protein-classification

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.3403
• Accuracy: 0.905

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: 10
• eval_batch_size: 4
• 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
No log	1.0	180	0.4654	0.855
No log	2.0	360	0.4703	0.835
0.521	3.0	540	0.3485	0.88
0.521	4.0	720	0.3523	0.865
0.521	5.0	900	0.3411	0.89
0.3122	6.0	1080	0.2946	0.91
0.3122	7.0	1260	0.3441	0.865
0.3122	8.0	1440	0.2862	0.915
0.2774	9.0	1620	0.3051	0.9
0.2774	10.0	1800	0.3358	0.895
0.2774	11.0	1980	0.3127	0.915
0.2688	12.0	2160	0.3054	0.895
0.2688	13.0	2340	0.3246	0.89
0.2316	14.0	2520	0.3374	0.91
0.2316	15.0	2700	0.4155	0.875
0.2316	16.0	2880	0.3715	0.885
0.2066	17.0	3060	0.3345	0.91
0.2066	18.0	3240	0.3400	0.9
0.2066	19.0	3420	0.3190	0.9
0.1893	20.0	3600	0.3084	0.915
0.1893	21.0	3780	0.3875	0.89
0.1893	22.0	3960	0.3599	0.89
0.1835	23.0	4140	0.3633	0.905
0.1835	24.0	4320	0.2984	0.905
0.1665	25.0	4500	0.3005	0.91
0.1665	26.0	4680	0.3371	0.915
0.1665	27.0	4860	0.3860	0.89
0.1627	28.0	5040	0.3110	0.915
0.1627	29.0	5220	0.3365	0.905
0.1627	30.0	5400	0.3403	0.905

Framework versions

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