hushem_1x_deit_base_rms_001_fold1

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: 2.0515
• Accuracy: 0.4444

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	5.7295	0.2444
3.8216	2.0	12	3.3899	0.2444
3.8216	3.0	18	1.9221	0.2444
2.0397	4.0	24	1.5517	0.2667
1.7061	5.0	30	1.6113	0.2444
1.7061	6.0	36	1.5745	0.2667
1.5185	7.0	42	1.5819	0.2444
1.5185	8.0	48	1.5544	0.2667
1.5399	9.0	54	1.4265	0.2667
1.4704	10.0	60	1.4555	0.2444
1.4704	11.0	66	1.5190	0.2667
1.4477	12.0	72	1.4505	0.2444
1.4477	13.0	78	1.4958	0.2667
1.4478	14.0	84	1.3383	0.2667
1.4577	15.0	90	1.2186	0.4444
1.4577	16.0	96	1.3020	0.4
1.4812	17.0	102	1.3485	0.3333
1.4812	18.0	108	1.2420	0.4889
1.357	19.0	114	1.5123	0.2889
1.2669	20.0	120	1.0806	0.5111
1.2669	21.0	126	1.2866	0.3778
1.1674	22.0	132	1.2669	0.4
1.1674	23.0	138	1.2024	0.4222
1.0892	24.0	144	1.1560	0.5111
1.1158	25.0	150	1.3838	0.4
1.1158	26.0	156	1.3008	0.4667
0.9418	27.0	162	1.2353	0.4667
0.9418	28.0	168	1.1679	0.4667
0.9251	29.0	174	1.5982	0.4
0.8116	30.0	180	1.4467	0.4222
0.8116	31.0	186	1.4759	0.4222
0.7877	32.0	192	1.7036	0.4222
0.7877	33.0	198	1.5198	0.4667
0.8255	34.0	204	1.6264	0.4444
0.6324	35.0	210	1.5460	0.4444
0.6324	36.0	216	1.7519	0.4444
0.6569	37.0	222	2.1790	0.4222
0.6569	38.0	228	2.1337	0.4667
0.4724	39.0	234	1.5947	0.4889
0.4698	40.0	240	2.2448	0.4444
0.4698	41.0	246	2.0575	0.4444
0.3953	42.0	252	2.0515	0.4444
0.3953	43.0	258	2.0515	0.4444
0.4352	44.0	264	2.0515	0.4444
0.3453	45.0	270	2.0515	0.4444
0.3453	46.0	276	2.0515	0.4444
0.3689	47.0	282	2.0515	0.4444
0.3689	48.0	288	2.0515	0.4444
0.3565	49.0	294	2.0515	0.4444
0.3737	50.0	300	2.0515	0.4444

Framework versions

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