videomae-base-finetuned-ucf_crime

shazab/videomae-base-finetuned-ucf_crime

videomae-base-finetuned-ucf_crime is a video classification model from shazab. 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 cc-by-nc-4.0.

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

Downloads · 30 days

3% of all-time downloads

All-time downloads

665

Public

Repo size

23.1 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.bin345 MB · 100%

At a glance

Task: Video Classification
Library: transformers
License: cc-by-nc-4.0
Model type: videomae
Access: Public
Created: May 22, 2023
Updated: May 24, 2023
SHA: 523f1851

Model card

From the Hugging Face model README

videomae-base-finetuned-ucf_crime

This model is a fine-tuned version of MCG-NJU/videomae-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 2.0871
Accuracy: 0.5058

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: 2
eval_batch_size: 2
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: 6000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6487	0.02	120	0.6813	0.6989
0.4529	1.02	240	0.7467	0.5678
0.5165	2.02	360	0.7600	0.7264
0.9203	3.02	480	0.6189	0.6517
1.2667	4.02	600	0.6702	0.6511
1.0452	5.02	720	0.5703	0.7592
0.3893	6.02	840	0.5524	0.7397
1.2013	7.02	960	0.7470	0.6109
1.4903	8.02	1080	1.0094	0.5695
0.5434	9.02	1200	0.5189	0.7586
0.8344	10.02	1320	0.7674	0.5552
1.1621	11.02	1440	0.9610	0.5874
0.8743	12.02	1560	0.8160	0.6448
0.7118	13.02	1680	0.6015	0.6592
0.8404	14.02	1800	0.7149	0.7661
1.0397	15.02	1920	0.6360	0.7770
1.3511	16.02	2040	0.9110	0.6523
1.0257	17.02	2160	1.0642	0.5891
1.0726	18.02	2280	0.7299	0.8213
0.5609	19.02	2400	0.8921	0.6408
0.495	20.02	2520	0.7762	0.7632
1.6306	21.02	2640	0.9976	0.7092
1.0072	22.02	2760	0.5697	0.8029
0.4503	23.02	2880	0.8448	0.6914
0.8306	24.02	3000	0.9617	0.7351
0.8824	25.02	3120	1.1839	0.6759
0.837	26.02	3240	1.6406	0.6075
0.4214	27.02	3360	1.1276	0.6920
0.5166	28.02	3480	1.0815	0.7626
0.6925	29.02	3600	1.1492	0.7086
0.2864	30.02	3720	1.1919	0.7345
0.463	31.02	3840	1.3524	0.6937
1.1162	32.02	3960	1.8301	0.5822
0.0033	33.02	4080	1.4447	0.6891
0.002	34.02	4200	1.6565	0.6960
0.0017	35.02	4320	1.5357	0.7282
0.2289	36.02	4440	1.9812	0.6397
0.4801	37.02	4560	2.2316	0.6167
1.0323	38.02	4680	2.1380	0.5822
0.0715	39.02	4800	1.9264	0.6345
0.0024	40.02	4920	2.6257	0.5023
0.4734	41.02	5040	1.8666	0.6316
0.3108	42.02	5160	1.5493	0.7299
0.2318	43.02	5280	2.0831	0.6333
0.6623	44.02	5400	2.3276	0.6029
0.0011	45.02	5520	2.0469	0.6684
0.32	46.02	5640	2.4646	0.6138
0.0054	47.02	5760	2.5998	0.5897
0.2733	48.02	5880	2.7594	0.5638
0.0005	49.02	6000	2.8625	0.55

Framework versions

Transformers 4.29.2
Pytorch 1.13.1+cu117
Datasets 2.12.0
Tokenizers 0.13.3

Task: Video Classification
Library: transformers
Type: videomae
License: cc-by-nc-4.0
Created: May 22, 2023
Updated: May 24, 2023