hushem_1x_deit_base_adamax_00001_fold2

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: 1.2691
• Accuracy: 0.5556

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	1.3747	0.2222
1.3344	2.0	12	1.3328	0.3333
1.3344	3.0	18	1.2950	0.3778
1.1589	4.0	24	1.2639	0.4222
1.0046	5.0	30	1.2362	0.4667
1.0046	6.0	36	1.2118	0.5111
0.8105	7.0	42	1.1924	0.5111
0.8105	8.0	48	1.1812	0.5111
0.6759	9.0	54	1.1724	0.4889
0.5237	10.0	60	1.1564	0.5333
0.5237	11.0	66	1.1490	0.5333
0.4159	12.0	72	1.1389	0.5333
0.4159	13.0	78	1.1320	0.5556
0.3256	14.0	84	1.1287	0.5333
0.2501	15.0	90	1.1375	0.5333
0.2501	16.0	96	1.1291	0.5333
0.2011	17.0	102	1.1342	0.5333
0.2011	18.0	108	1.1537	0.5333
0.1629	19.0	114	1.1612	0.5333
0.1297	20.0	120	1.1554	0.5333
0.1297	21.0	126	1.1646	0.5333
0.105	22.0	132	1.1740	0.5333
0.105	23.0	138	1.1835	0.5333
0.0899	24.0	144	1.1913	0.5333
0.0683	25.0	150	1.2006	0.5333
0.0683	26.0	156	1.2072	0.5556
0.0612	27.0	162	1.2095	0.5556
0.0612	28.0	168	1.2218	0.5556
0.0476	29.0	174	1.2356	0.5556
0.0433	30.0	180	1.2335	0.5556
0.0433	31.0	186	1.2371	0.5556
0.0387	32.0	192	1.2442	0.5556
0.0387	33.0	198	1.2507	0.5556
0.0349	34.0	204	1.2533	0.5556
0.0342	35.0	210	1.2544	0.5556
0.0342	36.0	216	1.2556	0.5556
0.0314	37.0	222	1.2581	0.5556
0.0314	38.0	228	1.2633	0.5556
0.0273	39.0	234	1.2676	0.5556
0.0293	40.0	240	1.2685	0.5556
0.0293	41.0	246	1.2690	0.5556
0.0278	42.0	252	1.2691	0.5556
0.0278	43.0	258	1.2691	0.5556
0.0273	44.0	264	1.2691	0.5556
0.0276	45.0	270	1.2691	0.5556
0.0276	46.0	276	1.2691	0.5556
0.0276	47.0	282	1.2691	0.5556
0.0276	48.0	288	1.2691	0.5556
0.0275	49.0	294	1.2691	0.5556
0.0279	50.0	300	1.2691	0.5556

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.14.7
• Tokenizers 0.15.0