README.md

metadata

language:
  - en
license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
  - f1
model-index:
  - name: bert-base-uncased-qqp
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE QQP
          type: glue
          args: qqp
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9099925797674994
          - name: F1
            type: f1
            value: 0.8788252139455897
      - task:
          type: natural-language-inference
          name: Natural Language Inference
        dataset:
          name: glue
          type: glue
          config: qqp
          split: validation
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9099925797674994
            verified: true
          - name: Precision
            type: precision
            value: 0.8712531361415555
            verified: true
          - name: Recall
            type: recall
            value: 0.8865300638226402
            verified: true
          - name: AUC
            type: auc
            value: 0.9690747048570257
            verified: true
          - name: F1
            type: f1
            value: 0.8788252139455897
            verified: true
          - name: loss
            type: loss
            value: 0.28284332156181335
            verified: true

bert-base-uncased-qqp

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

• Loss: 0.2829
• Accuracy: 0.9100
• F1: 0.8788
• Combined Score: 0.8944

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: 2e-05
• train_batch_size: 32
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Combined Score
0.2511	1.0	11371	0.2469	0.8969	0.8641	0.8805
0.1763	2.0	22742	0.2379	0.9071	0.8769	0.8920
0.1221	3.0	34113	0.2829	0.9100	0.8788	0.8944

Framework versions

• Transformers 4.20.0.dev0
• Pytorch 1.11.0+cu113
• Datasets 2.1.0
• Tokenizers 0.12.1