hushem_1x_deit_tiny_adamax_001_fold5

hkivancoral/hushem_1x_deit_tiny_adamax_001_fold5

hushem_1x_deit_tiny_adamax_001_fold5 is a image classification model from hkivancoral. Use it when you need a label for an image. It is set up for transformers. The card lists the license as apache-2.0.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

2% of all-time downloads

All-time downloads

552

Public

Parameters

5.5M

1.4 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors22.1 MB · 100%

At a glance

Task: Image Classification
Library: transformers
License: apache-2.0
Model type: vit
Access: Public
Created: Nov 13, 2023
Updated: Nov 13, 2023
SHA: 84f46843

Base models

facebook/deit-tiny-patch16-224

Datasets

imagefolder

Model card

From the Hugging Face model README

hushem_1x_deit_tiny_adamax_001_fold5

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: 0.9886
Accuracy: 0.7805

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: 0.0001
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	1.2270	0.3415
1.4194	2.0	12	1.0630	0.5122
1.4194	3.0	18	0.7493	0.7073
0.7944	4.0	24	0.7294	0.7561
0.3715	5.0	30	0.6953	0.6585
0.3715	6.0	36	0.5928	0.8293
0.1471	7.0	42	0.5485	0.8049
0.1471	8.0	48	0.8515	0.6829
0.0288	9.0	54	0.5381	0.8293
0.0065	10.0	60	0.8647	0.7317
0.0065	11.0	66	0.7563	0.7805
0.0018	12.0	72	0.7678	0.8049
0.0018	13.0	78	0.8017	0.8049
0.0008	14.0	84	0.8475	0.7805
0.0005	15.0	90	0.8926	0.7805
0.0005	16.0	96	0.9216	0.7805
0.0004	17.0	102	0.9424	0.7805
0.0004	18.0	108	0.9465	0.7805
0.0003	19.0	114	0.9461	0.7805
0.0003	20.0	120	0.9448	0.7805
0.0003	21.0	126	0.9474	0.7805
0.0003	22.0	132	0.9525	0.7805
0.0003	23.0	138	0.9551	0.7805
0.0003	24.0	144	0.9581	0.7805
0.0002	25.0	150	0.9626	0.7805
0.0002	26.0	156	0.9650	0.7805
0.0002	27.0	162	0.9711	0.7805
0.0002	28.0	168	0.9713	0.7805
0.0002	29.0	174	0.9730	0.7805
0.0002	30.0	180	0.9754	0.7805
0.0002	31.0	186	0.9786	0.7805
0.0002	32.0	192	0.9820	0.7805
0.0002	33.0	198	0.9835	0.7805
0.0002	34.0	204	0.9850	0.7805
0.0002	35.0	210	0.9850	0.7805
0.0002	36.0	216	0.9860	0.7805
0.0002	37.0	222	0.9866	0.7805
0.0002	38.0	228	0.9873	0.7805
0.0002	39.0	234	0.9879	0.7805
0.0002	40.0	240	0.9883	0.7805
0.0002	41.0	246	0.9886	0.7805
0.0002	42.0	252	0.9886	0.7805
0.0002	43.0	258	0.9886	0.7805
0.0002	44.0	264	0.9886	0.7805
0.0002	45.0	270	0.9886	0.7805
0.0002	46.0	276	0.9886	0.7805
0.0002	47.0	282	0.9886	0.7805
0.0002	48.0	288	0.9886	0.7805
0.0002	49.0	294	0.9886	0.7805
0.0002	50.0	300	0.9886	0.7805

Framework versions

Transformers 4.35.0
Pytorch 2.1.0+cu118
Datasets 2.14.6
Tokenizers 0.14.1

Task: Image Classification
Library: transformers
Type: vit
License: apache-2.0
Created: Nov 13, 2023
Updated: Nov 13, 2023