https://github.com/incalia/schedy-client
All your Data Science Experiments in one place
https://github.com/incalia/schedy-client
deep-learning hyperparameter-optimization machine-learning
Last synced: 6 months ago
JSON representation
All your Data Science Experiments in one place
- Host: GitHub
- URL: https://github.com/incalia/schedy-client
- Owner: incalia
- License: mit
- Created: 2017-12-21T10:20:55.000Z (over 8 years ago)
- Default Branch: master
- Last Pushed: 2019-01-08T15:40:22.000Z (over 7 years ago)
- Last Synced: 2025-12-20T10:46:56.314Z (6 months ago)
- Topics: deep-learning, hyperparameter-optimization, machine-learning
- Language: Python
- Homepage: https://schedy.io
- Size: 323 KB
- Stars: 6
- Watchers: 6
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.rst
- License: LICENSE.txt
Awesome Lists containing this project
README
Welcome to Schedy's documentation!
==================================
**Schedy** is your machine learning assistant. It will help you record your
experiments, your results, visualize them, and it will even suggest new
parameters for your next experiments!
Schedy can do useful things for you:
- Record the hyperparameters and results of all your past models.
- Suggest new hyperparameters for your next models.
- Coordinate a pool of workers (e.g. in a cluster), by making sure they
stay busy trying to find the best combination of hyperparameters for
your task.
And all of that in just a few lines of code! Coordinating a cluster of workers
becomes as simple as this:
.. code-block:: python
import schedy
db = schedy.SchedyDB()
exp = db.get_experiment('My Task')
while True:
with exp.next_job() as job:
my_train_function(job)
You can follow the evolution of your experiments thanks to our `online dashboard
`_.
.. image:: https://raw.githubusercontent.com/incalia/schedy-client/master/docs/images/dashboard.png
:align: center
We also provide a command line tool, that will help you with the most
repetitive tasks.
.. _setup:
Installation and setup
----------------------
`Sign up here `_, install Schedy & get your API token:
.. code-block:: bash
pip3 install schedy
schedy gen-token
You are now ready to `get started `_!
You can also read the `API reference `_.