videomae-base-finetuned-subset-200epochs

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: 0.7635
• Accuracy: 0.7407

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: 8
• eval_batch_size: 8
• 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: 11100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6058	0.01	56	0.7442	0.7880
0.4908	1.01	112	0.7775	0.7558
0.5326	2.01	168	0.7973	0.7419
0.4768	3.01	224	0.8451	0.7281
0.4243	4.01	280	0.9361	0.6728
0.6921	5.01	336	0.8979	0.7097
0.3182	6.01	392	0.8852	0.7235
0.6085	7.01	448	0.9224	0.7097
0.4067	8.01	504	0.9631	0.6682
0.47	9.01	560	0.9193	0.7465
0.5058	10.01	616	0.8967	0.7650
0.4187	11.01	672	0.7403	0.7834
0.6033	12.01	728	1.0005	0.6221
0.5032	13.01	784	1.1420	0.5899
0.5967	14.01	840	1.2590	0.5484
0.3103	15.01	896	0.9723	0.6544
0.4201	16.01	952	1.1665	0.6406
0.6246	17.01	1008	1.2497	0.4977
0.6306	18.01	1064	1.3829	0.5668
0.4179	19.01	1120	1.0787	0.5806
0.5468	20.01	1176	1.1144	0.5714
0.4166	21.01	1232	0.7674	0.6912
0.3844	22.01	1288	0.9260	0.6959
0.5138	23.01	1344	0.9093	0.7097
0.792	24.01	1400	0.7327	0.7465
0.5944	25.01	1456	0.8933	0.7650
0.4855	26.01	1512	0.9830	0.6636
0.6896	27.01	1568	0.7896	0.6590
0.3617	28.01	1624	0.8900	0.6544
0.6362	29.01	1680	1.0237	0.6912
0.6475	30.01	1736	1.1399	0.6037
0.5088	31.01	1792	0.7190	0.7742
0.7271	32.01	1848	0.9492	0.6359
0.3171	33.01	1904	0.9431	0.7281
0.5847	34.01	1960	0.7997	0.7235
0.4703	35.01	2016	0.9506	0.7051
0.4995	36.01	2072	1.0830	0.7005
0.5682	37.01	2128	1.0100	0.7005
0.6424	38.01	2184	0.9587	0.6452
0.5897	39.01	2240	0.8807	0.7097
0.5222	40.01	2296	1.1219	0.6682
0.5239	41.01	2352	1.0848	0.6406
0.5957	42.01	2408	0.9640	0.6866
0.5279	43.01	2464	1.0291	0.5853
0.3545	44.01	2520	0.8908	0.6636
0.6066	45.01	2576	1.2505	0.6406
0.3658	46.01	2632	0.8362	0.6866
0.5454	47.01	2688	1.3975	0.5622
0.5956	48.01	2744	0.8236	0.6590
0.4107	49.01	2800	1.2610	0.6267
0.462	50.01	2856	1.2553	0.6406
0.4837	51.01	2912	1.0389	0.6359
0.621	52.01	2968	0.8281	0.7235
0.4293	53.01	3024	1.0426	0.6267
0.4255	54.01	3080	1.2942	0.5806
0.5607	55.01	3136	1.1234	0.6498
0.3104	56.01	3192	1.0643	0.6590
0.3335	57.01	3248	1.2160	0.6590
0.4232	58.01	3304	1.3532	0.5806
0.6238	59.01	3360	0.9208	0.7005
0.369	60.01	3416	1.2186	0.5530
0.3874	61.01	3472	1.1746	0.6452
0.3421	62.01	3528	1.2017	0.5945
0.4243	63.01	3584	1.0288	0.6728
0.2806	64.01	3640	0.8483	0.7419
0.5357	65.01	3696	1.0890	0.6359
0.5155	66.01	3752	1.1885	0.6359
0.4367	67.01	3808	1.0738	0.6820
0.48	68.01	3864	1.0894	0.6866
0.4703	69.01	3920	1.2252	0.6498
0.4531	70.01	3976	1.0584	0.6498
0.2898	71.01	4032	1.7486	0.5576
0.3684	72.01	4088	1.0524	0.6406
0.2752	73.01	4144	1.2744	0.6728
0.3092	74.01	4200	1.3918	0.5806
0.3507	75.01	4256	1.4599	0.6544
0.4722	76.01	4312	1.0549	0.7143
0.4059	77.01	4368	1.2727	0.6728
0.2734	78.01	4424	1.1258	0.6959
0.4168	79.01	4480	0.9788	0.7189
0.4456	80.01	4536	1.4757	0.6544
0.4519	81.01	4592	1.2796	0.6820
0.5283	82.01	4648	1.2542	0.7051
0.4738	83.01	4704	1.2781	0.6083
0.2128	84.01	4760	1.0077	0.6866
0.3262	85.01	4816	1.0287	0.6820
0.3631	86.01	4872	1.3574	0.6544
0.4085	87.01	4928	1.1976	0.7235
0.3582	88.01	4984	1.4126	0.6544
0.3564	89.01	5040	1.3488	0.6406
0.4207	90.01	5096	1.0565	0.7005
0.4307	91.01	5152	0.9833	0.7281
0.3863	92.01	5208	0.9340	0.6912
0.2949	93.01	5264	0.9835	0.7143
0.2957	94.01	5320	1.1397	0.7235
0.3767	95.01	5376	1.4135	0.6221
0.4949	96.01	5432	1.0483	0.7189
0.3058	97.01	5488	1.8241	0.5530
