videomae-base-finetuned-kinetics-0411_final_5sec_org_ab7_val_inside_train_04

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:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
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: 0.05
training_steps: 67400

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0013	0.0100	675	0.3002	0.9060
0.0086	1.0100	1350	0.3780	0.8731
0.0021	2.0100	2025	0.4782	0.8708
0.0018	3.0100	2700	0.2713	0.9162
0.0036	4.0100	3375	0.3119	0.9162
0.0002	5.0100	4050	0.3516	0.9123
0.0006	6.0100	4725	0.4322	0.9037
0.0058	7.0100	5400	0.3882	0.8857
0.0002	8.0100	6075	0.7430	0.8238
0.0007	9.0100	6750	0.5749	0.8645
0.0007	10.0100	7425	0.7784	0.8340
0.1665	11.0100	8100	0.4743	0.8841
0.0006	12.0100	8775	0.4195	0.8998
0.0089	13.0100	9450	0.3420	0.9076
0.0012	14.0100	10125	0.6617	0.8606
0.0003	15.0100	10800	0.5359	0.8896
0.0033	16.0100	11475	0.6140	0.8630
0.0006	17.0100	12150	0.5461	0.8927
0.0019	18.0100	12825	0.3898	0.9272
0.0007	19.0100	13500	0.4610	0.9060
0.0001	20.0100	14175	0.4082	0.9178
0.0001	21.0100	14850	0.4762	0.9076
0.0	22.0100	15525	0.6382	0.8951
0.0	23.0100	16200	0.9528	0.8590
0.0	24.0100	16875	0.9228	0.8598
0.0003	25.0100	17550	0.7193	0.8825
0.0	26.0100	18225	0.6632	0.8865
0.0001	27.0100	18900	0.8076	0.8747
0.0	28.0100	19575	0.6610	0.8935
0.0	29.0100	20250	0.5382	0.9092
0.0178	30.0100	20925	0.6884	0.8919
0.0001	31.0100	21600	0.5910	0.9107
0.0001	32.0100	22275	0.6303	0.9076
0.0007	33.0100	22950	0.6595	0.9052