smids_1x_deit_tiny_rms_001_fold4

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: 0.9960
• Accuracy: 0.7333

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
1.1598	1.0	75	4.0739	0.3033
1.1025	2.0	150	1.1472	0.335
1.0056	3.0	225	0.8843	0.5617
0.9034	4.0	300	0.8985	0.5117
0.8764	5.0	375	0.9613	0.5017
0.9617	6.0	450	0.9074	0.5317
0.8578	7.0	525	0.8240	0.5717
0.8424	8.0	600	0.8437	0.5617
0.8025	9.0	675	0.7942	0.5833
0.7777	10.0	750	0.7683	0.57
0.8053	11.0	825	0.7474	0.5983
0.818	12.0	900	0.7555	0.61
0.8018	13.0	975	0.7629	0.5833
0.8411	14.0	1050	0.7216	0.635
0.6416	15.0	1125	0.8742	0.56
0.8084	16.0	1200	0.7814	0.6083
0.7505	17.0	1275	0.7600	0.6183
0.6996	18.0	1350	0.7346	0.6283
0.7648	19.0	1425	0.7240	0.6617
0.6916	20.0	1500	0.6768	0.6767
0.7556	21.0	1575	0.7263	0.6617
0.6471	22.0	1650	0.7297	0.6583
0.752	23.0	1725	0.7501	0.635
0.7349	24.0	1800	0.6751	0.6883
0.6802	25.0	1875	0.6689	0.6817
0.6239	26.0	1950	0.8871	0.5817
0.6865	27.0	2025	0.6485	0.7033
0.6138	28.0	2100	0.6457	0.7233
0.6707	29.0	2175	0.6937	0.6833
0.6824	30.0	2250	0.6688	0.7033
0.5913	31.0	2325	0.6725	0.715
0.5797	32.0	2400	0.6508	0.7167
0.5524	33.0	2475	0.7048	0.7
0.4736	34.0	2550	0.6807	0.6933
0.5263	35.0	2625	0.6317	0.7233
0.5348	36.0	2700	0.6398	0.7367
0.5082	37.0	2775	0.6440	0.7183
0.4972	38.0	2850	0.6697	0.7167
0.4567	39.0	2925	0.6947	0.73
0.4313	40.0	3000	0.6527	0.7383
0.4762	41.0	3075	0.6875	0.74
0.4293	42.0	3150	0.7259	0.7333
0.4594	43.0	3225	0.7531	0.7367
0.379	44.0	3300	0.7792	0.7383
0.3265	45.0	3375	0.7882	0.74
0.2807	46.0	3450	0.8615	0.7367
0.2733	47.0	3525	0.9438	0.73
0.2273	48.0	3600	0.9312	0.73
0.2189	49.0	3675	0.9889	0.7383
0.1609	50.0	3750	0.9960	0.7333

Framework versions

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