CTMAE-P2-V2-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: 1.4945
• Accuracy: 0.6667

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
0.6454	0.0202	131	0.8373	0.5556
0.4683	1.0202	262	2.0593	0.5556
1.185	2.0202	393	2.0221	0.5556
0.6278	3.0202	524	0.8492	0.5556
1.3943	4.0202	655	1.5432	0.5556
0.8242	5.0202	786	1.6719	0.5556
0.8047	6.0202	917	1.7035	0.5556
1.4137	7.0202	1048	0.9401	0.5333
0.763	8.0202	1179	1.8927	0.5556
0.7333	9.0202	1310	0.9205	0.6
0.7436	10.0202	1441	1.5193	0.6222
0.3139	11.0202	1572	1.1921	0.6222
0.3918	12.0202	1703	0.7039	0.6222
0.667	13.0202	1834	0.8153	0.6222
0.967	14.0202	1965	1.5073	0.5556
0.0771	15.0202	2096	1.5141	0.5778
0.5622	16.0202	2227	1.5615	0.6
1.0141	17.0202	2358	1.6540	0.6222
2.1512	18.0202	2489	1.2866	0.6444
1.0352	19.0202	2620	1.9383	0.6
0.2068	20.0202	2751	1.8477	0.6
0.2804	21.0202	2882	1.4945	0.6667
0.4271	22.0202	3013	1.8007	0.6222
0.8258	23.0202	3144	1.8842	0.5778
0.6762	24.0202	3275	1.8649	0.6222
1.1151	25.0202	3406	2.7759	0.5333
0.1755	26.0202	3537	2.0492	0.6444
0.4496	27.0202	3668	2.2949	0.5556
0.4336	28.0202	3799	2.3240	0.5333
0.9503	29.0202	3930	2.0642	0.6222
0.4413	30.0202	4061	2.3833	0.5556
0.0132	31.0202	4192	2.5514	0.5778
0.2513	32.0202	4323	2.4046	0.5778
0.4845	33.0202	4454	2.5703	0.6
0.8916	34.0202	4585	2.5372	0.6
0.3173	35.0202	4716	2.6754	0.6
0.4552	36.0202	4847	2.6613	0.5778
0.0155	37.0202	4978	2.4057	0.6222
0.6358	38.0202	5109	2.4891	0.6
0.834	39.0202	5240	2.6045	0.6222
0.0008	40.0202	5371	2.5713	0.6222
0.523	41.0202	5502	2.6842	0.5778
0.5313	42.0202	5633	2.7778	0.5556
0.0002	43.0202	5764	2.5852	0.6222
0.0002	44.0202	5895	2.2989	0.6444
0.0004	45.0202	6026	2.4952	0.5778
0.1427	46.0202	6157	2.2855	0.6667
0.1096	47.0202	6288	2.5532	0.5778
0.0048	48.0202	6419	2.5509	0.5778
0.0004	49.0125	6500	2.5656	0.5778

Framework versions

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