|
--- |
|
tags: |
|
- autotrain |
|
- text-classification |
|
widget: |
|
- text: I love AutoTrain |
|
datasets: |
|
- pruhtopia/multilingual-bert-toc-95k-dataset |
|
license: apache-2.0 |
|
--- |
|
|
|
# Model Trained Using AutoTrain |
|
|
|
- Problem type: Text Classification |
|
- Task: Legal Document Sequence Classification w/ [bert-base-multilingual-cased](https://huggingface.co/google-bert/bert-base-multilingual-cased) |
|
- id2label: [0: 'Caption', |
|
1: 'Footnote', |
|
2: 'Formula', |
|
3: 'List-item', |
|
4: 'Page-footer', |
|
5: 'Page-header', |
|
6: 'Picture', |
|
7: 'Section-header', |
|
8: 'Table', |
|
9: 'Text', |
|
10: 'Title'] |
|
- sample usage notebook [here](https://colab.research.google.com/drive/1tSpV0RC12LDNFbWEq6kdiUCIJ8DkHHU_?usp=sharing) |
|
|
|
## Validation Metrics |
|
loss: 0.5102838277816772 |
|
|
|
f1_macro: 0.605011586308457 |
|
|
|
f1_micro: 0.8910038281582305 |
|
|
|
f1_weighted: 0.8870714364293508 |
|
|
|
precision_macro: 0.6869883411452264 |
|
|
|
precision_micro: 0.8910038281582305 |
|
|
|
precision_weighted: 0.8858066104824025 |
|
|
|
recall_macro: 0.5550753643871188 |
|
|
|
recall_micro: 0.8910038281582305 |
|
|
|
recall_weighted: 0.8910038281582305 |
|
|
|
accuracy: 0.8910038281582305 |