cont-vvt-gs-rot-flip-wtoken-f198-4.4-h768-t8.16.16

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: 0.6569
• Accuracy: 0.7407

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: 2e-06
• train_batch_size: 4
• eval_batch_size: 4
• 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
0.868	0.0402	221	0.7216	0.6931
0.715	1.0402	442	0.6954	0.7037
0.7041	2.0402	663	0.7269	0.7460
0.5228	3.0402	884	0.7158	0.7407
0.7336	4.0402	1105	0.6833	0.7196
0.5987	5.0402	1326	0.6155	0.7725
0.8574	6.0402	1547	0.6601	0.7302
0.6805	7.0402	1768	0.6374	0.7460
0.8086	8.0402	1989	0.6896	0.6984
0.6552	9.0402	2210	0.6535	0.7090
0.7846	10.0402	2431	0.6646	0.7354
0.6114	11.0402	2652	0.6111	0.7619
0.7435	12.0402	2873	0.6779	0.7354
0.7742	13.0402	3094	0.7390	0.6878
0.6558	14.0402	3315	0.6284	0.7354
0.4822	15.0402	3536	0.7071	0.7196
0.7686	16.0402	3757	0.6982	0.7302
0.7945	17.0402	3978	0.6336	0.7566
0.5755	18.0402	4199	0.5924	0.7460
0.6895	19.0402	4420	0.6227	0.7513
0.4775	20.0402	4641	0.5846	0.7672
0.8137	21.0402	4862	0.6724	0.7354
0.4226	22.0402	5083	0.6772	0.7460
0.6616	23.0402	5304	0.6856	0.7407
0.5246	24.0356	5500	0.6569	0.7407

Framework versions

• Transformers 4.41.2
• Pytorch 1.13.0+cu117
• Datasets 2.20.0
• Tokenizers 0.19.1