MAE-CT-M1N0-M12_v8_split1_v3

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: 0.4942
• Accuracy: 0.7838

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: 4
• eval_batch_size: 4
• 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: 10500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6862	0.0068	71	0.6571	0.6622
0.665	1.0068	142	0.6370	0.6622
0.7033	2.0068	213	0.6254	0.6622
0.6524	3.0068	284	0.6091	0.6757
0.5611	4.0068	355	0.5565	0.6622
0.4274	5.0068	426	0.5154	0.7162
0.4797	6.0068	497	0.5644	0.6757
0.3758	7.0068	568	0.4942	0.7838
0.4243	8.0068	639	0.5252	0.7568
0.5133	9.0068	710	0.6873	0.6757
0.3709	10.0068	781	0.6555	0.7568
0.2793	11.0068	852	0.7140	0.7568
0.6153	12.0068	923	1.3006	0.6892
0.7185	13.0068	994	1.6663	0.6892
0.4609	14.0068	1065	1.3522	0.7162
0.236	15.0068	1136	1.2228	0.7297
0.0519	16.0068	1207	1.0973	0.7568
0.0026	17.0068	1278	1.4476	0.7432
0.357	18.0068	1349	1.4487	0.7432
0.4262	19.0068	1420	1.1604	0.7838
0.0021	20.0068	1491	1.7720	0.7027
0.0132	21.0068	1562	1.7388	0.7297
0.1451	22.0068	1633	1.7954	0.6892
0.0099	23.0068	1704	2.1619	0.7162
0.0001	24.0068	1775	1.6524	0.7297
0.0005	25.0068	1846	1.8499	0.7162
0.0388	26.0068	1917	1.8792	0.7027
0.1798	27.0068	1988	1.2951	0.7568
0.2354	28.0068	2059	1.5408	0.7297
0.0024	29.0068	2130	1.9224	0.7162
0.0018	30.0068	2201	2.5244	0.6486
0.1072	31.0068	2272	2.8444	0.6486
0.0664	32.0068	2343	1.8277	0.7297
0.0122	33.0068	2414	2.1148	0.7297
0.1118	34.0068	2485	1.5536	0.7703
0.1987	35.0068	2556	2.2923	0.7027
0.0012	36.0068	2627	2.6785	0.6622
0.0027	37.0068	2698	2.2400	0.6757
0.0002	38.0068	2769	2.2459	0.7162
0.0099	39.0068	2840	2.3601	0.6622
0.0071	40.0068	2911	2.0561	0.7297
0.0086	41.0068	2982	2.1898	0.7027
0.0131	42.0068	3053	2.6086	0.6757
0.0002	43.0068	3124	2.1400	0.6892
0.0001	44.0068	3195	2.2608	0.7027
0.0549	45.0068	3266	2.0129	0.7432
0.0	46.0068	3337	2.0018	0.7297
0.0001	47.0068	3408	1.7209	0.7838
0.31	48.0068	3479	2.1962	0.7027
0.0001	49.0068	3550	1.6650	0.7568
0.0	50.0068	3621	1.8843	0.7568
0.0	51.0068	3692	1.9398	0.7703
0.0	52.0068	3763	1.7851	0.7568
0.0001	53.0068	3834	1.9574	0.7162
0.0002	54.0068	3905	2.6200	0.6351
0.0	55.0068	3976	2.2333	0.7027
0.0001	56.0068	4047	2.7799	0.6757
0.0001	57.0068	4118	2.1935	0.7027
0.0188	58.0068	4189	2.2272	0.7162
0.1013	59.0068	4260	2.3607	0.7027
0.0001	60.0068	4331	2.1223	0.7432
0.0026	61.0068	4402	1.9220	0.7568
0.193	62.0068	4473	2.2254	0.7027
0.0002	63.0068	4544	2.2682	0.6622
0.0	64.0068	4615	2.6857	0.6892
0.0	65.0068	4686	2.3791	0.7297
0.0076	66.0068	4757	2.8393	0.6757
0.0043	67.0068	4828	1.9305	0.7162
0.0003	68.0068	4899	1.9944	0.7297
