MCG-NJUvideomae-base-finetuned-kinetics-face

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.7299
• Accuracy: 0.9167

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: 1
• eval_batch_size: 1
• 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: 6000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.5489	3.0017	100	1.5109	0.6111
0.6115	6.0033	200	0.5476	1.0
0.042	9.005	300	0.0697	1.0
0.0243	13.0017	400	0.1967	0.9167
1.1668	16.0033	500	0.9849	0.8333
0.1354	19.005	600	0.7017	0.8333
0.8714	23.0017	700	0.0027	1.0
0.0009	26.0033	800	0.2290	0.9444
0.0003	29.005	900	0.1073	0.9722
1.0037	33.0017	1000	1.3122	0.75
0.4275	36.0033	1100	0.0041	1.0
0.0005	39.005	1200	0.3496	0.9444
0.0005	43.0017	1300	0.1043	0.9722
0.0239	46.0033	1400	1.4939	0.8333
0.0174	49.005	1500	0.0428	0.9722
0.001	53.0017	1600	0.6678	0.8889
0.0865	56.0033	1700	0.3907	0.9444
0.9919	59.005	1800	0.0019	1.0
0.0007	63.0017	1900	0.7224	0.8889
0.0002	66.0033	2000	0.0023	1.0
0.0002	69.005	2100	0.2816	0.9444
0.7678	73.0017	2200	1.0904	0.8056
0.0013	76.0033	2300	1.0187	0.8611
0.0004	79.005	2400	0.2458	0.9722
0.0094	83.0017	2500	0.9661	0.8333
0.0003	86.0033	2600	0.2274	0.9722
0.0011	89.005	2700	0.2844	0.9444
0.0001	93.0017	2800	0.8405	0.8611
0.0001	96.0033	2900	0.5875	0.9167
0.0001	99.005	3000	1.8345	0.7778
0.0001	103.0017	3100	0.5098	0.9167
0.0003	106.0033	3200	0.0062	1.0
0.3248	109.005	3300	0.4113	0.9444
0.0001	113.0017	3400	0.1064	0.9722
0.0001	116.0033	3500	0.0006	1.0
0.0003	119.005	3600	0.2552	0.9722
0.001	123.0017	3700	0.0202	1.0
0.0002	126.0033	3800	0.3475	0.9444
0.0001	129.005	3900	0.5493	0.9444
0.0001	133.0017	4000	0.5506	0.9444
0.0001	136.0033	4100	0.5711	0.9167
0.0001	139.005	4200	0.5181	0.9444
0.0021	143.0017	4300	0.7568	0.9167
0.9007	146.0033	4400	0.0072	1.0
0.0001	149.005	4500	0.2858	0.9444
0.0001	153.0017	4600	1.0247	0.8889
0.6131	156.0033	4700	0.0814	0.9722
0.0004	159.005	4800	1.8986	0.8056
0.0001	163.0017	4900	1.5607	0.8056
0.0001	166.0033	5000	1.5370	0.8056
0.0001	169.005	5100	1.4807	0.8056
0.0001	173.0017	5200	1.2996	0.8333
0.0	176.0033	5300	1.2259	0.8056
0.0001	179.005	5400	1.1819	0.8056
0.0	183.0017	5500	1.1047	0.8056
0.0001	186.0033	5600	1.0461	0.8333
0.0	189.005	5700	1.2544	0.8056
0.8628	193.0017	5800	1.1260	0.8056
0.0	196.0033	5900	0.7299	0.9167
0.0	199.005	6000	0.7299	0.9167

Framework versions

• Transformers 4.45.2
• Pytorch 2.4.1+cu118
• Datasets 3.0.1
• Tokenizers 0.20.1