distilbert-base-uncased-finetuned-emotion

Model description

This model is a fine-tuned version of distilbert-base-uncased on emotion dataset. The dataset consists of 1658616 examples of text and emotion labels. The emotion labels are one of the following: 'joy', 'sadness', 'fear', 'anger', 'surprise'. The model was fine-tuned using the Trainer API from the transformers library on the emotion dataset.

Intended uses & limitations

This model is intended to be used for emotion classification tasks. The model is trained on English text data and may not perform well on other languages. The model is trained on the emotion dataset and may not perform well on other emotion classification tasks.

Training and evaluation data

The model was trained on the emotion dataset. The dataset consists of 1658616 examples of text and emotion labels. The emotion labels are one of the following: 'joy', 'sadness', 'fear', 'anger', 'surprise'.

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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.7763	1.0	250	0.2839	0.9145	0.9141
0.2336	2.0	500	0.2054	0.931	0.9309

It achieves the following results on the evaluation set:

• Loss: 0.2054
• Accuracy: 0.931
• F1: 0.9309

Framework versions

• Transformers 4.48.0
• Pytorch 2.5.1+cpu
• Datasets 3.2.0
• Tokenizers 0.21.0