|
|
--- |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
datasets: |
|
|
- squad_v2 |
|
|
model-index: |
|
|
- name: extractive_reader_nq_squad_v2 |
|
|
results: [] |
|
|
language: |
|
|
- en |
|
|
--- |
|
|
|
|
|
<!-- 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. --> |
|
|
|
|
|
# extractive_reader_nq_squad_v2 |
|
|
|
|
|
This model is a fine-tuned version of [ToluClassics/extractive_reader_nq](https://huggingface.co/ToluClassics/extractive_reader_nq) on the squad_v2 dataset. |
|
|
|
|
|
## Model description |
|
|
|
|
|
More information needed |
|
|
|
|
|
## Intended uses & limitations |
|
|
|
|
|
More information needed |
|
|
|
|
|
## Training and evaluation data |
|
|
|
|
|
More information needed |
|
|
|
|
|
## Training procedure |
|
|
|
|
|
### Training hyperparameters |
|
|
|
|
|
The following hyperparameters were used during training: |
|
|
- learning_rate: 3e-05 |
|
|
- train_batch_size: 64 |
|
|
- eval_batch_size: 64 |
|
|
- seed: 42 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- num_epochs: 5.0 |
|
|
|
|
|
### Training results |
|
|
|
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.26.0 |
|
|
- Pytorch 1.13.1+cu117 |
|
|
- Datasets 2.8.0 |
|
|
- Tokenizers 0.13.2 |
|
|
|
|
|
|
|
|
### Code Examples |
|
|
|
|
|
```python |
|
|
import torch |
|
|
import numpy as np |
|
|
from transformers import AutoTokenizer, AutoModelForQuestionAnswering |
|
|
|
|
|
tokenizer = AutoTokenizer.from_pretrained("ToluClassics/extractive_reader_nq_squad_v2") |
|
|
|
|
|
model = AutoModelForQuestionAnswering.from_pretrained("ToluClassics/extractive_reader_nq_squad_v2") |
|
|
|
|
|
question = "" |
|
|
context = "" |
|
|
|
|
|
inputs = tokenizer.encode(question, context, add_special_tokens=True, return_tensors="pt") |
|
|
|
|
|
output = model(inputs) |
|
|
|
|
|
answer_start = torch.argmax(output.start_logits) |
|
|
answer_end = torch.argmax(output.end_logits) |
|
|
if answer_end >= answer_start: |
|
|
print(tokenizer.decode(inputs[0][answer_start:answer_end+1])) |
|
|
``` |
