mobilebert_qqp / README.md
lewtun's picture
lewtun HF staff
Add evaluation results on the qqp config of glue
a93c799
|
raw
history blame
2.22 kB
metadata
language:
  - en
license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
  - f1
model-index:
  - name: qqp
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE QQP
          type: glue
          args: qqp
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8988869651249073
          - name: F1
            type: f1
            value: 0.8670050100852366
      - task:
          type: natural-language-inference
          name: Natural Language Inference
        dataset:
          name: glue
          type: glue
          config: qqp
          split: validation
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8989859015582489
            verified: true
          - name: Precision
            type: precision
            value: 0.8407470502870844
            verified: true
          - name: Recall
            type: recall
            value: 0.8951965065502183
            verified: true
          - name: AUC
            type: auc
            value: 0.9590670523994457
            verified: true
          - name: F1
            type: f1
            value: 0.8671178499381792
            verified: true
          - name: loss
            type: loss
            value: 0.2457672506570816
            verified: true

qqp

This model is a fine-tuned version of google/mobilebert-uncased on the GLUE QQP dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2458
  • Accuracy: 0.8989
  • F1: 0.8670
  • Combined Score: 0.8829

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: 5e-05
  • train_batch_size: 64
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 2.5

Training results

Framework versions

  • Transformers 4.20.0.dev0
  • Pytorch 1.11.0
  • Datasets 2.2.2
  • Tokenizers 0.12.1