CTMAE-P2-V2-S5

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: 1.2006
• Accuracy: 0.75

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
1.5874	0.02	261	2.2577	0.5682
0.581	1.02	522	2.4954	0.5682
1.5552	2.02	783	2.2144	0.5682
0.7597	3.02	1044	2.1388	0.5682
1.8176	4.02	1305	1.5857	0.5682
0.9596	5.02	1566	1.9454	0.5682
0.8402	6.02	1827	2.0550	0.5682
1.0823	7.02	2088	1.7864	0.5682
1.0229	8.02	2349	1.8592	0.5682
0.7113	9.02	2610	1.4045	0.5682
1.3068	10.02	2871	1.4536	0.5682
1.7964	11.02	3132	1.8695	0.5682
1.6925	12.02	3393	0.7860	0.5682
0.3966	13.02	3654	2.1610	0.5682
0.0112	14.02	3915	2.7138	0.5682
0.5847	15.02	4176	0.8433	0.7045
0.6547	16.02	4437	1.7384	0.6136
0.7854	17.02	4698	1.3477	0.6818
1.0052	18.02	4959	1.4197	0.7045
1.4927	19.02	5220	2.2046	0.6136
0.5386	20.02	5481	1.2006	0.75
0.7256	21.02	5742	1.5015	0.7273
0.8462	22.02	6003	1.6405	0.6591
0.64	23.02	6264	2.2160	0.5682
1.0358	24.02	6525	2.6674	0.5682
0.0003	25.02	6786	3.2237	0.5682
1.449	26.02	7047	2.9910	0.5455
0.6425	27.02	7308	2.9668	0.5682
0.0038	28.02	7569	3.2074	0.5455
0.4198	29.02	7830	3.4554	0.5455
0.0002	30.02	8091	2.2222	0.6591
0.0087	31.02	8352	2.7093	0.5455
0.2823	32.02	8613	2.8994	0.5909
0.0009	33.02	8874	2.9261	0.5909
0.0064	34.02	9135	2.4037	0.6818
0.7506	35.02	9396	2.8436	0.6364
0.6686	36.02	9657	3.1198	0.5682
0.0089	37.02	9918	2.2353	0.6591
0.6753	38.02	10179	3.0288	0.6364
0.0003	39.02	10440	2.4052	0.6591
0.295	40.02	10701	3.7579	0.5682
0.0002	41.02	10962	3.3831	0.5909
0.5379	42.02	11223	3.5119	0.5455
0.0001	43.02	11484	3.3207	0.5909
0.0001	44.02	11745	3.1331	0.6136
0.0002	45.02	12006	3.1938	0.5909
0.0001	46.02	12267	3.2387	0.5909
0.6632	47.02	12528	3.3889	0.5909
0.2849	48.02	12789	3.3584	0.6364
0.0001	49.02	13050	3.2970	0.6136

Framework versions

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