videomae-base-ipm_all_videos_gb2

This model is a fine-tuned version of MCG-NJU/videomae-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.5087
• Accuracy: 0.6957

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 9600

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.4413	0.01	60	2.5408	0.0696
2.3949	1.01	120	2.5420	0.0435
2.5429	2.01	180	2.5626	0.0957
2.4678	3.01	240	2.5721	0.0783
2.3535	4.01	300	2.5703	0.0783
2.3525	5.01	360	2.5966	0.0609
2.2312	6.01	420	2.3565	0.1913
2.0797	7.01	480	2.0738	0.1826
2.1423	8.01	540	2.0182	0.2435
1.8594	9.01	600	2.9555	0.0957
2.2635	10.01	660	2.1157	0.1565
2.0527	11.01	720	1.7646	0.2870
1.4499	12.01	780	2.2083	0.2696
1.3273	13.01	840	2.4202	0.2609
1.4349	14.01	900	1.9185	0.3043
1.476	15.01	960	2.1430	0.2261
1.2768	16.01	1020	1.6487	0.3391
1.2488	17.01	1080	1.7203	0.3130
1.5273	18.01	1140	1.9167	0.2783
1.6865	19.01	1200	2.1734	0.2522
1.448	20.01	1260	2.2406	0.3043
1.3169	21.01	1320	1.8596	0.2261
1.3004	22.01	1380	2.1954	0.2957
1.2201	23.01	1440	1.8007	0.3391
1.7577	24.01	1500	2.2078	0.2696
1.3741	25.01	1560	1.8426	0.3217
1.3676	26.01	1620	1.8888	0.3826
1.5892	27.01	1680	2.0376	0.3043
1.1962	28.01	1740	1.7738	0.3130
1.4768	29.01	1800	1.3115	0.4522
1.4112	30.01	1860	1.4297	0.3739
1.2148	31.01	1920	1.9232	0.2870
1.1125	32.01	1980	1.8406	0.3217
0.9814	33.01	2040	2.0529	0.3913
1.0787	34.01	2100	1.5659	0.3391
1.4073	35.01	2160	1.7671	0.3478
1.2131	36.01	2220	1.5678	0.3130
1.1894	37.01	2280	1.5435	0.4087
1.2001	38.01	2340	1.6149	0.3913
1.518	39.01	2400	1.7457	0.2957
1.1231	40.01	2460	1.7148	0.4
0.9362	41.01	2520	1.5611	0.4174
1.1348	42.01	2580	1.2901	0.3826
0.9504	43.01	2640	1.4024	0.4
1.2008	44.01	2700	1.6685	0.4609
1.0468	45.01	2760	1.6202	0.4174
0.7304	46.01	2820	1.4007	0.4522
0.8522	47.01	2880	1.5439	0.4174
0.9106	48.01	2940	1.6536	0.4783
0.7837	49.01	3000	1.4113	0.4609
0.6869	50.01	3060	1.2071	0.5391
0.8787	51.01	3120	1.3023	0.5130
0.8072	52.01	3180	1.2058	0.6
0.9491	53.01	3240	1.5370	0.4957
0.7642	54.01	3300	1.2301	0.5652
0.6676	55.01	3360	1.4549	0.5391
0.8502	56.01	3420	1.6117	0.4522
1.0006	57.01	3480	1.3982	0.4957
0.8304	58.01	3540	1.3233	0.4783
0.9832	59.01	3600	1.2982	0.5478
0.3973	60.01	3660	1.3903	0.5478
0.9487	61.01	3720	1.4241	0.5304
0.9319	62.01	3780	1.4913	0.5565
0.6713	63.01	3840	1.4731	0.5826
0.7139	64.01	3900	1.0942	0.6870
0.7852	65.01	3960	1.2570	0.6348
1.0018	66.01	4020	1.1249	0.5913
0.7371	67.01	4080	1.4665	0.5565
0.6106	68.01	4140	1.7390	0.4957
0.8815	69.01	4200	1.5044	0.5652
0.6724	70.01	4260	1.8060	0.4957
0.5907	71.01	4320	1.5552	0.5391
0.6218	72.01	4380	1.6037	0.5826
0.7698	73.01	4440	1.4280	0.5913
0.6719	74.01	4500	1.6870	0.5565
0.3956	75.01	4560	1.6326	0.5217
0.6272	76.01	4620	1.3282	0.6
0.4354	77.01	4680	1.5181	0.5913
0.8649	78.01	4740	1.4137	0.5913
