liyuanyaun

when i train voca, i found that volatile gpu-util is only 0%~6% which is so low? why this happend? @TimoBolkart

i try to make batchsize>1 to inference, but failed. only change run_Ringnet.py "self.batch_size" not help? could you give me any solution? @soubhiksanyal @TimoBolkart

the environment of runing your code and the dataset

2

may you offer the environment to run your code? such as the version of tensorflow. and may you tell when you will release the dataset? thank you very much! @ZhaoJ9014

hi, i follow your Google Drive link to download the pre-trained model, but when i decompress the file "checkpoint_step309.tar", the error is this file is not a tar file? could...

could you please share a demo for fpn, and describe how you built fpn? thank you very much @soeaver

how long would you release the training code?

2

hi, could you tell me how long will you release the training code? especially the preprocess code of data? @lelechen63

training time is too long for data process

1

i use ssd_pascal.py for training, and my training log is as following: I1207 14:33:58.859539 17469 solver.cpp:259] Train net output #0: mbox_loss = 12.2819 (* 1 = 12.2819 loss) I1207 14:34:00.304049...

did any one know how to use python to render reconstructed mesh with generated texture? @yanght321