smids_5x_deit_base_rms_001_fold3

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

• Loss: 0.7864
• Accuracy: 0.775

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
0.8829	1.0	375	0.8947	0.4967
0.7778	2.0	750	0.8858	0.5283
0.8311	3.0	1125	0.8355	0.55
0.8294	4.0	1500	0.7887	0.6067
0.8799	5.0	1875	0.8054	0.6217
0.7828	6.0	2250	0.8032	0.5883
0.7591	7.0	2625	0.7342	0.6533
0.7195	8.0	3000	0.7320	0.6317
0.6862	9.0	3375	0.8798	0.5583
0.6609	10.0	3750	0.6983	0.6717
0.6813	11.0	4125	0.7308	0.67
0.7021	12.0	4500	0.7207	0.63
0.6223	13.0	4875	0.6947	0.685
0.5883	14.0	5250	0.6340	0.7217
0.6307	15.0	5625	0.6616	0.7033
0.6001	16.0	6000	0.6868	0.6983
0.6448	17.0	6375	0.6323	0.7233
0.6618	18.0	6750	0.6385	0.7233
0.5927	19.0	7125	0.6305	0.71
0.5637	20.0	7500	0.6279	0.7033
0.5052	21.0	7875	0.6248	0.7217
0.5232	22.0	8250	0.6213	0.7133
0.586	23.0	8625	0.6387	0.7217
0.6497	24.0	9000	0.5879	0.7367
0.5103	25.0	9375	0.6071	0.7317
0.5081	26.0	9750	0.6415	0.725
0.5448	27.0	10125	0.5744	0.725
0.5273	28.0	10500	0.6005	0.7333
0.5173	29.0	10875	0.6315	0.725
0.4436	30.0	11250	0.5604	0.7617
0.5135	31.0	11625	0.5861	0.7633
0.5481	32.0	12000	0.5892	0.7467
0.5389	33.0	12375	0.5940	0.7533
0.5388	34.0	12750	0.5721	0.74
0.4177	35.0	13125	0.6397	0.7317
0.4044	36.0	13500	0.6157	0.74
0.4195	37.0	13875	0.6245	0.75
0.4242	38.0	14250	0.5771	0.7567
0.4027	39.0	14625	0.5507	0.7533
0.3811	40.0	15000	0.6123	0.7483
0.3735	41.0	15375	0.6056	0.7583
0.3892	42.0	15750	0.6319	0.7633
0.3928	43.0	16125	0.6475	0.7683
0.3524	44.0	16500	0.6335	0.7783
0.2896	45.0	16875	0.6766	0.7683
0.3014	46.0	17250	0.7029	0.775
0.2555	47.0	17625	0.6873	0.7967
0.2601	48.0	18000	0.7571	0.78
0.2142	49.0	18375	0.7689	0.7833
0.2176	50.0	18750	0.7864	0.775

Framework versions

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