smids_1x_deit_small_sgd_0001_fold3

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: 0.8065
• Accuracy: 0.6733

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: 0.0001
• 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
1.08	1.0	75	1.0697	0.42
1.068	2.0	150	1.0540	0.4567
1.0267	3.0	225	1.0400	0.4817
0.9978	4.0	300	1.0279	0.5117
1.0069	5.0	375	1.0166	0.5167
0.9928	6.0	450	1.0059	0.5333
0.9852	7.0	525	0.9957	0.535
0.9289	8.0	600	0.9859	0.54
0.9567	9.0	675	0.9764	0.5533
0.9517	10.0	750	0.9672	0.5583
0.9256	11.0	825	0.9583	0.565
0.9357	12.0	900	0.9498	0.565
0.9377	13.0	975	0.9415	0.57
0.9369	14.0	1050	0.9334	0.575
0.8836	15.0	1125	0.9259	0.5667
0.922	16.0	1200	0.9184	0.575
0.8518	17.0	1275	0.9112	0.5833
0.8741	18.0	1350	0.9043	0.5883
0.8604	19.0	1425	0.8978	0.5933
0.8822	20.0	1500	0.8914	0.5983
0.8613	21.0	1575	0.8855	0.61
0.8383	22.0	1650	0.8797	0.6117
0.8466	23.0	1725	0.8742	0.615
0.8373	24.0	1800	0.8689	0.6217
0.8379	25.0	1875	0.8638	0.6283
0.8435	26.0	1950	0.8590	0.6217
0.7992	27.0	2025	0.8544	0.6267
0.8242	28.0	2100	0.8500	0.6317
0.8405	29.0	2175	0.8460	0.635
0.8059	30.0	2250	0.8421	0.6383
0.8193	31.0	2325	0.8384	0.6383
0.8107	32.0	2400	0.8351	0.64
0.777	33.0	2475	0.8318	0.6433
0.799	34.0	2550	0.8288	0.6483
0.7972	35.0	2625	0.8260	0.6533
0.8308	36.0	2700	0.8234	0.6533
0.761	37.0	2775	0.8210	0.6567
0.8092	38.0	2850	0.8187	0.6567
0.8047	39.0	2925	0.8167	0.66
0.7661	40.0	3000	0.8149	0.6633
0.7897	41.0	3075	0.8132	0.6633
0.7801	42.0	3150	0.8117	0.665
0.8176	43.0	3225	0.8105	0.6683
0.7701	44.0	3300	0.8093	0.6717
0.8017	45.0	3375	0.8084	0.6717
0.7892	46.0	3450	0.8077	0.6717
0.7778	47.0	3525	0.8071	0.6717
0.7798	48.0	3600	0.8068	0.6733
0.7777	49.0	3675	0.8066	0.6733
0.7532	50.0	3750	0.8065	0.6733

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0