search

cwq159 / PyTorch-Spiking-YOLOv3

A PyTorch implementation of Spiking-YOLOv3. Two branches are provided, based on two common PyTorch implementation of YOLOv3(ultralytics/yolov3 & eriklindernoren/PyTorch-YOLOv3), with support for Spiking-YOLOv3-Tiny at present.
GNU General Public License v3.0
208 stars 58 forks source link

Training on VOC dataset #12

Open polarcrow opened 3 years ago

polarcrow commented 3 years ago

Hello,

I want to train the Yolov3-tiny-ours model on the VOC dataset but didn't find the code and dataset download script. How did you train the model on the VOC dataset ?

Also, do you plan to integrate channel-wise parameters normalization and leaky ReLu spiking activations like exposed in the Spiking-YOLO paper as I didn't find references to them in your code ?

Thanks in advance.

cwq159 commented 3 years ago

Please download the VOC dataset by yourself. This code has integrated channel-wise parameters normalization. You can use it with - -channel-wise. This code hasn’t included leaky ReLU spiking activations because we only use positive spike(+1) not negative spike(-1). Leaky ReLU uses negative spike(-1) which we will code for it in the future.