vit-base-patch16-224-Trial008-YEL_STEM3

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: 0.0916
• 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: 30
• eval_batch_size: 30
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 120
• 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
0.7743	1.0	1	0.8267	0.3636
0.7964	2.0	2	0.7547	0.3636
0.6369	3.0	3	0.6399	0.7273
0.5344	4.0	4	0.5082	0.9091
0.4342	5.0	5	0.4664	0.9091
0.3056	6.0	6	0.2145	0.9091
0.257	7.0	7	0.1395	0.9091
0.2064	8.0	8	0.1990	0.9091
0.2609	9.0	9	0.0916	1.0
0.1758	10.0	10	0.0321	1.0
0.1152	11.0	11	0.0256	1.0
0.1343	12.0	12	0.0413	1.0
0.0955	13.0	13	0.0319	1.0
0.0723	14.0	14	0.0112	1.0
0.13	15.0	15	0.0073	1.0
0.1918	16.0	16	0.0057	1.0
0.2469	17.0	17	0.0052	1.0
0.1001	18.0	18	0.0051	1.0
0.1331	19.0	19	0.0039	1.0
0.1511	20.0	20	0.0031	1.0
0.0956	21.0	21	0.0027	1.0
0.0952	22.0	22	0.0027	1.0
0.1679	23.0	23	0.0025	1.0
0.1075	24.0	24	0.0023	1.0
0.1507	25.0	25	0.0024	1.0
0.1267	26.0	26	0.0027	1.0
0.1141	27.0	27	0.0030	1.0
0.0767	28.0	28	0.0031	1.0
0.1746	29.0	29	0.0029	1.0
0.1101	30.0	30	0.0032	1.0
0.1632	31.0	31	0.0036	1.0
0.1346	32.0	32	0.0038	1.0
0.1024	33.0	33	0.0038	1.0
0.1198	34.0	34	0.0037	1.0
0.1217	35.0	35	0.0033	1.0
0.1433	36.0	36	0.0030	1.0
0.1255	37.0	37	0.0029	1.0
0.1369	38.0	38	0.0027	1.0
0.091	39.0	39	0.0026	1.0
0.1318	40.0	40	0.0025	1.0
0.0964	41.0	41	0.0025	1.0
0.1164	42.0	42	0.0024	1.0
0.0935	43.0	43	0.0023	1.0
0.0564	44.0	44	0.0022	1.0
0.1136	45.0	45	0.0021	1.0
0.1306	46.0	46	0.0021	1.0
0.0757	47.0	47	0.0021	1.0
0.0475	48.0	48	0.0020	1.0
0.1455	49.0	49	0.0020	1.0
0.1892	50.0	50	0.0020	1.0

Framework versions

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