CTMAE-P2-V3-3G-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: 0.7553
• Accuracy: 0.8409

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: 2
• eval_batch_size: 2
• 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: 6500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6815	0.0202	131	0.7046	0.5227
0.4922	1.0202	262	0.9263	0.5227
0.6484	2.0202	393	0.7719	0.5227
0.6707	3.0202	524	0.7503	0.5227
1.1086	4.0202	655	0.7398	0.5227
0.6679	5.0202	786	0.6244	0.5227
0.7521	6.0202	917	0.7773	0.5227
1.7822	7.0202	1048	1.4883	0.5227
0.7714	8.0202	1179	1.0682	0.5455
0.5244	9.0202	1310	0.7003	0.5227
0.7346	10.0202	1441	1.0285	0.5
0.4019	11.0202	1572	0.6361	0.6364
0.7769	12.0202	1703	0.4666	0.7727
0.6358	13.0202	1834	1.3155	0.6364
0.5129	14.0202	1965	1.1485	0.6818
1.1003	15.0202	2096	0.5102	0.7955
0.7943	16.0202	2227	0.6434	0.7273
0.363	17.0202	2358	0.9532	0.7955
0.2225	18.0202	2489	0.6883	0.7273
0.2107	19.0202	2620	0.6559	0.8182
1.0048	20.0202	2751	0.9784	0.7727
0.4285	21.0202	2882	0.9721	0.7273
0.0459	22.0202	3013	0.6200	0.8182
0.9143	23.0202	3144	1.4102	0.6818
0.426	24.0202	3275	1.4732	0.6591
0.8278	25.0202	3406	1.0037	0.7955
0.7064	26.0202	3537	0.7553	0.8409
0.6217	27.0202	3668	1.5788	0.7045
0.6104	28.0202	3799	1.4799	0.7045
0.4625	29.0202	3930	2.0381	0.6136
0.1935	30.0202	4061	1.8624	0.6136
0.3657	31.0202	4192	1.4570	0.7273
0.0952	32.0202	4323	1.2123	0.8182
0.6802	33.0202	4454	1.1155	0.8409
0.1602	34.0202	4585	1.0990	0.8182
0.269	35.0202	4716	1.3373	0.7727
0.2087	36.0202	4847	1.0545	0.7955
0.4447	37.0202	4978	1.1182	0.7955
0.803	38.0202	5109	1.9385	0.7045
0.685	39.0202	5240	1.5079	0.75
0.001	40.0202	5371	1.3848	0.7955
0.7371	41.0202	5502	1.8426	0.7273
0.0008	42.0202	5633	1.9710	0.7273
0.2964	43.0202	5764	2.2087	0.7045
0.0008	44.0202	5895	1.5878	0.7727
0.0005	45.0202	6026	1.8426	0.7273
0.4393	46.0202	6157	1.7193	0.75
0.0036	47.0202	6288	1.9741	0.7273
0.0005	48.0202	6419	1.8383	0.75
0.1776	49.0125	6500	1.8362	0.75

Framework versions

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