csdllab / pyneval Goto Github PK

• Add conda recipe to build pyneval into a conda package
• Create a channel called csdl on anaconda to host the package

Note: use the conda-forge channel while building the package.

All conda recipe files can be put under a separate folder named recipe-pyneval.

If the argument of --output ends with .json, then automatically save results in the JSON format.

The option can be:

--detail <output_folder>

• If <output_folder> does not exist, the program should automatically create <output_folder>;
• if <output_folder> does not exist, it should ask the user to confirm if the results should be written to the existing folder.
  -- If the user answers no, give the user further options: 1. quit the program; 2. proceeding without saving details; 3. input a new folder.

Rename metric options to make them less verbose. The changes are: