|
--- |
|
license: apache-2.0 |
|
tags: |
|
- Question Answering |
|
metrics: |
|
- squad |
|
model-index: |
|
- name: question-answering-roberta-base-s |
|
results: [] |
|
--- |
|
|
|
# Question Answering |
|
The model is intended to be used for Q&A task, given the question & context, the model would attempt to infer the answer text, answer span & confidence score.<br> |
|
Model is encoder-only (roberta-base) with QuestionAnswering LM Head, fine-tuned on SQUADx dataset with **exact_match:** 86.14 & **f1:** 92.330 performance scores. |
|
|
|
[Live Demo: Question Answering Encoders vs Generative](https://huggingface.co/spaces/consciousAI/question_answering) |
|
|
|
Please follow this link for [Encoder based Question Answering V2](https://huggingface.co/consciousAI/question-answering-roberta-base-s-v2/) |
|
<br>Please follow this link for [Generative Question Answering](https://huggingface.co/consciousAI/question-answering-generative-t5-v1-base-s-q-c/) |
|
|
|
Example code: |
|
``` |
|
from transformers import pipeline |
|
|
|
model_checkpoint = "consciousAI/question-answering-roberta-base-s" |
|
|
|
context = """ |
|
🤗 Transformers is backed by the three most popular deep learning libraries — Jax, PyTorch and TensorFlow — with a seamless integration |
|
between them. It's straightforward to train your models with one before loading them for inference with the other. |
|
""" |
|
question = "Which deep learning libraries back 🤗 Transformers?" |
|
|
|
question_answerer = pipeline("question-answering", model=model_checkpoint) |
|
question_answerer(question=question, context=context) |
|
|
|
``` |
|
|
|
## Training and evaluation data |
|
|
|
SQUAD Split |
|
|
|
## Training procedure |
|
|
|
Preprocessing: |
|
1. SQUAD Data longer chunks were sub-chunked with input context max-length 384 tokens and stride as 128 tokens. |
|
2. Target answers readjusted for sub-chunks, sub-chunks with no-answers or partial answers were set to target answer span as (0,0) |
|
|
|
Metrics: |
|
1. Adjusted accordingly to handle sub-chunking. |
|
2. n best = 20 |
|
3. skip answers with length zero or higher than max answer length (30) |
|
|
|
### Training hyperparameters |
|
Custom Training Loop: |
|
The following hyperparameters were used during training: |
|
- learning_rate: 2e-5 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 32 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 2 |
|
|
|
### Training results |
|
|
|
| Epoch | F1 | Exact Match | |
|
|:-----:|:--------:|:-----------:| |
|
| 1.0 | 91.3085 | 84.5412 | |
|
| 2.0 | 92.3304 | 86.1400 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.23.0.dev0 |
|
- Pytorch 1.12.1+cu113 |
|
- Datasets 2.5.2 |
|
- Tokenizers 0.13.0 |
|
|
