CTMAE-P2-V3-3G-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: 0.5835
• Accuracy: 0.8696

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: 2
• eval_batch_size: 2
• 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: 6500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6279	0.02	130	0.7452	0.4783
0.6448	1.02	260	0.9604	0.4783
0.675	2.02	390	0.7935	0.4783
0.5523	3.02	520	0.8305	0.4783
0.9104	4.02	650	0.8216	0.4783
0.5657	5.02	780	1.2400	0.4783
1.1352	6.02	910	0.6858	0.4783
0.77	7.02	1040	0.9103	0.5217
1.4024	8.02	1170	0.9320	0.6522
0.7694	9.02	1300	1.1192	0.5652
0.663	10.02	1430	0.8375	0.6739
1.0107	11.02	1560	0.9901	0.6087
0.8404	12.02	1690	0.4649	0.7826
0.7372	13.02	1820	1.2412	0.6739
1.2033	14.02	1950	1.5908	0.6304
0.7936	15.02	2080	0.8874	0.7174
1.3059	16.02	2210	0.6237	0.7826
1.0162	17.02	2340	0.6233	0.8043
0.9241	18.02	2470	1.5554	0.6304
0.9925	19.02	2600	0.6251	0.8043
0.844	20.02	2730	1.0150	0.7174
0.6418	21.02	2860	0.5920	0.8043
1.0493	22.02	2990	0.8085	0.7826
0.6551	23.02	3120	1.5049	0.6957
0.596	24.02	3250	0.7728	0.7826
0.5281	25.02	3380	0.7842	0.8043
0.4911	26.02	3510	0.6299	0.8043
0.2105	27.02	3640	0.8429	0.7826
0.6859	28.02	3770	1.0266	0.7391
0.5047	29.02	3900	1.0786	0.7609
0.8081	30.02	4030	0.6552	0.8478
0.4284	31.02	4160	0.5835	0.8696
0.3249	32.02	4290	0.9921	0.7826
0.8226	33.02	4420	0.7955	0.8261
0.0009	34.02	4550	1.0117	0.8043
0.6603	35.02	4680	1.4238	0.7609
0.144	36.02	4810	1.0399	0.7826
0.4473	37.02	4940	0.9877	0.8043
0.0012	38.02	5070	0.9295	0.8043
0.1138	39.02	5200	1.1066	0.7826
0.4031	40.02	5330	1.2339	0.7826
0.1228	41.02	5460	1.0563	0.8261
0.224	42.02	5590	1.0712	0.8043
0.0445	43.02	5720	1.2617	0.7609
0.1742	44.02	5850	1.1758	0.7826
0.5739	45.02	5980	1.4026	0.7391
0.1828	46.02	6110	1.3709	0.7609
0.0153	47.02	6240	1.2819	0.8043
0.0002	48.02	6370	1.3020	0.8043
0.0022	49.02	6500	1.2445	0.8043

Framework versions

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