CTMAE-P2-V3-3G-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: 0.9771
• Accuracy: 0.8

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: 4
• eval_batch_size: 4
• 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: 3250

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6262	0.0203	66	0.8422	0.4
0.5961	1.0203	132	0.9177	0.4
0.6691	2.0203	198	0.8617	0.4
0.6242	3.0203	264	0.9506	0.4
0.631	4.0203	330	0.7678	0.4
0.6747	5.0203	396	0.6632	0.6
0.6345	6.0203	462	0.7091	0.5556
0.6357	7.0203	528	1.0969	0.5111
0.4717	8.0203	594	0.6719	0.6444
0.827	9.0203	660	0.5489	0.7778
0.3107	10.0203	726	0.7150	0.6889
0.3366	11.0203	792	0.7248	0.7111
0.8919	12.0203	858	0.6667	0.7778
0.4823	13.0203	924	2.1050	0.4222
0.3742	14.0203	990	1.0017	0.5778
0.3399	15.0203	1056	1.5679	0.5556
0.6571	16.0203	1122	1.3521	0.6
0.2434	17.0203	1188	0.7812	0.7778
0.5967	18.0203	1254	0.9575	0.6889
0.1982	19.0203	1320	1.1721	0.6667
0.2631	20.0203	1386	2.5733	0.4667
0.3235	21.0203	1452	0.9771	0.8
0.1786	22.0203	1518	1.1978	0.7111
0.1352	23.0203	1584	0.8692	0.7778
0.1709	24.0203	1650	1.1424	0.7556
0.0873	25.0203	1716	1.8760	0.6222
0.1418	26.0203	1782	1.0964	0.8
0.0075	27.0203	1848	1.9130	0.6222
0.4534	28.0203	1914	1.1176	0.7778
0.0019	29.0203	1980	2.0684	0.6222
0.4098	30.0203	2046	1.9198	0.6667
0.0006	31.0203	2112	1.2724	0.7333
0.1891	32.0203	2178	1.8213	0.6444
0.0604	33.0203	2244	2.6845	0.5556
0.104	34.0203	2310	2.7468	0.6222
0.028	35.0203	2376	1.4458	0.7333
0.2479	36.0203	2442	2.1613	0.6222
0.0967	37.0203	2508	1.3895	0.7778
0.0668	38.0203	2574	2.0147	0.6667
0.0004	39.0203	2640	1.5766	0.6889
0.0027	40.0203	2706	2.3533	0.6444
0.1436	41.0203	2772	2.1496	0.6444
0.0327	42.0203	2838	2.2866	0.6444
0.0349	43.0203	2904	2.2496	0.6667
0.1178	44.0203	2970	1.8929	0.6889
0.0001	45.0203	3036	1.9030	0.7111
0.1289	46.0203	3102	1.9212	0.6889
0.137	47.0203	3168	1.6077	0.7556
0.0004	48.0203	3234	1.7398	0.7333
0.0001	49.0049	3250	1.7423	0.7333

Framework versions

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