Efficient Violence Detection Using 3D Convolutional Neural Networks
MIT License
15
stars
10
forks
source link
Question about the experimental setup with VIolent-Flows Dataset in the paper "Efficient Violence Detection Using 3D Convolutional Neural Networks" #13
First of all, I would like to thank you for your contribution to the community with your great work.
I'm really interested in your study and I am trying to implement your code with the same experimental setup with VIolent-Flows Dataset like in your paper. However there are something ambiguous for me here :
+, The learning rate and Batchsize in the paper in VIolent-Flows Dataset is 10^-4 and 16. I followed the Experiment Setup of paper but the result was only "94.35±2.28% " compared to the result of paper is : "97.17±0.95%". If I use learning rate = 10^-3 and Batchsize = 32 the result will be :" 96,768 ± 2.38%", is approximate to the result of paper. So I wonder if the Experiment Setup of paper is correct? Can you help me check it again?
Thank you!
First of all, I would like to thank you for your contribution to the community with your great work. I'm really interested in your study and I am trying to implement your code with the same experimental setup with VIolent-Flows Dataset like in your paper. However there are something ambiguous for me here : +, The learning rate and Batchsize in the paper in VIolent-Flows Dataset is 10^-4 and 16. I followed the Experiment Setup of paper but the result was only "94.35±2.28% " compared to the result of paper is : "97.17±0.95%". If I use learning rate = 10^-3 and Batchsize = 32 the result will be :" 96,768 ± 2.38%", is approximate to the result of paper. So I wonder if the Experiment Setup of paper is correct? Can you help me check it again? Thank you!