0.3406	98.01	5544	1.7386	0.5760
0.2319	99.01	5600	1.4739	0.6175
0.5261	100.01	5656	1.0822	0.7143
0.4181	101.01	5712	1.2876	0.6728
0.243	102.01	5768	1.0783	0.7235
0.2603	103.01	5824	1.4557	0.6129
0.4892	104.01	5880	1.2557	0.6912
0.3073	105.01	5936	1.3899	0.5991
0.3601	106.01	5992	1.2048	0.6820
0.4371	107.01	6048	1.3645	0.6866
0.5712	108.01	6104	1.2281	0.6636
0.3697	109.01	6160	1.4402	0.6544
0.2978	110.01	6216	1.3769	0.6912
0.303	111.01	6272	1.3096	0.6959
0.4606	112.01	6328	1.2236	0.7005
0.2554	113.01	6384	1.2662	0.6959
0.3033	114.01	6440	1.2476	0.6406
0.3025	115.01	6496	1.0474	0.7143
0.3513	116.01	6552	1.4692	0.6452
0.4205	117.01	6608	1.2675	0.6912
0.3898	118.01	6664	1.4018	0.6590
0.2184	119.01	6720	1.2402	0.6959
0.319	120.01	6776	1.0747	0.7097
0.2455	121.01	6832	1.3515	0.7051
0.2138	122.01	6888	1.5175	0.6682
0.3805	123.01	6944	1.4817	0.6820
0.3942	124.01	7000	1.5235	0.6221
0.2207	125.01	7056	1.6295	0.5945
0.2217	126.01	7112	1.3348	0.6912
0.3173	127.01	7168	1.3566	0.7097
0.4952	128.01	7224	1.2188	0.7327
0.3238	129.01	7280	1.2574	0.7143
0.1525	130.01	7336	1.5508	0.6313
0.2518	131.01	7392	1.3058	0.6912
0.4523	132.01	7448	1.7539	0.6313
0.3732	133.01	7504	1.4478	0.6820
0.2432	134.01	7560	1.3595	0.6912
0.2798	135.01	7616	1.5007	0.6866
0.3436	136.01	7672	1.3162	0.7465
0.3033	137.01	7728	1.3700	0.7051
0.3457	138.01	7784	1.1052	0.7465
0.1381	139.01	7840	1.5786	0.6959
0.3067	140.01	7896	1.5155	0.6912
0.269	141.01	7952	1.2751	0.7512
0.2646	142.01	8008	1.6017	0.6774
0.3933	143.01	8064	1.4294	0.7005
0.6315	144.01	8120	1.3814	0.6866
0.2814	145.01	8176	1.1689	0.7512
0.2749	146.01	8232	1.3208	0.7005
0.3966	147.01	8288	1.2817	0.7189
0.1787	148.01	8344	1.4568	0.7189
0.3006	149.01	8400	1.3312	0.7143
0.2871	150.01	8456	1.5808	0.6452
0.2018	151.01	8512	1.6682	0.6267
0.2698	152.01	8568	1.4281	0.6590
0.162	153.01	8624	1.4369	0.7051
0.3961	154.01	8680	1.3771	0.7143
0.4034	155.01	8736	1.5444	0.6452
0.2462	156.01	8792	1.4677	0.6728
0.2564	157.01	8848	1.6085	0.6590
0.2905	158.01	8904	1.3037	0.6912
0.2762	159.01	8960	1.3974	0.7051
0.1604	160.01	9016	1.5176	0.6959
0.2399	161.01	9072	1.4504	0.7143
0.3398	162.01	9128	1.4675	0.6728
0.2495	163.01	9184	1.3757	0.7005
0.3076	164.01	9240	1.3699	0.7051
0.2491	165.01	9296	1.4333	0.7005
0.1666	166.01	9352	1.6465	0.6313
0.1871	167.01	9408	1.6614	0.6544
0.2169	168.01	9464	1.8141	0.6175
0.3918	169.01	9520	1.3402	0.7097
0.2697	170.01	9576	1.4295	0.6774
0.2261	171.01	9632	1.5952	0.6452
0.1894	172.01	9688	1.5468	0.6590
0.1714	173.01	9744	1.4434	0.6636
0.3137	174.01	9800	1.5525	0.6313
0.267	175.01	9856	1.6447	0.6452
0.0797	176.01	9912	1.5593	0.6682
0.2698	177.01	9968	1.3952	0.7005
0.1364	178.01	10024	1.6720	0.6498
0.2342	179.01	10080	1.6315	0.6682
0.1909	180.01	10136	1.5374	0.7051
0.2234	181.01	10192	1.5861	0.7097
0.3425	182.01	10248	1.5664	0.6912
0.4092	183.01	10304	1.6135	0.6774
0.2427	184.01	10360	1.5366	0.6866
0.3751	185.01	10416	1.5561	0.6959
0.1831	186.01	10472	1.6049	0.7005
0.2207	187.01	10528	1.6072	0.6959
0.1096	188.01	10584	1.5016	0.7097
0.2417	189.01	10640	1.5027	0.7097
0.2974	190.01	10696	1.4897	0.7097
0.2296	191.01	10752	1.4927	0.7235
0.3323	192.01	10808	1.4947	0.7235
0.3002	193.01	10864	1.5225	0.7143
0.23	194.01	10920	1.4965	0.7189
0.3147	195.01	10976	1.5123	0.7051
0.1344	196.01	11032	1.5192	0.7051
0.1843	197.01	11088	1.5235	0.7097
0.1902	198.0	11100	1.5238	0.7097

Framework versions

• Transformers 4.36.2
• Pytorch 1.13.1
• Datasets 2.16.1
• Tokenizers 0.15.0