MAE-CT-M1N0-M12_v8_split3_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: 1.0548
• Accuracy: 0.8462

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: 10350

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6676	0.0068	70	0.6765	0.6538
0.6735	1.0068	140	0.6379	0.6667
0.6391	2.0068	210	0.6261	0.6667
0.8085	3.0068	280	0.6050	0.6667
0.4774	4.0068	350	0.5725	0.6538
0.5668	5.0068	420	0.5514	0.7179
0.6607	6.0068	490	0.4834	0.7564
1.0769	7.0068	560	0.7450	0.6667
0.5425	8.0068	630	0.4623	0.7949
0.2959	9.0068	700	0.5113	0.7564
0.7674	10.0068	770	0.5330	0.7308
0.3015	11.0068	840	0.6790	0.7436
0.6252	12.0068	910	1.6537	0.6667
0.4554	13.0068	980	0.8944	0.7564
0.364	14.0068	1050	0.8103	0.7179
0.444	15.0068	1120	0.7103	0.7821
0.104	16.0068	1190	0.9000	0.7179
0.5647	17.0068	1260	1.4782	0.7051
0.783	18.0068	1330	0.8539	0.8205
0.5938	19.0068	1400	0.8426	0.7564
0.5341	20.0068	1470	0.9862	0.7949
0.3391	21.0068	1540	1.1159	0.7692
0.2071	22.0068	1610	1.5833	0.7692
0.1159	23.0068	1680	1.0205	0.8205
0.1579	24.0068	1750	1.4633	0.7692
0.1042	25.0068	1820	1.5864	0.7564
0.1466	26.0068	1890	1.2990	0.7692
0.0006	27.0068	1960	1.5597	0.7564
0.001	28.0068	2030	1.6434	0.7564
0.0096	29.0068	2100	1.4161	0.7821
0.0015	30.0068	2170	1.3385	0.7949
0.3627	31.0068	2240	1.7864	0.7436
0.1541	32.0068	2310	1.5618	0.7821
0.1285	33.0068	2380	1.3062	0.7949
0.2193	34.0068	2450	1.5554	0.7821
0.0002	35.0068	2520	1.5445	0.7949
0.0003	36.0068	2590	1.7189	0.7821
0.092	37.0068	2660	1.4980	0.7692
0.2403	38.0068	2730	2.0993	0.7436
0.0001	39.0068	2800	1.8646	0.7436
0.0001	40.0068	2870	1.6642	0.7436
0.2108	41.0068	2940	1.4566	0.7949
0.1085	42.0068	3010	1.2915	0.7821
0.0019	43.0068	3080	1.2691	0.8333
0.0072	44.0068	3150	2.2703	0.7179
0.0002	45.0068	3220	1.5151	0.7564
0.4057	46.0068	3290	1.4170	0.7564
0.0862	47.0068	3360	2.1072	0.7308
0.0478	48.0068	3430	1.6826	0.7949
0.001	49.0068	3500	1.4349	0.7564
0.0001	50.0068	3570	1.4423	0.7949
0.0056	51.0068	3640	2.2833	0.7051
0.0004	52.0068	3710	1.0548	0.8462
0.1768	53.0068	3780	1.2974	0.8077
0.0001	54.0068	3850	1.2797	0.8333
0.1027	55.0068	3920	1.5159	0.8077
0.1638	56.0068	3990	1.9403	0.7564
0.0001	57.0068	4060	1.5075	0.8077
0.0003	58.0068	4130	2.1291	0.7436
0.004	59.0068	4200	1.6104	0.7949
0.0214	60.0068	4270	1.7017	0.7949
0.0	61.0068	4340	1.6485	0.8205
0.0	62.0068	4410	1.6668	0.8077
0.1604	63.0068	4480	1.7437	0.7949
0.0002	64.0068	4550	1.6770	0.7821
0.0	65.0068	4620	1.7766	0.7821
0.0214	66.0068	4690	1.6351	0.7821
0.0	67.0068	4760	1.6878	0.7821
0.0	68.0068	4830	1.8764	0.7564
0.2082	69.0068	4900	1.7799	0.7564
0.0	70.0068	4970	1.7388	0.7949
