MAE-CT-M1N0-M12_v8_split2_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: 2.5210
• Accuracy: 0.7808

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
0.6593	0.0068	71	0.6615	0.6301
0.6079	1.0068	142	0.6573	0.6301
0.6428	2.0068	213	0.6510	0.6301
0.7179	3.0068	284	0.6321	0.6301
0.6131	4.0068	355	0.6464	0.6301
0.6769	5.0068	426	0.5554	0.6712
0.7054	6.0068	497	0.5056	0.7534
0.758	7.0068	568	0.5272	0.7397
0.5288	8.0068	639	0.5494	0.6986
0.3878	9.0068	710	0.5180	0.7260
0.2466	10.0068	781	0.7316	0.6986
0.8338	11.0068	852	1.1721	0.6712
0.603	12.0068	923	0.7357	0.7534
0.2309	13.0068	994	1.1961	0.6986
0.2656	14.0068	1065	1.1105	0.7123
0.5578	15.0068	1136	1.3217	0.7123
0.2875	16.0068	1207	1.2618	0.6986
0.4332	17.0068	1278	1.3750	0.7260
0.4794	18.0068	1349	1.5369	0.7260
0.3151	19.0068	1420	1.2066	0.7397
0.2433	20.0068	1491	1.6149	0.6986
0.1373	21.0068	1562	1.6916	0.7397
0.0864	22.0068	1633	2.3674	0.6849
0.2188	23.0068	1704	2.4041	0.6712
0.089	24.0068	1775	1.8638	0.7123
0.0911	25.0068	1846	2.0675	0.6986
0.137	26.0068	1917	1.8598	0.7123
0.1882	27.0068	1988	1.6897	0.7534
0.1562	28.0068	2059	2.6265	0.6849
0.0003	29.0068	2130	1.6721	0.6986
0.1783	30.0068	2201	2.0134	0.7260
0.0041	31.0068	2272	1.8352	0.7260
0.0001	32.0068	2343	2.3171	0.7123
0.0001	33.0068	2414	2.2544	0.6986
0.0443	34.0068	2485	2.0805	0.7260
0.0052	35.0068	2556	2.5061	0.6849
0.1231	36.0068	2627	2.2596	0.6438
0.0001	37.0068	2698	2.4168	0.7260
0.0001	38.0068	2769	2.4288	0.7123
0.0667	39.0068	2840	2.6743	0.6849
0.0001	40.0068	2911	2.4385	0.7123
0.0001	41.0068	2982	2.0221	0.7397
0.0561	42.0068	3053	2.1503	0.6712
0.005	43.0068	3124	3.0311	0.6575
0.0	44.0068	3195	2.2170	0.7260
0.2196	45.0068	3266	2.0672	0.6986
0.1848	46.0068	3337	2.5003	0.6575
0.2445	47.0068	3408	2.0344	0.6849
0.3096	48.0068	3479	2.4040	0.6986
0.0	49.0068	3550	2.3224	0.7123
0.0	50.0068	3621	2.6102	0.6849
0.2334	51.0068	3692	3.0010	0.6849
0.0001	52.0068	3763	2.2647	0.7397
0.0001	53.0068	3834	2.2806	0.7123
0.0	54.0068	3905	2.5543	0.7260
0.0	55.0068	3976	2.6203	0.6986
0.2117	56.0068	4047	2.5486	0.6849
0.0001	57.0068	4118	2.2072	0.6986
0.0001	58.0068	4189	2.4930	0.6986
0.0001	59.0068	4260	2.3262	0.7123
0.3114	60.0068	4331	3.0585	0.6575
0.0001	61.0068	4402	2.0491	0.7260
0.0057	62.0068	4473	2.0623	0.7397
0.0	63.0068	4544	2.5215	0.6986
0.0319	64.0068	4615	2.4648	0.7123
0.2879	65.0068	4686	2.4885	0.6712
0.0001	66.0068	4757	1.8654	0.6986
0.0001	67.0068	4828	2.3100	0.6986
0.0001	68.0068	4899	2.0873	0.7260
0.1614	69.0068	4970	2.0189	0.7260
0.0001	70.0068	5041	2.5160	0.7123
