hushem_1x_deit_tiny_adamax_lr00001_fold3

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

• Loss: 1.0802
• Accuracy: 0.4651

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: 1e-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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.67	1	1.5307	0.2093
No log	2.0	3	1.3769	0.3023
No log	2.67	4	1.3327	0.3721
No log	4.0	6	1.2794	0.4419
No log	4.67	7	1.2620	0.4419
No log	6.0	9	1.2352	0.4884
1.4092	6.67	10	1.2244	0.4884
1.4092	8.0	12	1.2093	0.4884
1.4092	8.67	13	1.2029	0.4884
1.4092	10.0	15	1.1956	0.4651
1.4092	10.67	16	1.1914	0.4651
1.4092	12.0	18	1.1838	0.4651
1.4092	12.67	19	1.1805	0.4651
1.1598	14.0	21	1.1690	0.4419
1.1598	14.67	22	1.1624	0.4419
1.1598	16.0	24	1.1483	0.4186
1.1598	16.67	25	1.1431	0.4186
1.1598	18.0	27	1.1284	0.4186
1.1598	18.67	28	1.1216	0.4419
0.9892	20.0	30	1.1096	0.4419
0.9892	20.67	31	1.1035	0.4651
0.9892	22.0	33	1.0952	0.4651
0.9892	22.67	34	1.0922	0.4651
0.9892	24.0	36	1.0880	0.4651
0.9892	24.67	37	1.0863	0.4651
0.9892	26.0	39	1.0835	0.4651
0.8902	26.67	40	1.0825	0.4651
0.8902	28.0	42	1.0818	0.4651
0.8902	28.67	43	1.0817	0.4651
0.8902	30.0	45	1.0810	0.4651
0.8902	30.67	46	1.0810	0.4651
0.8902	32.0	48	1.0805	0.4651
0.8902	32.67	49	1.0803	0.4651
0.8497	33.33	50	1.0802	0.4651

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1