smids_3x_deit_small_sgd_0001_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: 0.5927
• Accuracy: 0.7820

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
1.0529	1.0	225	1.0464	0.4542
1.0393	2.0	450	1.0215	0.4759
1.0194	3.0	675	0.9971	0.5158
0.9608	4.0	900	0.9729	0.5541
0.9743	5.0	1125	0.9487	0.6023
0.9002	6.0	1350	0.9258	0.6206
0.8961	7.0	1575	0.9030	0.6373
0.9282	8.0	1800	0.8813	0.6539
0.856	9.0	2025	0.8605	0.6705
0.8441	10.0	2250	0.8407	0.6772
0.8723	11.0	2475	0.8225	0.6839
0.7789	12.0	2700	0.8048	0.6955
0.7952	13.0	2925	0.7885	0.7055
0.7937	14.0	3150	0.7729	0.7155
0.8007	15.0	3375	0.7585	0.7255
0.769	16.0	3600	0.7449	0.7238
0.7262	17.0	3825	0.7325	0.7255
0.7259	18.0	4050	0.7208	0.7238
0.7176	19.0	4275	0.7099	0.7255
0.6791	20.0	4500	0.6998	0.7271
0.7106	21.0	4725	0.6905	0.7338
0.6951	22.0	4950	0.6819	0.7371
0.7193	23.0	5175	0.6739	0.7471
0.6759	24.0	5400	0.6663	0.7521
0.6975	25.0	5625	0.6593	0.7537
0.6391	26.0	5850	0.6529	0.7571
0.6617	27.0	6075	0.6469	0.7604
0.6434	28.0	6300	0.6413	0.7604
0.6619	29.0	6525	0.6362	0.7587
0.6444	30.0	6750	0.6315	0.7571
0.6161	31.0	6975	0.6270	0.7604
0.6193	32.0	7200	0.6230	0.7671
0.5926	33.0	7425	0.6193	0.7654
0.5861	34.0	7650	0.6159	0.7754
0.6256	35.0	7875	0.6127	0.7770
0.6099	36.0	8100	0.6099	0.7754
0.5932	37.0	8325	0.6073	0.7770
0.5988	38.0	8550	0.6049	0.7804
0.574	39.0	8775	0.6028	0.7787
0.5835	40.0	9000	0.6009	0.7787
0.5292	41.0	9225	0.5992	0.7787
0.586	42.0	9450	0.5977	0.7804
0.5537	43.0	9675	0.5964	0.7820
0.5573	44.0	9900	0.5953	0.7837
0.5715	45.0	10125	0.5945	0.7820
0.6072	46.0	10350	0.5938	0.7820
0.5714	47.0	10575	0.5933	0.7837
0.5684	48.0	10800	0.5929	0.7820
0.5949	49.0	11025	0.5927	0.7820
0.5423	50.0	11250	0.5927	0.7820

Framework versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2