|
|
--- |
|
|
library_name: transformers |
|
|
tags: [] |
|
|
--- |
|
|
|
|
|
# Time Saving Stated Aim Classifier |
|
|
|
|
|
This is a roberta-base model that is trained to classify whether a set of explicit stated aims extracted from a British [historical patent](https://huggingface.co/datasets/matthewleechen/300YearsOfBritishPatents) includes a time-saving objective. |
|
|
|
|
|
Labels were manually generated and then checked with Gemini 2.0 Flash with the attached [prompt](https://huggingface.co/matthewleechen/time-saving_stated_aim_classifier/blob/main/time-saving_prompt.txt). |
|
|
|
|
|
Hyperparameters: |
|
|
lr = 3e-5 |
|
|
batch size = 128 |
|
|
|
|
|
Test set results: |
|
|
|
|
|
```text |
|
|
{'eval_loss': 0.5350350737571716, |
|
|
'eval_accuracy': 0.89, |
|
|
'eval_precision': 0.8972633319553535, |
|
|
'eval_recall': 0.89, |
|
|
'eval_f1': 0.8891426257718393, |
|
|
'eval_runtime': 0.4081, |
|
|
'eval_samples_per_second': 245.053, |
|
|
'eval_steps_per_second': 2.451} |
|
|
``` |
|
|
|
|
|
|
