hushem_40x_deit_base_sgd_001_fold5

hkivancoral/hushem_40x_deit_base_sgd_001_fold5

hushem_40x_deit_base_sgd_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

4% of all-time downloads

All-time downloads

213

Public

Repo size

10.6 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin343 MB · 100%

At a glance

Task: Image Classification
Library: transformers
License: apache-2.0
Model type: vit
Access: Public
Created: Dec 24, 2023
Updated: Dec 24, 2023
SHA: 34252a0e

Base models

facebook/deit-base-patch16-224

Datasets

imagefolder

Model card

From the Hugging Face model README

hushem_40x_deit_base_sgd_001_fold5

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: 0.5235
Accuracy: 0.8049

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: 0.001
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
1.2691	1.0	220	1.3176	0.3415
1.1324	2.0	440	1.2295	0.4634
1.003	3.0	660	1.1173	0.6341
0.8718	4.0	880	0.9888	0.6585
0.7662	5.0	1100	0.8700	0.6829
0.6305	6.0	1320	0.7780	0.6585
0.552	7.0	1540	0.7068	0.6829
0.4791	8.0	1760	0.6670	0.6829
0.413	9.0	1980	0.6302	0.6829
0.3827	10.0	2200	0.6050	0.7073
0.3215	11.0	2420	0.5880	0.7073
0.2953	12.0	2640	0.5689	0.7073
0.2691	13.0	2860	0.5551	0.7073
0.255	14.0	3080	0.5391	0.7317
0.2205	15.0	3300	0.5338	0.7561
0.2031	16.0	3520	0.5276	0.8049
0.1827	17.0	3740	0.5158	0.8049
0.178	18.0	3960	0.5117	0.8049
0.1722	19.0	4180	0.5070	0.8293
0.1354	20.0	4400	0.5054	0.8293
0.1154	21.0	4620	0.5008	0.8293
0.1032	22.0	4840	0.5031	0.8293
0.123	23.0	5060	0.5052	0.8293
0.0925	24.0	5280	0.5012	0.8049
0.1004	25.0	5500	0.5002	0.8293
0.1106	26.0	5720	0.5000	0.8293
0.0932	27.0	5940	0.5018	0.8293
0.0974	28.0	6160	0.5069	0.8293
0.0749	29.0	6380	0.5067	0.8293
0.0626	30.0	6600	0.5071	0.8293
0.058	31.0	6820	0.5023	0.8293
0.0771	32.0	7040	0.5068	0.8293
0.0537	33.0	7260	0.5089	0.8049
0.0443	34.0	7480	0.5110	0.8049
0.0529	35.0	7700	0.5102	0.8049
0.056	36.0	7920	0.5123	0.8293
0.0373	37.0	8140	0.5147	0.8293
0.0662	38.0	8360	0.5122	0.8293
0.0489	39.0	8580	0.5155	0.8293
0.0389	40.0	8800	0.5166	0.8293
0.0414	41.0	9020	0.5205	0.8049
0.0455	42.0	9240	0.5225	0.8293
0.0397	43.0	9460	0.5226	0.8049
0.0345	44.0	9680	0.5228	0.8049
0.0281	45.0	9900	0.5217	0.8049
0.0392	46.0	10120	0.5231	0.8049
0.0436	47.0	10340	0.5235	0.8293
0.0347	48.0	10560	0.5238	0.8049
0.0331	49.0	10780	0.5237	0.8049
0.0457	50.0	11000	0.5235	0.8049

Framework versions

Transformers 4.32.1
Pytorch 2.1.0+cu121
Datasets 2.12.0
Tokenizers 0.13.2

Task: Image Classification
Library: transformers
Type: vit
License: apache-2.0
Created: Dec 24, 2023
Updated: Dec 24, 2023