Metadata-Version: 2.1
Name: accern-xyme
Version: 0.0.11
Summary: AccernXYME is a library for easily accessing XYME via python.
Home-page: https://github.com/Accern/accern-xyme
Author: Accern Corp.
Author-email: josua.krause@accern.com
License: MIT
Keywords: XYME AI machine learning client
Platform: UNKNOWN
Classifier: Development Status :: 4 - Beta
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.6
Classifier: Programming Language :: Python :: 3.7
Classifier: Programming Language :: Python :: 3.8
Requires-Dist: numpy (>=1.17.3)
Requires-Dist: pandas (>=0.25.3)
Requires-Dist: quick-server (>=0.7.8)
Requires-Dist: requests (>=2.22.0)
Requires-Dist: typing-extensions (>=3.7.4.1)
Provides-Extra: dev
Provides-Extra: test

Accern-XYME
===========

*accern\_xyme* is a python library for accessing XYME functionality.

|CircleCI|

.. |CircleCI| image:: https://circleci.com/gh/Accern/accern-xyme.svg?style=svg
   :target: https://circleci.com/gh/Accern/accern-xyme

Usage
-----

You can install *accern\_xyme* with pip:

.. code:: sh

    pip install --user accern-xyme

Import it in python via:

.. code:: python

    from accern_xyme import create_xyme_client

    client = accern_xyme.create_xyme_client(
        "https://xyme.accern.com/", "<USERNAME>", "<PASSWORD>")
    print(client.get_user_info())

:code:`<USERNAME>` and :code:`<PASSWORD>` are the login credentials for XYME.
The values can also be set to :code:`None` in which case the values must
be set in the environment variables :code:`ACCERN_USER`
and :code:`ACCERN_PASSWORD`. A login token can also be provided.

You will need python3.6 or later.

Exploring Workspaces
--------------------

The workspaces of the user can be retrieved via:

.. code:: python

    for (workspace, count) in client.get_workspaces().items():
        print(f"{workspace} contains {count} jobs"

And jobs in a given workspace can be retrieved via:

.. code:: python

    for job in client.get_jobs(workspace):
        print(f"{job.get_job_id()}: {job.get_name()} - {job.get_status()}"

Or directly by Job ID:

.. code:: python

    job = client.get_job_id("username_example_com/job_id")

Starting Jobs
-------------

A new job can be started via:

.. code:: python

    # creating the job
    job = client.create_job(schema=schema_obj, name="my job")

    with job.update_schema() as cur:
        # updating the schema
        cur["M"]["params"]["hidden_layer_sizes"] = [100, 100, 100]

    # starting the job
    job.start()

    import time
    time.sleep(30)

    print(job.get_status())

Computing Predictions
---------------------

Predictions can be obtained for a finished or running job:

.. code:: python

    # predict_proba is also available
    predictions, stdout = job.predict(df)
    print(stdout)

    print("prediction of first row: ", predictions.iloc[0])