0.0114	71.0068	5112	2.0018	0.7260
0.0823	72.0068	5183	2.2905	0.7260
0.0001	73.0068	5254	2.2782	0.7260
0.0	74.0068	5325	2.4495	0.6986
0.3044	75.0068	5396	2.4417	0.7123
0.0	76.0068	5467	2.5168	0.6849
0.0007	77.0068	5538	2.9406	0.6986
0.0001	78.0068	5609	2.6533	0.7123
0.0	79.0068	5680	2.4312	0.6986
0.0	80.0068	5751	2.5024	0.7123
0.0	81.0068	5822	2.4178	0.6986
0.0	82.0068	5893	2.5872	0.7123
0.0	83.0068	5964	2.0274	0.7671
0.1991	84.0068	6035	2.5663	0.6986
0.0	85.0068	6106	2.6205	0.6849
0.1705	86.0068	6177	2.7275	0.6575
0.0	87.0068	6248	2.9716	0.6438
0.0	88.0068	6319	2.9101	0.6438
0.0	89.0068	6390	2.4764	0.6986
0.0	90.0068	6461	2.5322	0.6986
0.0	91.0068	6532	2.7223	0.6986
0.0	92.0068	6603	2.6965	0.6849
0.0	93.0068	6674	2.6896	0.6849
0.0001	94.0068	6745	2.7115	0.7123
0.0	95.0068	6816	2.6126	0.6849
0.0	96.0068	6887	2.6572	0.6986
0.0	97.0068	6958	2.7067	0.6986
0.0	98.0068	7029	3.0455	0.6712
0.0	99.0068	7100	2.8097	0.6849
0.0	100.0068	7171	2.8568	0.6849
0.0	101.0068	7242	2.9188	0.6849
0.0	102.0068	7313	2.9678	0.6849
0.0	103.0068	7384	3.0187	0.6712
0.0	104.0068	7455	3.0423	0.6712
0.0	105.0068	7526	3.0592	0.6712
0.0	106.0068	7597	3.0764	0.6712
0.0	107.0068	7668	3.0949	0.6712
0.0	108.0068	7739	3.1251	0.6712
0.0	109.0068	7810	3.1508	0.6712
0.0	110.0068	7881	2.7330	0.6986
0.0	111.0068	7952	2.5366	0.7534
0.0	112.0068	8023	2.7911	0.7123
0.0	113.0068	8094	2.6362	0.7260
0.0	114.0068	8165	2.4667	0.6849
0.0	115.0068	8236	2.5247	0.6849
0.0	116.0068	8307	2.6152	0.6712
0.0	117.0068	8378	2.6153	0.7260
0.0	118.0068	8449	2.5005	0.7397
0.0	119.0068	8520	2.5096	0.7397
0.0	120.0068	8591	2.5173	0.7397
0.0	121.0068	8662	2.5226	0.7397
0.0	122.0068	8733	2.5313	0.7397
0.0	123.0068	8804	2.6165	0.7397
0.0	124.0068	8875	2.4491	0.7260
0.0	125.0068	8946	2.3421	0.7671
0.0	126.0068	9017	2.3381	0.7671
0.0	127.0068	9088	2.3615	0.7671
0.0	128.0068	9159	2.4429	0.7534
0.0	129.0068	9230	2.6266	0.7397
0.0	130.0068	9301	2.6281	0.7397
0.0	131.0068	9372	2.6321	0.7260
0.0	132.0068	9443	2.6350	0.7260
0.0	133.0068	9514	2.5210	0.7808
0.0	134.0068	9585	2.5572	0.7671
0.0	135.0068	9656	2.5419	0.7671
0.0	136.0068	9727	2.5428	0.7534
0.0	137.0068	9798	2.5649	0.7534
0.0	138.0068	9869	2.7969	0.7123
0.0	139.0068	9940	2.8026	0.7123
0.0	140.0068	10011	2.8066	0.7123
0.0	141.0068	10082	2.6293	0.7397
0.0	142.0068	10153	2.6859	0.7260
0.0	143.0068	10224	2.6886	0.7260
0.0	144.0068	10295	2.7223	0.7260
0.0	145.0068	10366	2.7872	0.7260
0.0	146.0068	10437	2.7887	0.7260
0.0	147.006	10500	2.7888	0.7260

Framework versions

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