vit-base-patch16-224-Trial008-YEL_STEM1

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: 0.1847
• 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
0.5774	1.0	1	0.6707	0.5294
0.3598	2.0	3	0.5525	0.8235
0.1477	3.0	5	0.3968	0.9412
0.3936	4.0	6	0.4487	0.6471
0.3762	5.0	7	0.3782	0.7647
0.2275	6.0	9	0.1847	1.0
0.099	7.0	11	0.2053	0.9412
0.2703	8.0	12	0.1943	0.9412
0.2363	9.0	13	0.1228	0.9412
0.1336	10.0	15	0.0758	1.0
0.0772	11.0	17	0.0553	1.0
0.2227	12.0	18	0.0449	1.0
0.1975	13.0	19	0.0417	1.0
0.1401	14.0	21	0.0391	1.0
0.0541	15.0	23	0.0216	1.0
0.1465	16.0	24	0.0232	1.0
0.1583	17.0	25	0.0274	1.0
0.1226	18.0	27	0.0372	1.0
0.0826	19.0	29	0.0333	1.0
0.1634	20.0	30	0.0219	1.0
0.1904	21.0	31	0.0135	1.0
0.0755	22.0	33	0.0080	1.0
0.055	23.0	35	0.0071	1.0
0.1598	24.0	36	0.0072	1.0
0.1805	25.0	37	0.0068	1.0
0.1093	26.0	39	0.0062	1.0
0.0446	27.0	41	0.0061	1.0
0.1377	28.0	42	0.0062	1.0
0.1474	29.0	43	0.0063	1.0
0.0817	30.0	45	0.0066	1.0
0.0527	31.0	47	0.0067	1.0
0.1161	32.0	48	0.0067	1.0
0.1972	33.0	49	0.0067	1.0
0.0708	33.33	50	0.0067	1.0

Framework versions

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