ChiWeiHsiao / DeepVO-pytorch

PyTorch Implementation of DeepVO

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Train loss and test loss

zhaowenZhou opened this issue · comments

During training the trian loss is very small like around 0.3, but the smallest test loss is aroung 90.
Then I used trained model provided by alexart13 on test this model on training dataset ['01','02', '05', '08', '09']. The test.py still gave me three-digit even four-digit loss.
Anyone know why?

can you help me debug my code, when i run it my training loss comes around 28 for some reason , I downloaded the same repo and set the paths and everything but my training loss comes out to around 28, my img_mean is also different for some reason when I use the preprocessing.py to calculate the mean and std my values are around 80 and not between 0 and 1 although I am using the image_3 data set from KITTI like described in the repo. can someone guide in this regard ?