Skip to content

Latest commit

 

History

History
64 lines (49 loc) · 1.86 KB

cli.md

File metadata and controls

64 lines (49 loc) · 1.86 KB
Raw

Entropy projects and the CLI

When running multiple experiments in the lab, it's good to have a way to divide and conquer. Entropy allows you to divides such work packages into projects.

Starting a project

When working from the command prompt, initializing a project is similar to initializing a git repository. You need to create a new folder for your project, change the working directory to that folder and then use the entropy Command Line Interface (CLI) to initialize the project.

mkdir my_proj
cd my_proj
entropy init

The directory will now contain a .entropy sub-directory where the sqlite database file and the hdf5 data file are stored. The project directory's name is effectively the project's name.

📁 my_proj    <--- project directory
  📁 .entropy
    📄 entropy.db
    📄 entropy.hdf5

You can then place the experiment files under my_proj:

📁 my_proj    <--- project directory
  📁 .entropy
    📄 entropy.db
    📄 entropy.hdf5
  📄 experiment1.py
  📄 experiment2.py

The CLI

The entropy CLI in installed when you install entropy with pip.

pip install entropylab

The CLI currently support two commands: init and upgrade.

init

entropy init <path to project directory>

Initializes a new project in the given directory (as described above)

upgrade

entropy upgrade <path to entropy project directory>

Takes an entropy db that predates the project structure (before version 0.3.0) and updates it as needed.

These are the steps executed:

  1. Moves the .db file (and corresponding .hdf5 file, if it exists) to a new project directory. The directory name will be the original .db file's name.
  2. Upgrades the .db file to the latest version of Entropy (if needed).
  3. Migrates experiment results and metadata from the .db file to .hdf5 (if needed).