distilbert-base-uncased-finetuned-emotions

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

• Loss: 0.1970
• Accuracy: 0.9425
• F1: 0.9426

Model description

distilbert-base-uncased base model finetuned on dair-ai/emotion dataset.

Intended uses

Can be used to classify emotions, sentiment analysis

Labels Mapping

0: Sadness 😢

1: Joy 😊

2: Love ❤️

3: Anger 😠

4: Fear 😨

5: Surprise 😲

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.0567	1.0	250	0.2055	0.9345	0.9348
0.0559	2.0	500	0.2037	0.9355	0.9358
0.067	3.0	750	0.1970	0.9425	0.9426

Framework versions

• Transformers 4.56.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.0