CTMAE2_CS_V7_1

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: 0.4206
• Accuracy: 0.9130

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: 38850

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0024	0.02	777	2.8391	0.4565
1.8322	1.02	1554	2.3824	0.4565
0.8493	2.02	2331	1.1138	0.4565
0.4768	3.02	3108	1.7405	0.5
2.0611	4.02	3885	1.0070	0.7826
0.4943	5.02	4662	3.2575	0.4565
0.708	6.02	5439	2.1191	0.4565
0.5518	7.02	6216	1.4170	0.7609
1.2084	8.02	6993	0.6194	0.8478
1.4958	9.02	7770	0.6078	0.8478
1.7393	10.02	8547	0.7951	0.7609
1.0987	11.02	9324	0.7860	0.8261
0.8703	12.02	10101	0.9879	0.7826
1.0987	13.02	10878	1.0684	0.7826
0.0074	14.02	11655	1.2901	0.7391
1.39	15.02	12432	1.6628	0.6304
0.2423	16.02	13209	0.4206	0.9130
1.4765	17.02	13986	0.5211	0.8478
0.4732	18.02	14763	0.8912	0.8261
0.8663	19.02	15540	0.4561	0.8696
1.0169	20.02	16317	0.6229	0.8696
0.0059	21.02	17094	0.7054	0.8478
0.8826	22.02	17871	1.0915	0.7391
0.4632	23.02	18648	0.6940	0.8913
0.4901	24.02	19425	1.0199	0.8261
0.3927	25.02	20202	0.8669	0.8261
0.3564	26.02	20979	0.7220	0.8696
0.0017	27.02	21756	1.2441	0.7826
1.1658	28.02	22533	0.6606	0.8696
0.6756	29.02	23310	1.6571	0.7174
0.0165	30.02	24087	0.7047	0.8696
1.7687	31.02	24864	0.6157	0.8696
0.0036	32.02	25641	1.0989	0.8478
0.6257	33.02	26418	0.9701	0.8261
0.8758	34.02	27195	0.7716	0.8696
0.5316	35.02	27972	0.9176	0.8478
0.0003	36.02	28749	1.0744	0.8478
0.0005	37.02	29526	0.9155	0.8696
0.2137	38.02	30303	0.5414	0.9130
0.0001	39.02	31080	0.9475	0.8261
0.0022	40.02	31857	1.0578	0.8478
0.0002	41.02	32634	0.9568	0.8696
0.0004	42.02	33411	0.9224	0.8696
0.0	43.02	34188	0.8144	0.8913
0.0001	44.02	34965	0.9011	0.8913
0.6661	45.02	35742	1.0514	0.8478
0.0001	46.02	36519	0.8273	0.8913
0.0	47.02	37296	0.7097	0.8913
0.6658	48.02	38073	0.8315	0.8913
0.0	49.02	38850	0.8576	0.8913

Framework versions

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