distilbert-base-uncased-lora-text-classification

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

• Loss: 1.2612
• Accuracy: {'accuracy': 0.8287247214197276}

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: 0.001
• train_batch_size: 4
• eval_batch_size: 4
• 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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6087	1.0	606	0.6192	{'accuracy': 0.8018984729673958}
0.4539	2.0	1212	0.5939	{'accuracy': 0.8312009905076352}
0.3768	3.0	1818	0.7302	{'accuracy': 0.8283120099050764}
0.3249	4.0	2424	0.7608	{'accuracy': 0.8287247214197276}
0.1923	5.0	3030	0.8825	{'accuracy': 0.8283120099050764}
0.1518	6.0	3636	1.0603	{'accuracy': 0.8332645480808915}
0.1068	7.0	4242	1.1702	{'accuracy': 0.8262484523318201}
0.0673	8.0	4848	1.2515	{'accuracy': 0.8217086256706562}
0.072	9.0	5454	1.2673	{'accuracy': 0.8303755674783326}
0.0315	10.0	6060	1.2612	{'accuracy': 0.8287247214197276}

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.6.0+cpu
• Datasets 3.5.0
• Tokenizers 0.21.1