Brian Robinson

type: feat Fully soft-code optimizer choice between Adam and SGD as well as learning rate. Remove [0] index on line 79 to prevent IndexError: invalid index of a 0-dim tensor.

First of all, thanks for taking the time to put this pytorch implementation together. I am getting the following error. Also can you explain what the purpose of the torch.cat...

OS: Raspbian Bullseye Python: 3.9 Following instructions from: https://www.intel.com/content/www/us/en/support/articles/000057005/boards-and-kits.html Build was successful and environment successfully sourced. Entered OMZ demos (human_pose_estimation_3d_demo) and executed `omz_downloader --list models.lst` Resulting errors indicate versioning problem...

fastrtps compiler error: fatal error: 'tinyxml2.h' file not found

1

All the tinyxml2 sources including tinyxml2.h are in the thirdparty folder and eProsima CMakeLists.txt looks for it in thirdparty (eprosima_find_thirdparty(TinyXML2 tinyxml2). /home/brian/ros2_android_ws/src/eProsima/Fast-DDS/src/cpp/rtps/xmlparser/XMLProfileManager.cpp:15:10: fatal error: 'tinyxml2.h' file not found #include ^~~~~~~~~~~~...

Enable training with multiple GPU's as well as saving model generically when using dataparallel. Also call lr_scheduler after optimizer to bring in line with Pytorch 1.4 requirements.

Update call from deprecated name for Pytorch 1.4

NLLLoss2d is NLLLoss in Pytorch 1.4

First of all, thank you for your hard work on this and for sharing the training data. Can you please explain the cvset argument? I see it is calling different...

Update glossary.mdx with RL and DRL

1

Added 'reinforcement learning' and 'deep reinforcement learning' to the glossary.