0.0	71.0068	5040	1.6719	0.7949
0.0001	72.0068	5110	1.6066	0.7949
0.0001	73.0068	5180	2.1181	0.7564
0.0	74.0068	5250	2.1773	0.7564
0.0	75.0068	5320	2.5632	0.7179
0.0	76.0068	5390	1.1549	0.8077
0.0006	77.0068	5460	2.2296	0.7436
0.0	78.0068	5530	2.5382	0.7308
0.0001	79.0068	5600	1.5726	0.7821
0.0002	80.0068	5670	1.5704	0.8077
0.0199	81.0068	5740	1.5503	0.8077
0.0001	82.0068	5810	1.3654	0.8077
0.0	83.0068	5880	1.4727	0.8077
0.0001	84.0068	5950	2.2248	0.7179
0.0	85.0068	6020	1.7188	0.7821
0.0101	86.0068	6090	1.9271	0.7436
0.0001	87.0068	6160	1.5356	0.8333
0.0	88.0068	6230	1.6683	0.8077
0.0	89.0068	6300	2.5566	0.7308
0.0	90.0068	6370	1.7776	0.8077
0.0028	91.0068	6440	1.7494	0.7949
0.0	92.0068	6510	1.6141	0.8077
0.0	93.0068	6580	2.8243	0.7179
0.0	94.0068	6650	1.9026	0.7051
0.0002	95.0068	6720	2.2368	0.6667
0.0998	96.0068	6790	2.4985	0.7179
0.0	97.0068	6860	2.0686	0.7436
0.0074	98.0068	6930	2.7098	0.7051
0.0	99.0068	7000	2.2765	0.7308
0.0	100.0068	7070	2.2793	0.7308
0.0	101.0068	7140	2.2027	0.7308
0.0	102.0068	7210	2.2387	0.7308
0.0	103.0068	7280	2.1971	0.7308
0.0	104.0068	7350	2.3246	0.7179
0.0	105.0068	7420	1.5935	0.8077
0.0	106.0068	7490	1.4796	0.8077
0.0001	107.0068	7560	1.7052	0.7821
0.0	108.0068	7630	1.6022	0.8077
0.0002	109.0068	7700	1.6749	0.8077
0.0	110.0068	7770	1.7948	0.7436
0.0	111.0068	7840	1.8455	0.7949
0.0	112.0068	7910	1.8600	0.7949
0.0	113.0068	7980	1.8183	0.7949
0.0	114.0068	8050	1.7862	0.7949
0.0	115.0068	8120	1.8597	0.7821
0.0	116.0068	8190	1.8203	0.7821
0.0	117.0068	8260	1.8343	0.7949
0.0	118.0068	8330	1.8417	0.7949
0.0	119.0068	8400	1.7663	0.7821
0.0	120.0068	8470	1.9611	0.7821
0.0	121.0068	8540	1.9584	0.7949
0.0	122.0068	8610	1.5671	0.8205
0.0	123.0068	8680	2.3456	0.7564
0.0	124.0068	8750	2.3453	0.7564
0.0	125.0068	8820	2.4120	0.7436
0.0	126.0068	8890	2.3774	0.7436
0.0	127.0068	8960	2.3609	0.7436
0.0	128.0068	9030	2.3531	0.7564
0.0	129.0068	9100	1.9910	0.7821
0.0	130.0068	9170	2.0032	0.7821
0.0	131.0068	9240	2.0645	0.7692
0.0	132.0068	9310	2.0598	0.7692
0.0	133.0068	9380	2.0594	0.7692
0.0	134.0068	9450	2.0568	0.7692
0.0	135.0068	9520	2.0522	0.7821
0.0	136.0068	9590	1.9971	0.7564
0.0	137.0068	9660	1.9977	0.7564
0.0	138.0068	9730	2.0896	0.7692
0.0	139.0068	9800	2.1550	0.7692
0.0	140.0068	9870	2.1875	0.7692
0.0	141.0068	9940	2.1874	0.7692
0.0	142.0068	10010	2.1822	0.7692
0.0	143.0068	10080	2.1818	0.7692
0.0	144.0068	10150	2.1806	0.7692
0.0	145.0068	10220	2.1803	0.7692
0.0	146.0068	10290	2.1804	0.7692
0.0	147.0058	10350	2.1803	0.7692

Framework versions

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