metadata

license: mit
tags:
  - generated_from_trainer
datasets:
  - tweets_hate_speech_detection
metrics:
  - accuracy
  - f1
model-index:
  - name: FirstTry
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: tweets_hate_speech_detection
          type: tweets_hate_speech_detection
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9821679962458939
          - name: F1
            type: f1
            value: 0.8692660550458716

FirstTry

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

Loss: 0.0847
Accuracy: 0.9822
F1: 0.8693

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.1159	1.0	1599	0.1019	0.9759	0.8270
0.0727	2.0	3198	0.0965	0.9795	0.8424
0.044	3.0	4797	0.0847	0.9822	0.8693
0.0301	4.0	6396	0.1121	0.9811	0.8660
0.0206	5.0	7995	0.1718	0.9700	0.8110
0.0176	6.0	9594	0.1453	0.9811	0.8591

Framework versions

Transformers 4.26.1
Pytorch 2.0.1
Datasets 2.10.1
Tokenizers 0.13.3