hushem_1x_deit_small_rms_001_fold2

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

• Loss: 1.2524
• Accuracy: 0.4

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: 0.001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• 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	1.0	6	7.5665	0.2444
4.6598	2.0	12	1.8034	0.2444
4.6598	3.0	18	1.7719	0.2444
1.754	4.0	24	1.5619	0.2667
1.5561	5.0	30	1.5155	0.2444
1.5561	6.0	36	1.5905	0.2444
1.5161	7.0	42	1.4606	0.2444
1.5161	8.0	48	1.5057	0.2667
1.4837	9.0	54	1.4997	0.2444
1.456	10.0	60	1.4411	0.2444
1.456	11.0	66	1.4980	0.2667
1.4256	12.0	72	1.4097	0.2444
1.4256	13.0	78	1.4518	0.2667
1.4488	14.0	84	1.3937	0.2667
1.4354	15.0	90	1.4044	0.2444
1.4354	16.0	96	1.3767	0.2667
1.4383	17.0	102	1.4222	0.2444
1.4383	18.0	108	1.4806	0.2444
1.4107	19.0	114	1.4789	0.2444
1.3761	20.0	120	1.2485	0.4444
1.3761	21.0	126	1.3600	0.2667
1.3385	22.0	132	1.4500	0.4
1.3385	23.0	138	1.3814	0.3778
1.3465	24.0	144	1.4692	0.2667
1.323	25.0	150	1.1674	0.4667
1.323	26.0	156	1.3636	0.2889
1.2871	27.0	162	1.3963	0.4
1.2871	28.0	168	1.3023	0.4444
1.1938	29.0	174	1.2034	0.4222
1.2252	30.0	180	1.2237	0.4444
1.2252	31.0	186	1.2906	0.4
1.2127	32.0	192	1.2853	0.4
1.2127	33.0	198	1.3006	0.3556
1.131	34.0	204	1.3803	0.2889
1.1689	35.0	210	1.2981	0.3556
1.1689	36.0	216	1.4728	0.2889
1.1285	37.0	222	1.3455	0.3333
1.1285	38.0	228	1.2593	0.4
1.0174	39.0	234	1.2539	0.3556
1.0651	40.0	240	1.2296	0.4
1.0651	41.0	246	1.2510	0.3778
1.0297	42.0	252	1.2524	0.4
1.0297	43.0	258	1.2524	0.4
0.9982	44.0	264	1.2524	0.4
1.047	45.0	270	1.2524	0.4
1.047	46.0	276	1.2524	0.4
0.9969	47.0	282	1.2524	0.4
0.9969	48.0	288	1.2524	0.4
1.0686	49.0	294	1.2524	0.4
1.0034	50.0	300	1.2524	0.4

Framework versions

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