README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - emotion
metrics:
  - f1
  - accuracy
model-index:
  - name: distilbert-base-uncased-fine-tuned-by-emotion
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: emotion
          type: emotion
          config: default
          split: train
          args: default
        metrics:
          - name: F1
            type: f1
            value: 0.9345054748683583
          - name: Accuracy
            type: accuracy
            value: 0.9345

distilbert-base-uncased-fine-tuned-by-emotion

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

• Loss: 0.1678
• F1: 0.9345
• Accuracy: 0.9345

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Accuracy
0.4805	1.0	1000	0.2051	0.9250	0.9245
0.1541	2.0	2000	0.1678	0.9345	0.9345

Framework versions

• Transformers 4.21.2
• Pytorch 1.13.0.dev20220824
• Datasets 2.4.0
• Tokenizers 0.12.1