CTMAE-P2-V4-S2

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: 2.1238
• Accuracy: 0.7333

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
1.5797	0.02	261	2.3910	0.5556
0.6074	1.02	522	2.6631	0.5556
1.5457	2.02	783	2.0037	0.5556
0.7292	3.02	1044	2.4139	0.5556
1.5417	4.02	1305	1.8115	0.5556
0.9259	5.02	1566	1.9903	0.5556
0.7081	6.02	1827	2.2051	0.5556
1.0459	7.02	2088	1.6674	0.5778
0.9128	8.02	2349	2.2127	0.5556
0.5241	9.02	2610	1.8464	0.5556
1.1457	10.02	2871	2.0479	0.5556
1.2264	11.02	3132	1.9070	0.5556
1.4812	12.02	3393	0.9185	0.7111
0.2804	13.02	3654	1.6409	0.6
0.0047	14.02	3915	1.9807	0.6
0.0406	15.02	4176	1.6699	0.6889
1.102	16.02	4437	1.7953	0.5778
0.6474	17.02	4698	1.8526	0.5556
0.0028	18.02	4959	1.4928	0.6667
1.1598	19.02	5220	2.2158	0.6222
0.0028	20.02	5481	1.4304	0.6444
0.6598	21.02	5742	1.8181	0.6222
1.3712	22.02	6003	2.1180	0.6
0.526	23.02	6264	1.9783	0.6
1.4996	24.02	6525	1.8767	0.6222
0.0002	25.02	6786	1.8231	0.6222
0.0062	26.02	7047	1.8086	0.6222
0.7647	27.02	7308	1.8100	0.6222
0.0002	28.02	7569	2.3123	0.6444
0.3838	29.02	7830	2.0822	0.6889
0.3837	30.02	8091	2.0536	0.7111
0.0003	31.02	8352	2.2919	0.6444
0.0002	32.02	8613	2.5788	0.6444
0.0007	33.02	8874	2.7998	0.6
0.0073	34.02	9135	2.9001	0.5556
0.1746	35.02	9396	2.5460	0.6667
0.0007	36.02	9657	2.4710	0.6889
0.0006	37.02	9918	2.1948	0.6667
0.0001	38.02	10179	2.4550	0.6667
0.3026	39.02	10440	2.1238	0.7333
0.0002	40.02	10701	2.3738	0.6222
0.7121	41.02	10962	2.8000	0.6
0.0001	42.02	11223	2.4507	0.6444
0.0001	43.02	11484	2.8657	0.6667
0.0001	44.02	11745	3.0916	0.5778
0.4721	45.02	12006	2.8460	0.6444
0.1128	46.02	12267	3.4904	0.6
0.0001	47.02	12528	3.4742	0.6
0.9105	48.02	12789	2.6414	0.6667
0.0001	49.02	13050	2.5768	0.6889

Framework versions

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