CTMAE-P2-V5-3g-S2

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: 1.1479
• Accuracy: 0.8222

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: 1
• eval_batch_size: 1
• 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: 13050

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.3342	0.02	261	1.3344	0.4667
0.5407	1.02	522	3.0926	0.4667
1.9217	2.02	783	2.3252	0.4667
0.6849	3.02	1044	2.6546	0.4667
1.7893	4.02	1305	1.1093	0.4667
0.8412	5.02	1566	1.2741	0.4667
1.4469	6.02	1827	2.1842	0.4667
1.2686	7.02	2088	1.8709	0.4667
1.4495	8.02	2349	1.8210	0.4667
0.7905	9.02	2610	0.6076	0.7556
0.8322	10.02	2871	0.9286	0.7111
0.8618	11.02	3132	0.9191	0.7111
1.1268	12.02	3393	1.7668	0.5778
0.7087	13.02	3654	1.8699	0.5333
1.5327	14.02	3915	1.4736	0.5556
0.531	15.02	4176	1.5145	0.6667
1.0064	16.02	4437	1.4191	0.6222
0.2797	17.02	4698	1.3301	0.7333
1.0355	18.02	4959	1.4829	0.7111
0.937	19.02	5220	1.8992	0.5556
0.0065	20.02	5481	1.1564	0.6444
0.9542	21.02	5742	1.4034	0.6889
0.4833	22.02	6003	1.3208	0.7111
1.1719	23.02	6264	1.2117	0.7333
0.9043	24.02	6525	1.0468	0.8
1.1914	25.02	6786	1.3830	0.7111
0.0032	26.02	7047	1.8348	0.7111
0.0022	27.02	7308	1.9530	0.6889
1.1481	28.02	7569	1.4339	0.7111
0.8332	29.02	7830	2.0841	0.6667
0.9692	30.02	8091	1.4363	0.7333
0.0095	31.02	8352	1.6087	0.7778
0.0007	32.02	8613	1.1479	0.8222
0.534	33.02	8874	1.7736	0.6667
1.149	34.02	9135	1.3923	0.8
0.0015	35.02	9396	1.7895	0.6889
0.0007	36.02	9657	1.6939	0.7556
0.6659	37.02	9918	1.9130	0.6889
0.2733	38.02	10179	1.8643	0.6667
0.3147	39.02	10440	1.8536	0.6889
0.0003	40.02	10701	1.6605	0.7778
0.4885	41.02	10962	1.7213	0.7333
0.0086	42.02	11223	1.9743	0.7111
0.0004	43.02	11484	1.5626	0.7556
0.0005	44.02	11745	1.5694	0.7333
0.0003	45.02	12006	1.6538	0.7778
0.0002	46.02	12267	1.8243	0.7111
0.0001	47.02	12528	1.7648	0.7556
0.0002	48.02	12789	1.6705	0.7333
0.0008	49.02	13050	1.6519	0.7556

Framework versions

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