omermazig/2-class-2-batch-50-epochs

videomae-base-finetuned-kinetics-finetuned-nba-binary-data-2-batch-50-epochs-new-database

This model is a fine-tuned version of MCG-NJU/videomae-base-finetuned-kinetics on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1744
• Accuracy: 0.965

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: 5e-05
• train_batch_size: 2
• eval_batch_size: 2
• 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: 10000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6618	0.02	200	0.6293	0.6875
0.5781	1.02	400	1.4660	0.6042
0.8554	2.02	600	0.8740	0.6667
0.4445	3.02	800	1.0660	0.6667
0.3265	4.02	1000	0.6635	0.7708
0.5417	5.02	1200	0.4705	0.8542
0.5912	6.02	1400	1.0082	0.7708
0.5918	7.02	1600	2.6292	0.5625
0.8992	8.02	1800	0.8514	0.7708
0.172	9.02	2000	0.4568	0.875
0.493	10.02	2200	0.7354	0.7917
0.3622	11.02	2400	1.0386	0.7708
0.4966	12.02	2600	0.8979	0.7917
0.3541	13.02	2800	0.8220	0.7708
0.5386	14.02	3000	1.0256	0.7708
0.4615	15.02	3200	1.0447	0.7917
0.1624	16.02	3400	0.6448	0.8542
1.0388	17.02	3600	0.9992	0.7708
0.0442	18.02	3800	1.1626	0.7708
0.2449	19.02	4000	0.8174	0.8542
0.3024	20.02	4200	0.8500	0.7917
0.4879	21.02	4400	1.2219	0.7292
0.4035	22.02	4600	0.6436	0.8333
0.0334	23.02	4800	0.7433	0.8333
0.4849	24.02	5000	0.9911	0.8125
0.6075	25.02	5200	1.2249	0.7083
0.3441	26.02	5400	0.8563	0.8333
0.5653	27.02	5600	0.4557	0.8958
0.196	28.02	5800	0.4156	0.8542
0.0038	29.02	6000	0.4562	0.8542
0.2696	30.02	6200	0.8153	0.7917
0.0015	31.02	6400	0.5923	0.8958
0.0036	32.02	6600	0.7343	0.875
0.3623	33.02	6800	0.3089	0.9375
0.2142	34.02	7000	0.6142	0.8958
0.0008	35.02	7200	0.6010	0.875
0.0005	36.02	7400	0.6238	0.875
0.0002	37.02	7600	0.5966	0.875
0.5	38.02	7800	0.6371	0.8542
0.0004	39.02	8000	0.8515	0.8542
0.0001	40.02	8200	0.5120	0.875
0.0069	41.02	8400	0.8686	0.8542
0.0002	42.02	8600	0.8801	0.8542
0.0001	43.02	8800	0.8996	0.8542
0.0067	44.02	9000	0.7670	0.8542
0.0001	45.02	9200	0.9936	0.8333
0.0638	46.02	9400	0.6616	0.875
0.0001	47.02	9600	0.7978	0.8542
0.0001	48.02	9800	0.6737	0.8542
0.0001	49.02	10000	0.5887	0.875

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3