hushem_1x_deit_tiny_sgd_lr00001_fold3

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: 1.5383
• Accuracy: 0.3023

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	1.5437	0.3023
1.6283	2.0	12	1.5434	0.3023
1.6283	3.0	18	1.5431	0.3023
1.63	4.0	24	1.5428	0.3023
1.6367	5.0	30	1.5426	0.3023
1.6367	6.0	36	1.5423	0.3023
1.6273	7.0	42	1.5421	0.3023
1.6273	8.0	48	1.5419	0.3023
1.6489	9.0	54	1.5417	0.3023
1.5924	10.0	60	1.5414	0.3023
1.5924	11.0	66	1.5412	0.3023
1.6227	12.0	72	1.5411	0.3023
1.6227	13.0	78	1.5409	0.3023
1.6367	14.0	84	1.5407	0.3023
1.622	15.0	90	1.5405	0.3023
1.622	16.0	96	1.5403	0.3023
1.621	17.0	102	1.5401	0.3023
1.621	18.0	108	1.5400	0.3023
1.6386	19.0	114	1.5398	0.3023
1.6207	20.0	120	1.5397	0.3023
1.6207	21.0	126	1.5395	0.3023
1.6152	22.0	132	1.5394	0.3023
1.6152	23.0	138	1.5393	0.3023
1.6503	24.0	144	1.5392	0.3023
1.6219	25.0	150	1.5390	0.3023
1.6219	26.0	156	1.5389	0.3023
1.6152	27.0	162	1.5389	0.3023
1.6152	28.0	168	1.5388	0.3023
1.6192	29.0	174	1.5387	0.3023
1.6111	30.0	180	1.5386	0.3023
1.6111	31.0	186	1.5386	0.3023
1.6114	32.0	192	1.5385	0.3023
1.6114	33.0	198	1.5384	0.3023
1.6361	34.0	204	1.5384	0.3023
1.6146	35.0	210	1.5384	0.3023
1.6146	36.0	216	1.5383	0.3023
1.6254	37.0	222	1.5383	0.3023
1.6254	38.0	228	1.5383	0.3023
1.6124	39.0	234	1.5383	0.3023
1.6367	40.0	240	1.5383	0.3023
1.6367	41.0	246	1.5383	0.3023
1.6229	42.0	252	1.5383	0.3023
1.6229	43.0	258	1.5383	0.3023
1.6506	44.0	264	1.5383	0.3023
1.6148	45.0	270	1.5383	0.3023
1.6148	46.0	276	1.5383	0.3023
1.6242	47.0	282	1.5383	0.3023
1.6242	48.0	288	1.5383	0.3023
1.6087	49.0	294	1.5383	0.3023
1.6097	50.0	300	1.5383	0.3023

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1