Foot

This model is a fine-tuned version of facebook/dinov2-base-imagenet1k-1-layer on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0747
• Accuracy: 0.4865

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: 10
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 20
• 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
No log	1.0	2	1.2201	0.3378
No log	2.0	4	1.2243	0.2568
No log	3.0	6	1.2672	0.2703
No log	4.0	8	1.2501	0.2297
1.2006	5.0	10	1.1975	0.2973
1.2006	6.0	12	1.1270	0.3919
1.2006	7.0	14	1.0999	0.3243
1.2006	8.0	16	1.1497	0.3649
1.2006	9.0	18	1.1006	0.3108
1.1058	10.0	20	1.1271	0.3514
1.1058	11.0	22	1.1273	0.3784
1.1058	12.0	24	1.1639	0.2838
1.1058	13.0	26	1.1421	0.4054
1.1058	14.0	28	1.1190	0.3514
1.0489	15.0	30	1.1735	0.3243
1.0489	16.0	32	1.1422	0.3378
1.0489	17.0	34	1.1414	0.3649
1.0489	18.0	36	1.1033	0.4189
1.0489	19.0	38	1.0747	0.3919
0.9717	20.0	40	1.0952	0.3919
0.9717	21.0	42	1.1063	0.3784
0.9717	22.0	44	1.0822	0.3649
0.9717	23.0	46	1.0768	0.3784
0.9717	24.0	48	1.0753	0.4595
0.9816	25.0	50	1.0531	0.4054
0.9816	26.0	52	1.0624	0.4189
0.9816	27.0	54	1.0690	0.4459
0.9816	28.0	56	1.1392	0.3514
0.9816	29.0	58	1.0696	0.4054
0.9576	30.0	60	1.0747	0.4865

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.15.0
• Tokenizers 0.15.0