MAE-CT-CPC-Dicotomized-v3

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: 0.8472
• Accuracy: 0.7317

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• training_steps: 4000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5652	0.02	81	0.6260	0.6829
0.6109	1.02	162	0.6052	0.6829
0.7142	2.02	243	0.6207	0.6829
0.5037	3.02	324	0.6219	0.6829
0.5177	4.02	405	0.6456	0.6829
0.5363	5.02	486	0.5791	0.6829
0.6797	6.02	567	0.6295	0.6829
0.557	7.02	648	0.5182	0.7073
0.5645	8.02	729	0.4608	0.7073
0.4484	9.02	810	0.4721	0.7805
0.3354	10.02	891	0.7011	0.7317
0.2875	11.02	972	0.4498	0.7805
0.3535	12.02	1053	0.8161	0.7561
0.294	13.02	1134	0.5175	0.7805
0.1907	14.02	1215	0.3750	0.8537
0.2533	15.02	1296	0.6400	0.7805
0.2499	16.02	1377	0.9233	0.7805
0.2726	17.02	1458	1.0686	0.7805
0.4447	18.02	1539	1.0433	0.7073
0.1554	19.02	1620	1.0791	0.8293
0.0163	20.02	1701	1.0939	0.7317
0.0507	21.02	1782	1.1601	0.8049
0.0116	22.02	1863	1.3358	0.7073
0.015	23.02	1944	1.0973	0.8049
0.1993	24.02	2025	1.2985	0.7805
0.1496	25.02	2106	1.1559	0.7805
0.2059	26.02	2187	1.5124	0.7561
0.0176	27.02	2268	1.4691	0.7805
0.2477	28.02	2349	1.5908	0.7805
0.0007	29.02	2430	1.3929	0.7805
0.0003	30.02	2511	1.3020	0.7805
0.0002	31.02	2592	1.5539	0.7805
0.0001	32.02	2673	1.5173	0.8049
0.0001	33.02	2754	1.6113	0.8049
0.0918	34.02	2835	1.5837	0.7805
0.0001	35.02	2916	1.4928	0.8049
0.0001	36.02	2997	1.4138	0.8049
0.2197	37.02	3078	1.3980	0.8049
0.0001	38.02	3159	1.6004	0.8049
0.0002	39.02	3240	1.5590	0.8049
0.0246	40.02	3321	1.5889	0.7805
0.0002	41.02	3402	1.6164	0.7805
0.0001	42.02	3483	1.6072	0.7805
0.0001	43.02	3564	1.6259	0.7805
0.0001	44.02	3645	1.5990	0.7805
0.0002	45.02	3726	1.5517	0.7805
0.0	46.02	3807	1.5260	0.7805
0.0146	47.02	3888	1.5292	0.7805
0.0001	48.02	3969	1.5405	0.7805
0.0001	49.01	4000	1.5414	0.7805

Framework versions

• Transformers 4.37.2
• Pytorch 2.0.1+cu117
• Datasets 3.0.1
• Tokenizers 0.15.1