|
|
| 0.8229 | 3.1 | 14500 | 1.2410 | |
|
|
| 0.8598 | 3.21 | 15000 | 1.2420 | |
|
|
| 0.9295 | 3.32 | 15500 | 1.2414 | |
|
|
| 0.8477 | 3.43 | 16000 | 1.2386 | |
|
|
| 0.9302 | 3.53 | 16500 | 1.2382 | |
|
|
| 0.8284 | 3.64 | 17000 | 1.2374 | |
|
|
| 0.8242 | 3.75 | 17500 | 1.2410 | |
|
|
| 0.8422 | 3.85 | 18000 | 1.2346 | |
|
|
| 0.8742 | 3.96 | 18500 | 1.2362 | |
|
|
| 0.798 | 4.07 | 19000 | 1.2667 | |
|
|
| 0.7821 | 4.17 | 19500 | 1.2701 | |
|
|
| 0.7788 | 4.28 | 20000 | 1.2714 | |
|
|
| 0.7701 | 4.39 | 20500 | 1.2702 | |
|
|
| 0.7348 | 4.5 | 21000 | 1.2722 | |
|
|
| 0.762 | 4.6 | 21500 | 1.2705 | |
|
|
| 0.7385 | 4.71 | 22000 | 1.2705 | |
|
|
| 0.7837 | 4.82 | 22500 | 1.2695 | |
|
|
| 0.8371 | 4.92 | 23000 | 1.2703 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.27.4 |
|
|
- Pytorch 2.0.0+cu118 |
|
|
- Datasets 2.11.0 |
|
|
- Tokenizers 0.13.3 |
|
|
|
