ProteinClassificationModel

This model is a fine-tuned version of google/vit-base-patch16-224 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5405
• Accuracy: 0.9

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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.7133	1.0	113	0.3957	0.84
0.3139	2.0	226	0.3768	0.845
0.2595	3.0	339	0.3597	0.89
0.2337	4.0	452	0.3059	0.9
0.2152	5.0	565	0.3772	0.88
0.147	6.0	678	0.4119	0.885
0.1592	7.0	791	0.3554	0.895
0.1162	8.0	904	0.3889	0.88
0.1105	9.0	1017	0.5130	0.855
0.0968	10.0	1130	0.4606	0.87
0.1197	11.0	1243	0.4978	0.88
0.1103	12.0	1356	0.4679	0.885
0.089	13.0	1469	0.4860	0.885
0.0696	14.0	1582	0.5217	0.87
0.0667	15.0	1695	0.5027	0.885
0.0647	16.0	1808	0.4870	0.89
0.0577	17.0	1921	0.4739	0.875
0.0475	18.0	2034	0.4714	0.885
0.0479	19.0	2147	0.4766	0.875
0.0563	20.0	2260	0.5175	0.89
0.0356	21.0	2373	0.5672	0.9
0.052	22.0	2486	0.4877	0.895
0.0385	23.0	2599	0.5096	0.895
0.0321	24.0	2712	0.4934	0.905
0.0295	25.0	2825	0.4869	0.9
0.0347	26.0	2938	0.5327	0.905
0.025	27.0	3051	0.5653	0.9
0.0382	28.0	3164	0.5227	0.9
0.0167	29.0	3277	0.5447	0.9
0.0254	30.0	3390	0.5405	0.9

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.0
• Tokenizers 0.15.0