mert_cmp_single

This model is a fine-tuned version of m-a-p/MERT-v0 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.5373
• Accuracy: 0.4588
• Precision Micro: 0.4588
• Recall Micro: 0.4588
• F1 Micro: 0.4588
• Precision Macro: 0.2915
• Recall Macro: 0.3266
• F1 Macro: 0.2877
• Precision Weighted: 0.3892
• Recall Weighted: 0.4588
• F1 Weighted: 0.4111

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: 0.005
• 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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision Micro	Recall Micro	F1 Micro	Precision Macro	Recall Macro	F1 Macro	Precision Weighted	Recall Weighted	F1 Weighted
1.9863	1.0	167	1.8525	0.3421	0.3421	0.3421	0.3421	0.1125	0.1591	0.1151	0.2027	0.3421	0.2380
1.825	2.0	334	1.8022	0.3395	0.3395	0.3395	0.3395	0.1190	0.2242	0.1392	0.2482	0.3395	0.2694
1.7777	3.0	501	1.6796	0.3877	0.3877	0.3877	0.3877	0.1842	0.2353	0.1812	0.3123	0.3877	0.3247
1.7034	4.0	668	1.5932	0.4193	0.4193	0.4193	0.4193	0.2826	0.2972	0.2541	0.3787	0.4193	0.3767
1.5985	5.0	835	1.5373	0.4588	0.4588	0.4588	0.4588	0.2915	0.3266	0.2877	0.3892	0.4588	0.4111

Framework versions

• Transformers 4.44.0
• Pytorch 2.3.1+cu121
• Datasets 3.6.0
• Tokenizers 0.19.1