mvit_v2_s_Kinetics400_transf_BLANK_RWF-2000_ONLY_PERSON

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

• Loss: 0.2832
• Accuracy: 0.9158
• F1: 0.9157
• Precision: 0.9157

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: 1e-05
• train_batch_size: 10
• eval_batch_size: 10
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 2280

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision
0.5681	0.0632	144	0.4817	0.85	0.8494	0.8607
0.4612	1.0632	288	0.3374	0.875	0.8743	0.8797
0.3022	2.0632	432	0.2846	0.9	0.9	0.9
0.3056	3.0632	576	0.2733	0.9	0.8999	0.9003
0.3098	4.0632	720	0.3172	0.9167	0.9166	0.9211
0.3965	5.0632	864	0.2689	0.9333	0.9333	0.9354
0.1318	6.0632	1008	0.2828	0.9333	0.9333	0.9354
0.1719	7.0632	1152	0.3062	0.925	0.9250	0.9282
0.0967	8.0632	1296	0.3469	0.925	0.9250	0.9282
0.0456	9.0632	1440	0.3319	0.925	0.9250	0.9282
0.1613	10.0632	1584	0.3430	0.925	0.9250	0.9282

Framework versions

• Transformers 4.44.2
• Pytorch 2.0.1+cu118
• Datasets 2.21.0
• Tokenizers 0.19.1