deep-visualization-toolbox
deep-visualization-toolbox copied to clipboard
the LRN layer seems to highlight neurals and leads to contain many non-zerons pixels?
the LRN layer seems to highlight neurals and make it contains many non-zerons pixels. So there will be more neurals that can recongnize some patterns, but we hope only highlight the neural that can recongnize the main object. so can you tell me your motivation to add the LRN?