classify-sher

This model is a fine-tuned version of mufaddal-k/classify-sher on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6875
• Accuracy: 0.7512
• F1: 0.7511
• Precision: 0.7520
• Recall: 0.7512
• Accuracy Label Classical: 0.7578
• Accuracy Label Contemporary: 0.8101
• Accuracy Label Modernist: 0.6776

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
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 3357

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall	Accuracy Label Classical	Accuracy Label Contemporary	Accuracy Label Modernist
0.3859	0.2234	500	0.7063	0.7195	0.7195	0.7195	0.7195	0.7609	0.7568	0.6441
0.4271	0.4468	1000	0.6870	0.7265	0.7218	0.7238	0.7265	0.7694	0.8353	0.5666
0.4572	0.6702	1500	0.6906	0.7301	0.7316	0.7347	0.7301	0.7336	0.7607	0.6918
0.4384	0.8937	2000	0.6800	0.7369	0.7385	0.7420	0.7369	0.7454	0.7521	0.7126
0.3364	1.1171	2500	0.7259	0.7452	0.7438	0.7469	0.7452	0.7313	0.8399	0.6461
0.333	1.3405	3000	0.6875	0.7512	0.7511	0.7520	0.7512	0.7578	0.8101	0.6776

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1