hushem_1x_deit_small_rms_00001_fold4

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: 0.4388
• Accuracy: 0.8095

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: 1e-05
• 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	1.2550	0.4286
1.2114	2.0	12	1.1127	0.4762
1.2114	3.0	18	0.8436	0.6667
0.6039	4.0	24	0.7891	0.6429
0.2289	5.0	30	0.6119	0.7143
0.2289	6.0	36	0.5730	0.7381
0.0572	7.0	42	0.5854	0.7381
0.0572	8.0	48	0.4823	0.7619
0.0155	9.0	54	0.4273	0.8095
0.0057	10.0	60	0.4459	0.8095
0.0057	11.0	66	0.4283	0.8333
0.0036	12.0	72	0.4439	0.8333
0.0036	13.0	78	0.4381	0.8333
0.0028	14.0	84	0.4361	0.8095
0.0022	15.0	90	0.4297	0.8095
0.0022	16.0	96	0.4286	0.8333
0.0018	17.0	102	0.4333	0.8333
0.0018	18.0	108	0.4303	0.8333
0.0015	19.0	114	0.4275	0.8095
0.0014	20.0	120	0.4353	0.8095
0.0014	21.0	126	0.4311	0.8095
0.0012	22.0	132	0.4354	0.8095
0.0012	23.0	138	0.4378	0.8095
0.0011	24.0	144	0.4372	0.8095
0.001	25.0	150	0.4362	0.8095
0.001	26.0	156	0.4357	0.8095
0.0009	27.0	162	0.4417	0.8095
0.0009	28.0	168	0.4425	0.8095
0.0009	29.0	174	0.4408	0.8095
0.0008	30.0	180	0.4402	0.8095
0.0008	31.0	186	0.4406	0.8095
0.0008	32.0	192	0.4385	0.8095
0.0008	33.0	198	0.4397	0.8095
0.0007	34.0	204	0.4393	0.8095
0.0007	35.0	210	0.4395	0.8095
0.0007	36.0	216	0.4391	0.8095
0.0007	37.0	222	0.4387	0.8095
0.0007	38.0	228	0.4386	0.8095
0.0007	39.0	234	0.4388	0.8095
0.0007	40.0	240	0.4387	0.8095
0.0007	41.0	246	0.4388	0.8095
0.0007	42.0	252	0.4388	0.8095
0.0007	43.0	258	0.4388	0.8095
0.0006	44.0	264	0.4388	0.8095
0.0007	45.0	270	0.4388	0.8095
0.0007	46.0	276	0.4388	0.8095
0.0007	47.0	282	0.4388	0.8095
0.0007	48.0	288	0.4388	0.8095
0.0006	49.0	294	0.4388	0.8095
0.0007	50.0	300	0.4388	0.8095

Framework versions

• Transformers 4.35.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.7
• Tokenizers 0.14.1