distilbert-scoring-on-off

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3140
• Accuracy: 0.9248
• Precision: 0.9232
• Recall: 0.9248
• F1: 0.9236

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.4479	0.3083	500	0.2908	0.9084	0.9118	0.9084	0.9091
0.3248	0.6165	1000	0.2770	0.9060	0.9027	0.9060	0.8983
0.3051	0.9248	1500	0.2426	0.9229	0.9207	0.9229	0.9207
0.2634	1.2330	2000	0.3114	0.9072	0.9047	0.9072	0.9056
0.2296	1.5413	2500	0.3468	0.9162	0.9142	0.9162	0.9145
0.2347	1.8496	3000	0.2526	0.9254	0.9231	0.9254	0.9231
0.2016	2.1578	3500	0.2777	0.9202	0.9186	0.9202	0.9192
0.1516	2.4661	4000	0.3153	0.9177	0.9200	0.9177	0.9187
0.1553	2.7744	4500	0.3140	0.9248	0.9232	0.9248	0.9236

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.2
• Datasets 2.19.2
• Tokenizers 0.19.1