bert-emotion-classifier

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

• Loss: 0.1751
• Accuracy: 0.93
• Precision: 0.9296
• Recall: 0.93
• F1: 0.9295

Model description

This is my first huggingface push trial. Trained the bert model for emotion classification fune-tuning.

Intended uses & limitations

Please dont use this model for serious work. This is just for my own learning.

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: 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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.2447	1.0	1000	0.1953	0.9265	0.9317	0.9265	0.9279
0.1398	2.0	2000	0.1690	0.93	0.9298	0.93	0.9288
0.0873	3.0	3000	0.1751	0.93	0.9296	0.93	0.9295

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1