8_CTMAE_5

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: 0.5903
• Accuracy: 0.8222

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: 21900

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4256	0.02	438	0.8835	0.4667
0.4763	1.02	876	2.8235	0.4667
1.6809	2.02	1314	1.2676	0.4667
0.5491	3.02	1752	2.7803	0.4667
0.7575	4.02	2190	1.2661	0.4667
1.7515	5.02	2628	1.6746	0.4667
0.7512	6.02	3066	2.2369	0.4667
0.2794	7.02	3504	0.7320	0.5111
2.0758	8.02	3942	1.2627	0.4667
0.663	9.02	4380	0.5200	0.7556
1.297	10.02	4818	2.1279	0.4667
1.2477	11.02	5256	0.5839	0.8
0.2851	12.02	5694	0.5903	0.8222
0.5838	13.02	6132	1.1180	0.6444
0.5258	14.02	6570	0.6902	0.8222
0.8624	15.02	7008	0.6784	0.8222
0.0246	16.02	7446	2.5170	0.4889
0.0039	17.02	7884	1.2381	0.7778
0.2381	18.02	8322	2.6533	0.5556
0.0014	19.02	8760	1.5067	0.7333
0.0014	20.02	9198	1.5134	0.7333
0.0019	21.02	9636	1.5027	0.6444
0.5404	22.02	10074	3.3171	0.4667
0.6011	23.02	10512	1.4685	0.7111
0.5212	24.02	10950	3.0006	0.4889
0.4134	25.02	11388	1.3394	0.7333
0.0588	26.02	11826	1.3787	0.7333
0.0141	27.02	12264	2.4794	0.6444
0.0004	28.02	12702	1.6891	0.7556
0.0008	29.02	13140	1.2273	0.7778
0.6532	30.02	13578	2.1206	0.7333
0.0004	31.02	14016	1.8691	0.7111
0.2953	32.02	14454	1.9442	0.7111
0.0932	33.02	14892	2.8354	0.6
0.0004	34.02	15330	2.9664	0.6444
0.0003	35.02	15768	1.8551	0.7556
0.0004	36.02	16206	3.3321	0.6222
0.0003	37.02	16644	2.4348	0.6667
0.0001	38.02	17082	2.2297	0.7556
0.0001	39.02	17520	2.5960	0.6889
0.0002	40.02	17958	1.8083	0.7778
0.0	41.02	18396	2.5271	0.6889
0.0003	42.02	18834	3.0164	0.6222
0.0	43.02	19272	2.8862	0.6222
0.0001	44.02	19710	2.9471	0.6444
0.0001	45.02	20148	3.5813	0.6222
0.0003	46.02	20586	3.3598	0.6
0.4492	47.02	21024	3.5946	0.6
0.0	48.02	21462	3.1944	0.6444
0.0	49.02	21900	3.2094	0.6444

Framework versions

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