README.md

metadata

license: mit
tags:
  - generated_from_trainer
datasets:
  - squad_v2
  - quoref
  - adversarial_qa
  - ibm/duorc
task:
  - question-answering
model-index:
  - name: rob-base-superqa
    results:
      - task:
          type: question-answering
          name: Question Answering
        dataset:
          name: adversarial_qa
          type: adversarial_qa
          config: adversarialQA
          split: validation
        metrics:
          - type: exact_match
            value: 43.8667
            name: Exact Match
            verified: true
          - type: f1
            value: 55.135
            name: F1
            verified: true
      - task:
          type: question-answering
          name: Question Answering
        dataset:
          name: squad_v2
          type: squad_v2
          config: squad_v2
          split: validation
        metrics:
          - type: exact_match
            value: 79.2432
            name: Exact Match
            verified: true
          - type: f1
            value: 82.336
            name: F1
            verified: true
      - task:
          type: question-answering
          name: Question Answering
        dataset:
          name: quoref
          type: quoref
          config: default
          split: validation
        metrics:
          - type: exact_match
            value: 78.8581
            name: Exact Match
            verified: true
          - type: f1
            value: 82.8261
            name: F1
            verified: true

rob-base-superqa

This model is a fine-tuned version of roberta-base on the None 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: 7e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 256
• total_eval_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-06
• lr_scheduler_type: linear
• num_epochs: 3.0

Training results

Framework versions

• Transformers 4.21.1
• Pytorch 1.11.0a0+gita4c10ee
• Datasets 2.4.0
• Tokenizers 0.12.1