CTMAE-P2-V4-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: 1.4928
• Accuracy: 0.8043

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: 13050

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6322	0.0100	131	0.6958	0.5435
0.3448	1.0100	262	1.4634	0.5435
1.1225	2.0100	393	2.0266	0.5435
0.7246	3.0100	524	0.9006	0.5435
1.2784	4.0100	655	1.6206	0.5435
0.7234	5.0100	786	1.7217	0.5435
0.7544	6.0100	917	1.4504	0.5435
1.732	7.0100	1048	1.1581	0.5435
0.8227	8.0100	1179	1.9053	0.5435
0.7839	9.0100	1310	0.9410	0.5435
0.8302	10.0100	1441	1.5093	0.5435
0.6264	11.0100	1572	1.7408	0.5435
0.5032	12.0100	1703	0.7154	0.5
1.1847	13.0100	1834	1.1743	0.5435
0.9721	14.0100	1965	1.7714	0.5435
0.6003	15.0100	2096	0.8652	0.5870
0.4912	16.0100	2227	1.7541	0.5435
0.8106	17.0100	2358	1.0464	0.5652
1.2365	18.0100	2489	0.7472	0.6739
1.7469	19.0100	2620	1.3125	0.6304
0.2345	20.0100	2751	1.0220	0.6087
0.483	21.0100	2882	1.2559	0.6087
1.5409	22.0100	3013	1.6619	0.5435
1.1284	23.0100	3144	1.0519	0.6739
0.4471	24.0100	3275	2.1155	0.5652
0.2323	25.0100	3406	1.6991	0.6304
0.871	26.0100	3537	1.4254	0.6957
0.4976	27.0100	3668	1.8011	0.6304
0.5621	28.0100	3799	1.6148	0.6739
0.9854	29.0100	3930	1.4576	0.6522
0.0018	30.0100	4061	1.5995	0.7174
0.3031	31.0100	4192	1.5070	0.6957
0.8871	32.0100	4323	1.7620	0.6522
0.6212	33.0100	4454	1.7319	0.6739
0.5674	34.0100	4585	1.8520	0.6739
0.2845	35.0100	4716	1.8629	0.6522
0.1611	36.0100	4847	1.7524	0.6522
0.0779	37.0100	4978	1.5949	0.6739
0.6805	38.0100	5109	2.1198	0.6739
0.0297	39.0100	5240	2.1019	0.6739
0.0005	40.0100	5371	2.3706	0.6739
0.4209	41.0100	5502	1.3258	0.6957
0.2219	42.0100	5633	1.9883	0.6957
0.0184	43.0100	5764	2.0343	0.6522
0.001	44.0100	5895	1.9996	0.6957
0.4252	45.0100	6026	1.9136	0.6522
0.0456	46.0100	6157	1.9553	0.6739
0.375	47.0100	6288	1.9227	0.6957
0.6046	48.0100	6419	2.6295	0.6087
0.2836	49.0100	6550	1.7961	0.7174
0.1522	50.0100	6681	1.3961	0.7826
0.6705	51.0100	6812	1.7068	0.7391
0.0005	52.0100	6943	1.7986	0.7391
0.02	53.0100	7074	1.6991	0.7609
0.0037	54.0100	7205	1.5867	0.7391
0.2488	55.0100	7336	1.4928	0.8043
0.0297	56.0100	7467	1.8699	0.7174
0.0003	57.0100	7598	2.1014	0.7174
0.0008	58.0100	7729	1.9651	0.6739
0.2982	59.0100	7860	2.5969	0.6522
0.3197	60.0100	7991	2.3923	0.6087
0.012	61.0100	8122	2.4473	0.6522
0.0002	62.0100	8253	2.1692	0.6957
0.0002	63.0100	8384	2.3358	0.6739
0.0001	64.0100	8515	2.6785	0.6739
0.364	65.0100	8646	2.7085	0.6522
0.0001	66.0100	8777	2.8955	0.6522
0.0002	67.0100	8908	2.2053	0.7391
0.0002	68.0100	9039	2.6436	0.6739
0.0001	69.0100	9170	2.6494	0.6739
0.0046	70.0100	9301	2.2621	0.7391
0.0001	71.0100	9432	2.9285	0.6739
0.0001	72.0100	9563	2.4097	0.6957
0.0001	73.0100	9694	2.8739	0.6304
0.0004	74.0100	9825	2.8154	0.6304
0.3257	75.0100	9956	2.3350	0.6957
0.0001	76.0100	10087	1.9011	0.7391
0.0001	77.0100	10218	2.3655	0.7174
0.0001	78.0100	10349	2.6572	0.6739
0.0001	79.0100	10480	2.6350	0.6739
0.4185	80.0100	10611	2.4854	0.7174
0.0001	81.0100	10742	2.4658	0.7391
0.0	82.0100	10873	2.6691	0.6957
0.0001	83.0100	11004	2.7930	0.5870
0.0001	84.0100	11135	2.5645	0.6957
0.0001	85.0100	11266	2.5759	0.7174
0.0	86.0100	11397	2.6901	0.6957
0.0	87.0100	11528	2.6050	0.6957
0.0	88.0100	11659	3.0276	0.6087
0.0001	89.0100	11790	2.9324	0.6739
0.0	90.0100	11921	2.9194	0.6739
0.0	91.0100	12052	2.5726	0.7391
0.0	92.0100	12183	2.8832	0.6739
0.0001	93.0100	12314	3.0274	0.6304
0.0001	94.0100	12445	2.8242	0.6957
0.0	95.0100	12576	2.7715	0.6957
0.0	96.0100	12707	2.7907	0.6957
0.4392	97.0100	12838	2.7856	0.6957
0.0	98.0100	12969	2.7755	0.6957
0.0	99.0062	13050	2.7569	0.6957

Framework versions

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