Edit model card

Auto Question Generation

The model is intended to be used for Auto Question Generation task i.e. no hint are required as input. The model is expected to produce one or possibly more than one question from the provided context.

Live Demo: Question Generation

Including this there are four models trained with different training sets, demo provide comparison to all in one go. However, you can reach individual projects at below links:

Auto Question Generation v1

Auto Question Generation v2

Auto/Hints based Question Generation v1

Auto/Hints based Question Generation v2

This model can be used as below:

from transformers import (
    AutoModelForSeq2SeqLM,
    AutoTokenizer
)

model_checkpoint = "consciousAI/question-generation-auto-t5-v1-base-s-q-c"

model = AutoModelForSeq2SeqLM.from_pretrained(model_checkpoint)
tokenizer = AutoTokenizer.from_pretrained(model_checkpoint)

## Input with prompt
context="question_context: <context>"
encodings = tokenizer.encode(context, return_tensors='pt', truncation=True, padding='max_length').to(device)

## You can play with many hyperparams to condition the output, look at demo
output = model.generate(encodings, 
                        #max_length=300, 
                        #min_length=20, 
                        #length_penalty=2.0, 
                        num_beams=4,
                        #early_stopping=True,
                        #do_sample=True,
                        #temperature=1.1
                       )

## Multiple questions are expected to be delimited by '?' You can write a small wrapper to elegantly format. Look at the demo.
questions = [tokenizer.decode(id, clean_up_tokenization_spaces=False, skip_special_tokens=False) for id in output]

Training and evaluation data

Squad & QNLi combo.

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.0003
  • train_batch_size: 6
  • eval_batch_size: 6
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss Rouge1 Rouge2 Rougel Rougelsum
1.8732 1.0 942 1.4330 0.5558 0.3857 0.5219 0.5231
1.2361 2.0 1884 1.3720 0.5649 0.4042 0.5332 0.5338
0.9515 3.0 2826 1.3703 0.5699 0.4044 0.5373 0.5385
0.7383 4.0 3768 1.4039 0.5753 0.4159 0.5414 0.5426
0.6291 5.0 4710 1.4661 0.5809 0.4227 0.5488 0.5498

Framework versions

  • Transformers 4.23.0.dev0
  • Pytorch 1.12.1+cu113
  • Datasets 2.5.2
  • Tokenizers 0.13.0
Downloads last month
10
Safetensors
Model size
248M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Space using consciousAI/question-generation-auto-t5-v1-base-s-q-c 1