CTMAE-P2-V5-3g-S1

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: 0.8195
• Accuracy: 0.8261

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
1.7838	0.02	261	1.7955	0.4565
0.5703	1.02	522	3.1441	0.4565
1.7124	2.02	783	2.1934	0.4565
0.7551	3.02	1044	2.1766	0.4565
2.0593	4.02	1305	1.4703	0.4565
0.9954	5.02	1566	1.6884	0.4565
1.4644	6.02	1827	2.2782	0.4565
1.513	7.02	2088	1.9542	0.4565
1.4666	8.02	2349	2.0516	0.4565
0.8709	9.02	2610	1.4395	0.4565
1.3748	10.02	2871	2.2200	0.4565
1.7822	11.02	3132	1.9016	0.4565
0.5944	12.02	3393	0.5156	0.7391
0.3298	13.02	3654	2.3661	0.5
0.6941	14.02	3915	3.1725	0.4565
0.1119	15.02	4176	1.0384	0.7174
0.0121	16.02	4437	1.3320	0.6739
0.3107	17.02	4698	0.9006	0.8043
0.3681	18.02	4959	0.8195	0.8261
1.0861	19.02	5220	1.6541	0.6522
0.0026	20.02	5481	1.3390	0.7391
0.7763	21.02	5742	0.9517	0.7826
0.265	22.02	6003	1.7661	0.6739
0.4977	23.02	6264	1.2860	0.7174
0.4803	24.02	6525	1.4460	0.7174
1.2687	25.02	6786	0.9046	0.7826
0.7841	26.02	7047	0.8998	0.7826
1.1971	27.02	7308	0.8336	0.8043
0.6645	28.02	7569	1.0082	0.8261
0.1029	29.02	7830	1.5363	0.7391
0.5459	30.02	8091	1.4413	0.7609
0.5047	31.02	8352	1.9479	0.6957
0.961	32.02	8613	1.3866	0.7609
0.9575	33.02	8874	1.6708	0.7174
0.2937	34.02	9135	1.8605	0.6957
0.0017	35.02	9396	1.8789	0.6522
0.5069	36.02	9657	1.3560	0.8043
0.0008	37.02	9918	1.3608	0.7826
0.4073	38.02	10179	1.9196	0.7174
0.7112	39.02	10440	1.6879	0.7391
0.0002	40.02	10701	2.1146	0.7174
0.6376	41.02	10962	1.9108	0.7391
0.0006	42.02	11223	1.7010	0.7826
0.6593	43.02	11484	2.2273	0.6957
0.0002	44.02	11745	1.7225	0.7609
0.0006	45.02	12006	2.1480	0.7174
0.4689	46.02	12267	2.0820	0.7609
0.0002	47.02	12528	2.0152	0.7609
0.664	48.02	12789	1.9991	0.7609
0.0003	49.02	13050	1.9912	0.7609

Framework versions

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