Nathan Louis

Create two new directories: metrics and losses. Each metric or loss would be self-contained in a separate file. Avoids extremely long .py files

Minor change, printed experiment setup is more legible.

Reference to #50 Adding new config parameter: num_gpus (default: -1) -1: Use all available GPUs 0 : Do not push model onto GPU i.e. use CPU >1: Use specified number...

Add option for DataParallel training in PyTorch. It's pretty straight forward, the only issues are when accessing the state_dict and functions belonging to model (for multi-gpu training). It becomes`model.module.state_dict` instead...

In Wiki

Along with citations and links

Specify different learning rates for the layers of a network

Could this also be used for video annotations? When editing a image sequence (video): Instead of annotating each frame from scratch, duplicate the annotations from the previous frame and just...