microsoft_swinv2-tiny-patch4-window8-256-batch_16_epoch_4_classes_24_final_withAug

Zannatul/microsoft_swinv2-tiny-patch4-window8-256-batch_16_epoch_4_classes_24_final_withAug

microsoft_swinv2-tiny-patch4-window8-256-batch_16_epoch_4_classes_24_final_withAug is a image classification model from Zannatul. 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

15% of all-time downloads

All-time downloads

Public

Parameters

27.6M

111 MB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors110 MB · 100%

At a glance

Task: Image Classification
Library: transformers
License: apache-2.0
Model type: swinv2
Access: Public
Created: May 9, 2024
Updated: May 9, 2024
SHA: ec6c855a

Base models

microsoft/swinv2-tiny-patch4-window8-256

Datasets

imagefolder

Model card

From the Hugging Face model README

microsoft_swinv2-tiny-patch4-window8-256-batch_16_epoch_4_classes_24_final_withAug

This model is a fine-tuned version of microsoft/swinv2-tiny-patch4-window8-256 on the bengali_food_images dataset. It achieves the following results on the evaluation set:

Loss: 0.2321
Accuracy: 0.9457

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: 0.0002
train_batch_size: 16
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 4
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.7162	0.09	100	1.4225	0.7079
1.2286	0.17	200	0.9461	0.7935
1.0323	0.26	300	0.7366	0.8356
0.8678	0.34	400	0.6211	0.8628
0.7849	0.43	500	0.5354	0.8655
0.7105	0.51	600	0.4793	0.8899
0.6198	0.6	700	0.4319	0.9090
0.6276	0.68	800	0.4022	0.8981
0.5411	0.77	900	0.3816	0.9117
0.4984	0.85	1000	0.3824	0.9022
0.5665	0.94	1100	0.3460	0.9212
0.5741	1.02	1200	0.3336	0.9158
0.4039	1.11	1300	0.3204	0.9130
0.4347	1.19	1400	0.3038	0.9307
0.3639	1.28	1500	0.2955	0.9253
0.4282	1.36	1600	0.2948	0.9293
0.4375	1.45	1700	0.2868	0.9212
0.3063	1.53	1800	0.2861	0.9334
0.3549	1.62	1900	0.2826	0.9293
0.4326	1.71	2000	0.2698	0.9348
0.3697	1.79	2100	0.2602	0.9280
0.3155	1.88	2200	0.2523	0.9361
0.3348	1.96	2300	0.2506	0.9470
0.3854	2.05	2400	0.2565	0.9321
0.3951	2.13	2500	0.2482	0.9402
0.3531	2.22	2600	0.2455	0.9402
0.3643	2.3	2700	0.2513	0.9375
0.3393	2.39	2800	0.2492	0.9429
0.3635	2.47	2900	0.2394	0.9402
0.3624	2.56	3000	0.2425	0.9389
0.3608	2.64	3100	0.2390	0.9457
0.3215	2.73	3200	0.2483	0.9321
0.2971	2.81	3300	0.2455	0.9402
0.3838	2.9	3400	0.2363	0.9470
0.3036	2.98	3500	0.2422	0.9402
0.401	3.07	3600	0.2398	0.9429
0.3458	3.15	3700	0.2517	0.9429
0.2908	3.24	3800	0.2423	0.9457
0.3016	3.32	3900	0.2402	0.9443
0.2961	3.41	4000	0.2414	0.9457
0.3822	3.5	4100	0.2413	0.9416
0.2596	3.58	4200	0.2356	0.9457
0.3064	3.67	4300	0.2324	0.9497
0.3059	3.75	4400	0.2321	0.9457
0.42	3.84	4500	0.2556	0.9402
0.2959	3.92	4600	0.2491	0.9416

Framework versions

Transformers 4.39.3
Pytorch 2.1.2
Datasets 2.18.0
Tokenizers 0.15.2

Task: Image Classification
Library: transformers
Type: swinv2
License: apache-2.0
Created: May 9, 2024
Updated: May 9, 2024