Hugging Face's logo

research-backup
/
t5-large-subjqa-grocery-qg

Text2Text Generation
Transformers
PyTorch
English
t5
question generation
Eval Results
text-generation-inference
Inference Endpoints
Model card Files Community
t5-large-subjqa-grocery-qg / README.md
asahi417's picture
asahi417
model update
b5589e5
preview code
|
raw
history blame
4.89 kB
metadata
license: cc-by-4.0
metrics:
  - bleu4
  - meteor
  - rouge-l
  - bertscore
  - moverscore
language: en
datasets:
  - lmqg/qg_subjqa
pipeline_tag: text2text-generation
tags:
  - question generation
widget:
  - text: >-
      generate question: <hl> Beyonce <hl> further expanded her acting career,
      starring as blues singer Etta James in the 2008 musical biopic, Cadillac
      Records.
    example_title: Question Generation Example 1
  - text: >-
      generate question: Beyonce further expanded her acting career, starring as
      blues singer <hl> Etta James <hl> in the 2008 musical biopic, Cadillac
      Records.
    example_title: Question Generation Example 2
  - text: >-
      generate question: Beyonce further expanded her acting career, starring as
      blues singer Etta James in the 2008 musical biopic,  <hl> Cadillac Records
      <hl> .
    example_title: Question Generation Example 3
model-index:
  - name: lmqg/t5-large-subjqa-grocery
    results:
      - task:
          name: Text2text Generation
          type: text2text-generation
        dataset:
          name: lmqg/qg_subjqa
          type: grocery
          args: grocery
        metrics:
          - name: BLEU4
            type: bleu4
            value: 0.011335292363312374
          - name: ROUGE-L
            type: rouge-l
            value: 0.1740279794913675
          - name: METEOR
            type: meteor
            value: 0.20641848238590096
          - name: BERTScore
            type: bertscore
            value: 0.9139250615437825
          - name: MoverScore
            type: moverscore
            value: 0.6341318883185333
      - task:
          name: Text2text Generation
          type: text2text-generation
        dataset:
          name: lmqg/qg_squad
          type: default
          args: default
        metrics:
          - name: BLEU4
            type: bleu4
            value: 0.266398028296004
          - name: ROUGE-L
            type: rouge-l
            value: 0.5400055833410796
          - name: METEOR
            type: meteor
            value: 0.26916696517436683
          - name: BERTScore
            type: bertscore
            value: 0.9097899012334792
          - name: MoverScore
            type: moverscore
            value: 0.6514236028343862

Language Models Fine-tuning on Question Generation: lmqg/t5-large-subjqa-grocery

This model is fine-tuned version of lmqg/t5-large-squad for question generation task on the lmqg/qg_subjqa (dataset_name: grocery). This model is continuously fine-tuned with lmqg/t5-large-squad.

Overview

Usage 


from transformers import pipeline

model_path = 'lmqg/t5-large-subjqa-grocery'
pipe = pipeline("text2text-generation", model_path)

# Question Generation
input_text = 'generate question: <hl> Beyonce <hl> further expanded her acting career, starring as blues singer Etta James in the 2008 musical biopic, Cadillac Records.'
question = pipe(input_text)

Evaluation Metrics

Metrics

Dataset Type BLEU4 ROUGE-L METEOR BERTScore MoverScore Link
lmqg/qg_subjqa grocery 0.011335292363312374 0.1740279794913675 0.20641848238590096 0.9139250615437825 0.6341318883185333 link

Out-of-domain Metrics

Dataset Type BLEU4 ROUGE-L METEOR BERTScore MoverScore Link
lmqg/qg_squad default 0.266398028296004 0.5400055833410796 0.26916696517436683 0.9097899012334792 0.6514236028343862 link

Training hyperparameters

The following hyperparameters were used during fine-tuning:

  • dataset_path: lmqg/qg_subjqa
  • dataset_name: grocery
  • input_types: ['paragraph_answer']
  • output_types: ['question']
  • prefix_types: ['qg']
  • model: lmqg/t5-large-squad
  • max_length: 512
  • max_length_output: 32
  • epoch: 3
  • batch: 16
  • lr: 5e-05
  • fp16: False
  • random_seed: 1
  • gradient_accumulation_steps: 32
  • label_smoothing: 0.15

The full configuration can be found at fine-tuning config file.

Citation

TBA