CTMAE-P2-V4-S3

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

• Loss: 1.6581
• Accuracy: 0.7111

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: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 13050

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.3133	0.02	261	2.0786	0.5556
0.6004	1.02	522	2.5565	0.5556
1.6674	2.02	783	2.1021	0.5556
0.7822	3.02	1044	2.1914	0.5556
1.6707	4.02	1305	1.9963	0.5556
0.9301	5.02	1566	2.0110	0.5556
0.7333	6.02	1827	2.0193	0.5556
1.0221	7.02	2088	1.7499	0.5556
1.0956	8.02	2349	1.9463	0.5556
0.6189	9.02	2610	1.3986	0.5556
1.5383	10.02	2871	1.7564	0.5556
1.9417	11.02	3132	1.7499	0.5556
0.4756	12.02	3393	1.3450	0.6889
0.3515	13.02	3654	1.4140	0.6444
0.0096	14.02	3915	1.8714	0.6222
0.7724	15.02	4176	1.6674	0.6889
0.549	16.02	4437	2.0384	0.6
1.033	17.02	4698	1.6581	0.7111
1.0439	18.02	4959	1.9511	0.6222
1.7522	19.02	5220	1.9120	0.7111
0.0011	20.02	5481	2.1552	0.6222
0.7904	21.02	5742	2.0723	0.6444
0.0824	22.02	6003	2.3224	0.6
0.4656	23.02	6264	1.8394	0.6444
1.5142	24.02	6525	2.4023	0.6444
0.0002	25.02	6786	2.0959	0.6444
0.8021	26.02	7047	1.9081	0.6444
0.6738	27.02	7308	2.4259	0.6444
0.015	28.02	7569	2.5600	0.6222
0.001	29.02	7830	2.4097	0.6222
0.0003	30.02	8091	2.3336	0.6222
0.353	31.02	8352	2.4537	0.6444
0.0003	32.02	8613	2.4429	0.6889
0.5047	33.02	8874	2.5712	0.6222
0.0002	34.02	9135	2.3336	0.6444
1.3626	35.02	9396	2.5391	0.6222
0.5114	36.02	9657	2.6003	0.6222
0.0528	37.02	9918	2.4778	0.6
0.0027	38.02	10179	2.8720	0.6
0.703	39.02	10440	2.7653	0.6444
0.0001	40.02	10701	2.6459	0.6444
0.0001	41.02	10962	2.5735	0.6667
0.0	42.02	11223	2.6030	0.6889
0.0008	43.02	11484	2.7085	0.6667
0.0001	44.02	11745	2.8014	0.6444
0.0001	45.02	12006	2.7477	0.6667
0.0028	46.02	12267	2.8431	0.6444
0.0002	47.02	12528	2.8467	0.6444
0.003	48.02	12789	2.9743	0.6444
0.0503	49.02	13050	2.9219	0.6444

Framework versions

• Transformers 4.46.2
• Pytorch 2.0.1+cu117
• Datasets 3.0.1
• Tokenizers 0.20.0