Even
@WongKinYiu Actually, my problem is i did not figure out a way to implement a Group Softmax layer(an output layer) in PyTorch. ``` torch.softmax ``` This API can not do...
@dontLoveBugs It seems that 0.36m is the correct metric baseline.
@french-paragon @LifeBeyondExpectations @RollingIsland 0.36m
@sibozhang Hi, thanks for replying. I want to train ApolloScape stereo dataset with MonoDepthV2, and the metric baseline value is needed(In Kitti dataset, the value is 0.54meters).
The baseline metric distance of two stereo cameras(The metric distance between the left and right camera).
@Rachine Hi, I have launched a PR for Python3 support, you can check it later and determine whether to merge or not.
@gaoyangyiqiao Hi, i 'm also confused about this point, did u got it yet?
@nana1205 refer to https://github.com/CaptainEven/RepNet-MDNet-VehicleReID
@luckxiao 不需要提取码，百度云盘分享见readme
应该可以。在目标检测网络的基础上做多分支的多任务学习。