vit-base-patch16-224-Trial008-YEL_STEM

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

• Loss: 0.0593
• Accuracy: 1.0

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: 5e-05
• train_batch_size: 60
• eval_batch_size: 60
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 240
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.544	1.0	1	0.8179	0.4118
0.3416	2.0	3	0.7448	0.5294
0.1412	3.0	5	0.7606	0.5294
0.4868	4.0	6	0.5647	0.6471
0.3852	5.0	7	0.4646	0.8235
0.284	6.0	9	0.4300	0.8235
0.1075	7.0	11	0.4628	0.8235
0.3243	8.0	12	0.4687	0.7647
0.3317	9.0	13	0.4089	0.8235
0.146	10.0	15	0.3330	0.8824
0.0762	11.0	17	0.2941	0.8824
0.2351	12.0	18	0.3217	0.8824
0.2458	13.0	19	0.3705	0.8824
0.1431	14.0	21	0.3138	0.8824
0.0883	15.0	23	0.1510	0.9412
0.1601	16.0	24	0.1373	0.9412
0.2212	17.0	25	0.1175	0.9412
0.1311	18.0	27	0.1130	0.9412
0.0801	19.0	29	0.1506	0.9412
0.1857	20.0	30	0.1272	0.9412
0.241	21.0	31	0.0974	0.9412
0.1098	22.0	33	0.0593	1.0
0.0464	23.0	35	0.0574	1.0
0.1757	24.0	36	0.0554	1.0
0.1992	25.0	37	0.0605	1.0
0.1167	26.0	39	0.0818	0.9412
0.0703	27.0	41	0.1177	0.9412
0.1382	28.0	42	0.1281	0.9412
0.1563	29.0	43	0.1357	0.9412
0.1113	30.0	45	0.1417	0.8824
0.0639	31.0	47	0.1250	0.9412
0.1564	32.0	48	0.1107	0.9412
0.1877	33.0	49	0.1002	0.9412
0.06	33.33	50	0.0958	0.9412

Framework versions

• Transformers 4.30.0.dev0
• Pytorch 1.12.1
• Datasets 2.12.0
• Tokenizers 0.13.1