| | --- |
| | license: apache-2.0 |
| | tags: |
| | - generated_from_trainer |
| | metrics: |
| | - precision |
| | - recall |
| | - f1 |
| | - accuracy |
| | model-index: |
| | - name: bert-finetuned-ner |
| | results: [] |
| | language: |
| | - en |
| | pipeline_tag: token-classification |
| | --- |
| | |
| | <!-- 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. --> |
| |
|
| | # bert-finetuned-AAVE-PoS |
| |
|
| | This model is a version of [bert-base-cased](https://huggingface.co/bert-base-cased) fine-tuned on a [dataset](https://bitbucket.org/soegaard/aave-pos16/src/master/data) of African American Vernacular English (AAVE) which was published alongside [Jørgensen et al. 2016](https://aclanthology.org/N16-1130.pdf). |
| | It achieves the following results on the evaluation set: |
| | - Loss: 0.2582 |
| | - Precision: 0.8632 |
| | - Recall: 0.8730 |
| | - F1: 0.8681 |
| | - Accuracy: 0.9356 |
| |
|
| | ## Model description |
| |
|
| | More information needed |
| |
|
| | ## Intended uses & limitations |
| |
|
| | This model is intended to help close the gap in part-of-speech tagging performance between Standard American English (SAE) and African American English (AAVE) which differ liguistically in many [well-documented](http://www.johnrickford.com/portals/45/documents/papers/Rickford-1999e-Phonological-and-Grammatical-Features-of-AAVE.pdf) ways. It was fine-tuned on data gathered from Twitter, and is thus ingrained with what linguists call 'register bias'. |
| |
|
| | ## Training and evaluation data |
| |
|
| | Code hosted at [GitHub](https://github.com/DrewGalbraith/AAE-PoS/tree/main). |
| |
|
| | ## Training procedure |
| |
|
| | ### Training hyperparameters |
| |
|
| | The following hyperparameters were used during training: |
| | - learning_rate: 2e-05 |
| | - train_batch_size: 8 |
| | - eval_batch_size: 8 |
| | - seed: 42 |
| | - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
| | - lr_scheduler_type: linear |
| | - num_epochs: 3 (this amount of data overfits on 3+) |
| |
|
| | ### Training results |
| |
|
| | | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
| | |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
| | | No log | 1.0 | 223 | 0.2982 | 0.8196 | 0.8350 | 0.8272 | 0.9216 | |
| | | No log | 2.0 | 446 | 0.2625 | 0.8599 | 0.8680 | 0.8640 | 0.9326 | |
| | | 0.4647 | 3.0 | 669 | 0.2582 | 0.8632 | 0.8730 | 0.8681 | 0.9356 | |
| |
|
| |
|
| | ### Framework versions |
| |
|
| | - Transformers 4.29.2 |
| | - Pytorch 1.13.1+cpu |
| | - Datasets 2.12.0 |
| | - Tokenizers 0.13.3 |
