distilbert-base-uncased-finetuned-emotion-detector-from-text

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: 0.1628
• Accuracy: 0.9345
• F1: 0.9347

Model description

This model is trained on english tweets and can classify emotions in text files.

Intended uses & limitations

More information needed

Training and evaluation data

16,000 train samples 2,000 validation samples 2,000 test samples

Training procedure

Finetunning distilbert-base-uncased

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: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.1038	1.0	250	0.1757	0.9325	0.9329
0.094	2.0	500	0.1628	0.9345	0.9347

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1