CTMAE-P2-V3-3G-S4

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: 0.8000
• 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: 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
0.649	0.0202	131	0.8224	0.3778
0.4184	1.0202	262	1.4642	0.3778
0.7051	2.0202	393	1.1787	0.3778
0.5842	3.0202	524	0.9482	0.3778
0.9705	4.0202	655	1.0651	0.3778
0.6554	5.0202	786	0.9533	0.3778
0.8662	6.0202	917	1.2020	0.3778
1.8391	7.0202	1048	1.3808	0.3778
0.492	8.0202	1179	1.0200	0.3778
0.3752	9.0202	1310	0.8383	0.5778
0.8025	10.0202	1441	1.2180	0.5111
0.6779	11.0202	1572	1.9618	0.3778
0.598	12.0202	1703	0.7358	0.7111
1.4615	13.0202	1834	0.9323	0.6222
0.6769	14.0202	1965	1.1934	0.6222
0.398	15.0202	2096	1.3051	0.6444
0.2597	16.0202	2227	0.6407	0.7333
0.9731	17.0202	2358	0.8000	0.8222
0.9509	18.0202	2489	1.0755	0.7111
0.3026	19.0202	2620	1.6900	0.6444
0.3618	20.0202	2751	1.8778	0.6
0.2526	21.0202	2882	2.0385	0.6
1.7509	22.0202	3013	1.9079	0.6
1.0213	23.0202	3144	1.3900	0.7333
0.1836	24.0202	3275	1.8195	0.6222
0.2277	25.0202	3406	2.1068	0.5778
0.3344	26.0202	3537	2.1472	0.6222
0.5114	27.0202	3668	2.5289	0.5778
0.0018	28.0202	3799	2.3118	0.6444
0.1634	29.0202	3930	2.7060	0.5778
0.2339	30.0202	4061	2.3984	0.6222
0.0997	31.0202	4192	3.5809	0.5111
0.8974	32.0202	4323	2.8206	0.5556
0.4314	33.0202	4454	3.5183	0.4667
0.0006	34.0202	4585	2.4841	0.6222
0.185	35.0202	4716	3.3856	0.5556
0.1699	36.0202	4847	2.9514	0.5778
0.4543	37.0202	4978	2.5094	0.6444
0.0003	38.0202	5109	2.2945	0.6667
0.3989	39.0202	5240	2.7996	0.6
0.0853	40.0202	5371	3.2383	0.5778
0.2906	41.0202	5502	3.0334	0.6
0.0006	42.0202	5633	2.9619	0.6
0.0001	43.0202	5764	3.3620	0.6
0.0021	44.0202	5895	3.2390	0.5778
0.1838	45.0202	6026	3.3982	0.5778
0.0028	46.0202	6157	3.3721	0.5778
0.0002	47.0202	6288	3.4766	0.5778
0.067	48.0202	6419	3.3522	0.5778
0.0001	49.0125	6500	3.3565	0.5778

Framework versions

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