0.0	69.0068	4970	2.5842	0.7162
0.0001	70.0068	5041	2.6503	0.6622
0.0	71.0068	5112	2.7254	0.6757
0.0002	72.0068	5183	3.0429	0.6622
0.0	73.0068	5254	2.5716	0.7027
0.0671	74.0068	5325	2.5144	0.7027
0.0	75.0068	5396	2.8938	0.6622
0.0	76.0068	5467	2.8503	0.6622
0.0	77.0068	5538	2.8861	0.6622
0.0	78.0068	5609	2.8524	0.6892
0.0	79.0068	5680	2.7962	0.6757
0.0	80.0068	5751	2.8640	0.6622
0.0	81.0068	5822	2.8446	0.6757
0.0	82.0068	5893	2.6401	0.6892
0.0007	83.0068	5964	2.3987	0.7432
0.0	84.0068	6035	2.3642	0.7162
0.0	85.0068	6106	2.4710	0.6757
0.0004	86.0068	6177	3.0323	0.6486
0.0	87.0068	6248	3.0862	0.6351
0.2299	88.0068	6319	2.0283	0.7703
0.0	89.0068	6390	2.3752	0.6892
0.1842	90.0068	6461	2.2107	0.7568
0.0002	91.0068	6532	3.1361	0.6622
0.0	92.0068	6603	2.7366	0.7027
0.0	93.0068	6674	2.6850	0.6892
0.0	94.0068	6745	2.6965	0.6892
0.1894	95.0068	6816	2.5070	0.7162
0.0	96.0068	6887	2.5327	0.7162
0.0	97.0068	6958	2.8845	0.6892
0.0	98.0068	7029	2.0030	0.7838
0.1439	99.0068	7100	2.7892	0.6892
0.0	100.0068	7171	2.4622	0.7162
0.0016	101.0068	7242	2.4540	0.7297
0.0006	102.0068	7313	2.4853	0.7162
0.0	103.0068	7384	2.5101	0.7297
0.0	104.0068	7455	2.5136	0.7162
0.0	105.0068	7526	2.5028	0.7297
0.0039	106.0068	7597	2.6882	0.7162
0.0	107.0068	7668	2.8377	0.7162
0.0	108.0068	7739	2.8495	0.7162
0.0073	109.0068	7810	2.6625	0.7297
0.0	110.0068	7881	2.7063	0.7162
0.0	111.0068	7952	2.3949	0.7568
0.0	112.0068	8023	2.5956	0.7432
0.0001	113.0068	8094	2.9212	0.7027
0.0	114.0068	8165	2.8216	0.6892
0.0	115.0068	8236	2.8409	0.6892
0.0	116.0068	8307	2.8546	0.6892
0.0	117.0068	8378	2.8172	0.6892
0.0	118.0068	8449	2.4546	0.7432
0.0	119.0068	8520	2.3815	0.7568
0.0	120.0068	8591	2.4006	0.7432
0.0	121.0068	8662	2.4198	0.7432
0.0	122.0068	8733	2.4389	0.7432
0.0	123.0068	8804	2.4763	0.7432
0.0	124.0068	8875	2.4947	0.7432
0.0	125.0068	8946	2.5126	0.7297
0.0	126.0068	9017	2.5314	0.7297
0.0	127.0068	9088	2.5429	0.7297
0.0	128.0068	9159	2.5660	0.7297
0.0	129.0068	9230	2.5828	0.7162
0.0	130.0068	9301	2.5996	0.7162
0.0	131.0068	9372	2.6081	0.7162
0.0	132.0068	9443	2.6265	0.7162
0.0	133.0068	9514	2.6524	0.7162
0.0	134.0068	9585	2.6634	0.7162
0.0	135.0068	9656	2.6925	0.7162
0.0	136.0068	9727	2.7701	0.7162
0.0	137.0068	9798	2.7774	0.7027
0.0	138.0068	9869	2.7756	0.7162
0.0	139.0068	9940	2.7789	0.7162
0.0	140.0068	10011	2.7818	0.7162
0.0	141.0068	10082	2.7164	0.7162
0.0	142.0068	10153	2.9571	0.7027
0.0	143.0068	10224	2.9562	0.7027
0.0	144.0068	10295	2.9538	0.7027
0.0	145.0068	10366	2.9514	0.7027
0.0	146.0068	10437	2.9517	0.7027
0.0	147.006	10500	2.9518	0.7027

Framework versions

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