gravityai 0.1.4

The gravity-ai.com helper package

Gravity-ai.com

This package includes helper scripts for interfacing your machine learning code with the Gravity-ai.com containerization code.

To install from pypi:

pip install gravityai

To install from source:

python3 setup.py install

Example Implementation Code

from gravityai import gravityai as grav

# the following function is a callback (defined by you), that may be
# async or synchronous and that may be called multiple times, to
# transform input data against an algorithm or model (or whatever you want)
# If an error is experienced, return an error string message, or throw an exception.
def process_data(dataPath, outPath):
    # TODO:
    # Read data in from dataPath
    # Transform Data via model
    # Write data out to outPath
    # Return None if everything went ok.
    # return an error string if there was a problem.


# TODO Initialize models, etc. before calling wait_for_requests.

grav.wait_for_requests(process_data)

Running Multiple Files

This section explains how to use multiple files. You will require a csv with the column "uri", where each row is a uri where an input file is stored:

from gravityai import gravityai as grav

'''
In continuation of the section above, define a callback function:

def process_data(dataPath, outPath):
    #TODO:
    # 1) Read data in from dataPath (csv containing column "uri")
    # NOTE: dataPath will likely end with a non-standard file extension (like .dat), so make sure your code can handle
    # a filename with an extension different than expected. You may need to rename the input file to .zip. .csv, etc for it to load
    # properly in your code.
    # 2) Define a custom handling function to process the input file, with whatever arguments are necessary:
    def handle_fnc(filePath, row, **kwargs):
        # read file from filePath, read whatever data is necessary from the Pandas df row.
        # perform processing
        # return pandas df after processing
    grav.handle_csvs_with_uris(dataPath, outPath, handle_fnc, **kwargs)
'''

grav.wait_for_requests(process_data)

Building a new Version

To build a new version for pypi (only we do that):

python3 -m pip install --user --upgrade setuptools wheel

python3 -m pip install --user --upgrade twine

python3 setup.py sdist bdist_wheel

python3 -m twine upload --repository pypi dist/*

Use the saved credentials, and remember the username is __token__