File size: 1,219 Bytes
bd779f8 68a014f bd779f8 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 |
---
language:
- ms
datasets:
- squad_v2
metrics:
- exact_match
- f1
---
# Overview
This model is an experiment I and my friend did as a researcher internship at the National University of Singapore (NUS). We finetuned the model to our datasets in Finance and Healthcare domain, in the Malay Language.
# Details
- Finetuned from the base model by [zhufy](https://huggingface.co/zhufy/squad-ms-bert-base)
- The base datasets from [SQuAD2.0](https://rajpurkar.github.io/SQuAD-explorer/)
- Our [datasets](https://ids.nus.edu.sg/microsites/nzsg-nlp/datahub.html) in Finance and Healthcare domain
# Finetuned Detail
```py
from transformers import TrainingArguments
training_args = TrainingArguments(
output_dir='test_trainer',
evaluation_strategy='epoch',
num_train_epochs=20,
optim='adamw_torch',
report_to='all',
logging_steps=1,
)
```
# How to use the Model
```py
from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline
model_name = "primasr/malaybert-for-eqa-finetuned"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForQuestionAnswering.from_pretrained(model_name)
nlp = pipeline("question-answering", model=model, tokenizer=tokenizer)
``` |