---
license: cc-by-nc-4.0
---


[allenai/tk-instruct-base-def-pos](https://huggingface.co/allenai/tk-instruct-base-def-pos) model fined tuned on custom pyabsa english model from 'ATEPC_MULTILINGUAL_CHECKPOINT' auto annotated hospital reviews [keywords,polarity] dataset.


Training Results:

Epoch | Training Loss | Validation Loss

1 | 0.061900 | 0.047395

2 | 0.038300 | 0.035213

3 | 0.029700 | 0.028486

4 | 0.024900 | 0.028562

-----------------------------------------------------------------------------------------------------------------------------------------------

Evaluation Results:

Train Precision:  0.9682688696958318

Train Recall:  0.9692090837901332

Train F1:  0.968738748610698

Test Precision:  0.965

Test Recall:  0.9620517768129925

Test F1:  0.9635236331427651

UnseenTest Precision:  0.9670502092050209

UnseenTest Recall:  0.9685699319015191

UnseenTest F1:  0.9678094739596965