vivit-b-16x2-kinetics400-ft-3620

Temo27Anas/vivit-b-16x2-kinetics400-ft-3620

vivit-b-16x2-kinetics400-ft-3620 is a video classification model from Temo27Anas. Use it for the video classification task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as mit.

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

Downloads · 30 days

14% of all-time downloads

All-time downloads

Public

Parameters

32.8M

6.6 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors131 MB · 100%

At a glance

Task: Video Classification
Library: transformers
License: mit
Model type: vivit
Access: Public
Created: Jul 8, 2024
Updated: Jul 9, 2024
SHA: 2f6a95b9

Base models

google/vivit-b-16x2-kinetics400

Model card

From the Hugging Face model README

vivit-b-16x2-kinetics400-ft-3620

This model is a fine-tuned version of google/vivit-b-16x2-kinetics400 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.9281
Accuracy: 0.5566

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: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
training_steps: 5500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0684	0.0202	111	1.1114	0.3799
1.0415	1.0202	222	1.0135	0.5249
1.0271	2.0202	333	1.0630	0.4857
1.1609	3.0202	444	1.0203	0.4222
0.9824	4.0202	555	1.0219	0.5249
1.0247	5.0202	666	1.0210	0.5026
1.0824	6.0202	777	0.9947	0.4720
0.943	7.0202	888	1.1085	0.4381
0.8807	8.0202	999	0.9345	0.5767
1.1009	9.0202	1110	0.9855	0.5164
1.0292	10.0202	1221	1.0506	0.4339
0.9071	11.0202	1332	0.9926	0.5143
1.0001	12.0202	1443	1.0406	0.4931
0.9698	13.0202	1554	0.9440	0.5598
0.9405	14.0202	1665	0.9667	0.5323
0.8802	15.0202	1776	0.9011	0.5862
0.9154	16.0202	1887	0.9429	0.5598
0.929	17.0202	1998	0.9948	0.5132
0.9112	18.0202	2109	0.9056	0.5852
0.9202	19.0202	2220	0.9489	0.5524
0.9004	20.0202	2331	0.8995	0.5820
0.9318	21.0202	2442	0.9032	0.5958
0.8493	22.0202	2553	0.9975	0.5238
0.8587	23.0202	2664	1.0142	0.5259
0.958	24.0202	2775	0.9665	0.5376
0.996	25.0202	2886	0.9391	0.5704
0.823	26.0202	2997	0.9171	0.5778
0.8834	27.0202	3108	0.8923	0.5873
0.8615	28.0202	3219	0.9577	0.5471
0.9462	29.0202	3330	0.9468	0.5630
0.8909	30.0202	3441	0.9343	0.5672
0.8048	31.0202	3552	0.9107	0.5778
0.8109	32.0202	3663	0.9547	0.5492
0.9242	33.0202	3774	0.9275	0.5598
0.9046	34.0202	3885	0.9290	0.5831
0.7677	35.0202	3996	0.9208	0.5725
0.8501	36.0202	4107	0.9126	0.5810
0.8468	37.0202	4218	0.9053	0.5862
0.7814	38.0202	4329	0.8858	0.5905
0.9354	39.0202	4440	0.9207	0.5725
0.8849	40.0202	4551	0.9277	0.5651
0.7856	41.0202	4662	0.9130	0.5915
0.7133	42.0202	4773	0.9080	0.5884
0.932	43.0202	4884	0.9388	0.5577
0.6883	44.0202	4995	0.8925	0.5937
0.9944	45.0202	5106	0.9143	0.5820
0.8892	46.0202	5217	0.9103	0.5884
0.9071	47.0202	5328	0.9018	0.5905
0.7943	48.0202	5439	0.9022	0.5905
0.8034	49.0111	5500	0.9004	0.5947

Framework versions

Transformers 4.41.2
Pytorch 1.13.0+cu117
Datasets 2.20.0
Tokenizers 0.19.1

Task: Video Classification
Library: transformers
Type: vivit
License: mit
Created: Jul 8, 2024
Updated: Jul 9, 2024