vit-large-patch16-224-finetuned-landscape-test

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

• Loss: 0.3101
• Accuracy: 0.9094

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.3906	1.0	10	1.1521	0.4969
0.914	2.0	20	0.7812	0.6687
0.6704	3.0	30	0.5566	0.7688
0.4609	4.0	40	0.4363	0.8313
0.404	5.0	50	0.4807	0.8156
0.3948	6.0	60	0.4216	0.8531
0.3535	7.0	70	0.3281	0.8688
0.3107	8.0	80	0.2972	0.9
0.3086	9.0	90	0.3328	0.8812
0.2564	10.0	100	0.3517	0.8875
0.2654	11.0	110	0.3985	0.8594
0.2733	12.0	120	0.2870	0.9062
0.2511	13.0	130	0.4177	0.8875
0.2762	14.0	140	0.3579	0.8938
0.2188	15.0	150	0.3348	0.8906
0.2265	16.0	160	0.3046	0.9031
0.2054	17.0	170	0.3305	0.8969
0.1951	18.0	180	0.3576	0.8812
0.1762	19.0	190	0.3985	0.8812
0.2264	20.0	200	0.3711	0.9031
0.1958	21.0	210	0.3259	0.8875
0.1765	22.0	220	0.3804	0.8938
0.1859	23.0	230	0.3464	0.9
0.1915	24.0	240	0.3742	0.8906
0.1667	25.0	250	0.3200	0.9062
0.1744	26.0	260	0.3545	0.8938
0.1595	27.0	270	0.3101	0.9094
0.1793	28.0	280	0.3230	0.8969
0.1596	29.0	290	0.3268	0.9
0.169	30.0	300	0.3321	0.8969

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1