9_mae_5

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: 0.5883
• Accuracy: 0.8444

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
1.0343	0.02	777	2.7909	0.4667
1.862	1.02	1554	2.2958	0.4667
0.4912	2.02	2331	0.6226	0.5778
1.3007	3.02	3108	2.4763	0.4667
0.9231	4.02	3885	1.5136	0.6667
0.5727	5.02	4662	3.2810	0.4667
0.7861	6.02	5439	1.9614	0.4667
1.3624	7.02	6216	1.9749	0.5556
0.7581	8.02	6993	0.5883	0.8444
1.5744	9.02	7770	1.6895	0.6222
0.8279	10.02	8547	1.9930	0.4889
1.7719	11.02	9324	0.9402	0.7778
0.7774	12.02	10101	1.7397	0.7111
0.6499	13.02	10878	0.8965	0.7556
1.055	14.02	11655	2.5025	0.5333
0.4922	15.02	12432	1.3579	0.7111
0.5173	16.02	13209	1.3173	0.8
1.1031	17.02	13986	1.3298	0.7778
0.4406	18.02	14763	0.7908	0.8222
0.7463	19.02	15540	1.5625	0.7111
2.1131	20.02	16317	1.7190	0.6444
1.528	21.02	17094	1.6978	0.7111
1.1473	22.02	17871	1.7044	0.7333
0.5417	23.02	18648	1.2448	0.8
0.2006	24.02	19425	1.1313	0.8222
0.0071	25.02	20202	2.2483	0.6667
0.0016	26.02	20979	1.7836	0.7556
0.0004	27.02	21756	1.9108	0.7556
0.0014	28.02	22533	1.9915	0.7333
0.3871	29.02	23310	1.7853	0.7556
0.4299	30.02	24087	2.1849	0.7111
0.0044	31.02	24864	1.6871	0.7556
0.001	32.02	25641	1.5571	0.7778
1.1781	33.02	26418	1.8120	0.7333
0.5962	34.02	27195	2.3191	0.6222
0.0005	35.02	27972	2.4253	0.6889
0.746	36.02	28749	1.3087	0.8222
0.0001	37.02	29526	1.2850	0.8222
0.7738	38.02	30303	1.6447	0.7778
1.1191	39.02	31080	2.3239	0.7333
0.0001	40.02	31857	1.7029	0.8
0.7854	41.02	32634	1.6677	0.8
0.0	42.02	33411	2.2465	0.7333
0.0	43.02	34188	1.8314	0.7778
1.3186	44.02	34965	1.7976	0.7778
0.0	45.02	35742	1.8324	0.8
0.6813	46.02	36519	1.8297	0.7778
0.0	47.02	37296	1.7793	0.7778
0.0	48.02	38073	2.1542	0.7556
0.0	49.02	38850	2.1519	0.7556

Framework versions

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