song-coherency-classifier-v2

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

• Loss: 0.1341
• F1: [0.9784946236559139, 0.9789473684210526]

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	190	0.0924	[0.9760000000000001, 0.9761273209549072]
No log	2.0	380	0.0926	[0.9754768392370572, 0.9766233766233766]
0.1717	3.0	570	0.0825	[0.9810298102981029, 0.9817232375979111]
0.1717	4.0	760	0.0892	[0.9813333333333334, 0.9814323607427056]
0.1717	5.0	950	0.0788	[0.9838709677419355, 0.9842105263157895]
0.0737	6.0	1140	0.1032	[0.9813333333333334, 0.9814323607427056]
0.0737	7.0	1330	0.1212	[0.9783783783783783, 0.9790575916230367]
0.0538	8.0	1520	0.1010	[0.9786096256684492, 0.9788359788359788]
0.0538	9.0	1710	0.1186	[0.9811320754716981, 0.9816272965879265]
0.0538	10.0	1900	0.1341	[0.9784946236559139, 0.9789473684210526]

Framework versions

• Transformers 4.40.0
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1