hushem_1x_deit_tiny_rms_00001_fold1

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: 1.2620
• Accuracy: 0.6222

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	1.3330	0.3111
1.415	2.0	12	1.1496	0.4222
1.415	3.0	18	1.0095	0.6
0.6844	4.0	24	1.0528	0.5333
0.3289	5.0	30	0.8970	0.6
0.3289	6.0	36	1.2025	0.5111
0.1275	7.0	42	0.9016	0.6
0.1275	8.0	48	1.0450	0.5778
0.049	9.0	54	1.1767	0.5556
0.0201	10.0	60	1.2285	0.5333
0.0201	11.0	66	1.0471	0.6
0.0071	12.0	72	0.9300	0.6444
0.0071	13.0	78	1.1280	0.5778
0.0042	14.0	84	1.1318	0.5556
0.0029	15.0	90	1.1503	0.5556
0.0029	16.0	96	1.0998	0.5778
0.0023	17.0	102	1.1889	0.5778
0.0023	18.0	108	1.2431	0.5778
0.0018	19.0	114	1.2158	0.5778
0.0016	20.0	120	1.2220	0.6
0.0016	21.0	126	1.1974	0.6
0.0014	22.0	132	1.2207	0.6
0.0014	23.0	138	1.2242	0.6
0.0013	24.0	144	1.2118	0.6
0.0011	25.0	150	1.2264	0.6222
0.0011	26.0	156	1.2250	0.6
0.0011	27.0	162	1.2237	0.6
0.0011	28.0	168	1.2290	0.6
0.001	29.0	174	1.2254	0.6222
0.0009	30.0	180	1.2294	0.6222
0.0009	31.0	186	1.2336	0.6222
0.0009	32.0	192	1.2394	0.6222
0.0009	33.0	198	1.2441	0.6222
0.0008	34.0	204	1.2483	0.6
0.0008	35.0	210	1.2484	0.6
0.0008	36.0	216	1.2564	0.6
0.0008	37.0	222	1.2583	0.6222
0.0008	38.0	228	1.2617	0.6222
0.0007	39.0	234	1.2626	0.6222
0.0007	40.0	240	1.2627	0.6222
0.0007	41.0	246	1.2621	0.6222
0.0007	42.0	252	1.2620	0.6222
0.0007	43.0	258	1.2620	0.6222
0.0007	44.0	264	1.2620	0.6222
0.0007	45.0	270	1.2620	0.6222
0.0007	46.0	276	1.2620	0.6222
0.0007	47.0	282	1.2620	0.6222
0.0007	48.0	288	1.2620	0.6222
0.0007	49.0	294	1.2620	0.6222
0.0007	50.0	300	1.2620	0.6222

Framework versions

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