README.md

metadata

tags:
  - generated_from_trainer
datasets:
  - emotion
metrics:
  - accuracy
  - f1
base_model: distilbert-base-uncased
model-index:
  - name: bertweet-base-finetuned-emotion
    results:
      - task:
          type: text-classification
          name: Text Classification
        dataset:
          name: emotion
          type: emotion
          args: default
        metrics:
          - type: accuracy
            value: 0.9365
            name: Accuracy
          - type: f1
            value: 0.9371
            name: F1
      - task:
          type: text-classification
          name: Text Classification
        dataset:
          name: emotion
          type: emotion
          config: default
          split: test
        metrics:
          - type: accuracy
            value: 0.923
            name: Accuracy
            verified: true
          - type: precision
            value: 0.8676576686813523
            name: Precision Macro
            verified: true
          - type: precision
            value: 0.923
            name: Precision Micro
            verified: true
          - type: precision
            value: 0.9268406401714973
            name: Precision Weighted
            verified: true
          - type: recall
            value: 0.8945488803260702
            name: Recall Macro
            verified: true
          - type: recall
            value: 0.923
            name: Recall Micro
            verified: true
          - type: recall
            value: 0.923
            name: Recall Weighted
            verified: true
          - type: f1
            value: 0.8798961895301041
            name: F1 Macro
            verified: true
          - type: f1
            value: 0.923
            name: F1 Micro
            verified: true
          - type: f1
            value: 0.9241278880972197
            name: F1 Weighted
            verified: true
          - type: loss
            value: 0.24626904726028442
            name: loss
            verified: true

distilbert-base-uncased-finetuned-emotion

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1995
• Accuracy: 0.9365
• F1: 0.9371

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: 64
• eval_batch_size: 64
• 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	Accuracy	F1
0.475	1.0	503	0.2171	0.928	0.9292
0.1235	2.0	1006	0.1764	0.9365	0.9372
0.0802	3.0	1509	0.1788	0.938	0.9388
0.0531	4.0	2012	0.2005	0.938	0.9388
0.0367	5.0	2515	0.1995	0.9365	0.9371

Framework versions

• Transformers 4.13.0
• Pytorch 1.11.0+cu113
• Datasets 1.16.1
• Tokenizers 0.10.3