mvit_v2_s_Kinetics400_transf_BLANK_RWF-2000_SEGMENTATION

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

• Loss: 0.4089
• Accuracy: 0.885
• F1: 0.8850
• Precision: 0.8856

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
0.5225	0.0667	152	0.5302	0.8	0.8016	0.8091
0.3808	1.0667	304	0.3757	0.85	0.85	0.85
0.4703	2.0667	456	0.3731	0.8313	0.8304	0.8376
0.4689	3.0667	608	0.3386	0.8438	0.8436	0.8451
0.3932	4.0667	760	0.3443	0.8812	0.8812	0.8818
0.4814	5.0667	912	0.3364	0.8875	0.8873	0.8897
0.3567	6.0667	1064	0.3215	0.8625	0.8624	0.8634
0.1626	7.0667	1216	0.3214	0.8938	0.8937	0.8943
0.1784	8.0667	1368	0.3331	0.8938	0.8936	0.8953
0.2009	9.0667	1520	0.3242	0.9	0.9000	0.9003
0.2908	10.0667	1672	0.3588	0.9	0.9000	0.9003
0.3546	11.0667	1824	0.3930	0.8688	0.8686	0.8702
0.2209	12.0667	1976	0.3809	0.8875	0.8875	0.8875
0.2317	13.0667	2128	0.3858	0.8875	0.8875	0.8875
0.15	14.0667	2280	0.3865	0.8875	0.8875	0.8875

Framework versions

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