0.48	79.01	4800	1.6439	0.5913
0.9693	80.01	4860	1.6453	0.5739
0.3872	81.01	4920	1.5209	0.6696
0.913	82.01	4980	1.5002	0.6435
0.7185	83.01	5040	1.8319	0.5478
1.0149	84.01	5100	1.5270	0.5826
0.3811	85.01	5160	1.3813	0.6609
0.4902	86.01	5220	1.3160	0.6348
1.2717	87.01	5280	1.5052	0.6696
0.5379	88.01	5340	1.4357	0.6870
0.7101	89.01	5400	1.7699	0.5739
0.6517	90.01	5460	1.3428	0.6609
0.6213	91.01	5520	1.4725	0.6087
0.6995	92.01	5580	1.2645	0.6435
0.3997	93.01	5640	1.5827	0.5652
0.7778	94.01	5700	1.2344	0.7304
0.5093	95.01	5760	1.2908	0.6957
0.6022	96.01	5820	1.3528	0.6609
0.508	97.01	5880	1.4460	0.6783
0.4772	98.01	5940	1.1836	0.7478
0.8776	99.01	6000	1.4956	0.6435
0.7514	100.01	6060	1.4904	0.6609
0.1734	101.01	6120	1.6757	0.6087
0.5279	102.01	6180	1.8148	0.5913
0.2101	103.01	6240	1.4176	0.6348
0.6081	104.01	6300	1.7604	0.5913
0.2781	105.01	6360	1.7557	0.6087
0.2321	106.01	6420	1.3726	0.6696
0.4503	107.01	6480	1.6582	0.6348
0.4361	108.01	6540	2.0009	0.5913
0.4934	109.01	6600	1.9722	0.5217
0.3898	110.01	6660	1.5016	0.6696
0.4286	111.01	6720	1.5307	0.6783
0.2792	112.01	6780	1.5770	0.6696
0.2254	113.01	6840	1.7076	0.6522
0.1739	114.01	6900	2.0225	0.5826
0.1951	115.01	6960	1.8448	0.6174
0.614	116.01	7020	1.5507	0.6696
0.6894	117.01	7080	1.5430	0.6609
0.9059	118.01	7140	1.6563	0.6696
0.4592	119.01	7200	1.5566	0.7043
0.3895	120.01	7260	1.5251	0.7130
0.4897	121.01	7320	1.7417	0.6696
0.5362	122.01	7380	1.5845	0.6783
0.4484	123.01	7440	1.6405	0.6870
0.557	124.01	7500	1.5133	0.7130
0.4878	125.01	7560	1.3845	0.7391
0.2704	126.01	7620	1.4704	0.6957
0.7636	127.01	7680	1.4413	0.6957
0.4196	128.01	7740	1.4106	0.7043
0.5835	129.01	7800	1.2571	0.7391
0.6156	130.01	7860	1.8000	0.6609
0.3074	131.01	7920	1.7324	0.6435
0.4697	132.01	7980	1.5218	0.7043
0.2968	133.01	8040	1.3640	0.7391
0.452	134.01	8100	1.4916	0.7217
0.2699	135.01	8160	1.6554	0.6957
0.3889	136.01	8220	1.5015	0.7391
0.5006	137.01	8280	1.4134	0.7391
0.135	138.01	8340	1.3987	0.7565
0.3882	139.01	8400	1.4364	0.7304
0.194	140.01	8460	1.6716	0.6957
0.1185	141.01	8520	1.8543	0.6609
0.4103	142.01	8580	1.9628	0.6348
0.1577	143.01	8640	1.7975	0.6609
0.2213	144.01	8700	1.6324	0.6870
0.6129	145.01	8760	1.5654	0.7130
0.54	146.01	8820	1.4210	0.7565
0.357	147.01	8880	1.4255	0.7478
0.2451	148.01	8940	1.6774	0.6957
0.4752	149.01	9000	1.7326	0.6957
0.1847	150.01	9060	1.7124	0.6609
0.2618	151.01	9120	1.6317	0.6783
0.4884	152.01	9180	1.6136	0.6870
0.4929	153.01	9240	1.5062	0.7217
0.5781	154.01	9300	1.4666	0.7217
0.4633	155.01	9360	1.5033	0.7043
0.5355	156.01	9420	1.4821	0.6957
0.551	157.01	9480	1.4866	0.6957
0.3247	158.01	9540	1.5070	0.6957
0.5455	159.01	9600	1.5087	0.6957

Framework versions

• Transformers 4.29.1
• Pytorch 2.0.1+cu117
• Datasets 2.12.0
• Tokenizers 0.13.3