groove_midi

This model is a fine-tuned version of google-t5/t5-small on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3579
• Bleu: 0.0
• Gen Len: 20.0

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: 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: 40

Training results

Training Loss	Epoch	Step	Validation Loss	Bleu	Gen Len
No log	1.0	57	0.4015	0.0	20.0
No log	2.0	114	0.3978	0.0	20.0
No log	3.0	171	0.3930	0.0	20.0
No log	4.0	228	0.3895	0.0	20.0
No log	5.0	285	0.3871	0.0	20.0
No log	6.0	342	0.3858	0.0	20.0
No log	7.0	399	0.3805	0.0	20.0
No log	8.0	456	0.3795	0.0	20.0
0.443	9.0	513	0.3781	0.0	20.0
0.443	10.0	570	0.3762	0.0	20.0
0.443	11.0	627	0.3745	0.0	20.0
0.443	12.0	684	0.3724	0.0	20.0
0.443	13.0	741	0.3715	0.0	20.0
0.443	14.0	798	0.3696	0.0	20.0
0.443	15.0	855	0.3695	0.0	20.0
0.443	16.0	912	0.3676	0.0	20.0
0.443	17.0	969	0.3671	0.0	20.0
0.4172	18.0	1026	0.3657	0.0	20.0
0.4172	19.0	1083	0.3649	0.0	20.0
0.4172	20.0	1140	0.3641	0.0	20.0
0.4172	21.0	1197	0.3641	0.0	20.0
0.4172	22.0	1254	0.3643	0.0	20.0
0.4172	23.0	1311	0.3621	0.0	20.0
0.4172	24.0	1368	0.3616	0.0	20.0
0.4172	25.0	1425	0.3614	0.0	20.0
0.4172	26.0	1482	0.3612	0.0	20.0
0.406	27.0	1539	0.3603	0.0	20.0
0.406	28.0	1596	0.3606	0.0	20.0
0.406	29.0	1653	0.3600	0.0	20.0
0.406	30.0	1710	0.3599	0.0	20.0
0.406	31.0	1767	0.3592	0.0	20.0
0.406	32.0	1824	0.3588	0.0	20.0
0.406	33.0	1881	0.3582	0.0	20.0
0.406	34.0	1938	0.3585	0.0	20.0
0.406	35.0	1995	0.3583	0.0	20.0
0.401	36.0	2052	0.3580	0.0	20.0
0.401	37.0	2109	0.3578	0.0	20.0
0.401	38.0	2166	0.3579	0.0	20.0
0.401	39.0	2223	0.3578	0.0	20.0
0.401	40.0	2280	0.3579	0.0	20.0

Framework versions

• Transformers 4.48.2
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0