drucker-client

A Python gRPC client for Drucker.


Keywords
Drucker, Kubernetes, Python, client, gRPC, ml, rekcurd
License
Apache-2.0
Install
pip install drucker-client==0.4.0

Documentation

rekcurd-client

Build Status PyPI version codecov pypi supported versions

Rekcurd client is the project for integrating ML module. Any Rekcurd service is connectable. It can connect the Rekcurd service on Kubernetes.

Parent Project

https://github.com/rekcurd/community

Components

Installation

From source:

git clone --recursive https://github.com/rekcurd/python-client.git
cd python-client
python setup.py install

From PyPi directly:

pip install rekcurd_client

How to use

Example is available here.

Unittest

$ python -m unittest

Method definition

You need to use an appropriate method for your Rekcurd service. The methods are generated according to the input and output formats. V is the length of feature vector. M is the number of classes. If your algorithm is a binary classifier, you set M to 1. If your algorithm is a multi-class classifier, you set M to the number of classes.

method input: data
(required)
input: option output: label
(required)
output: score
(required)
output: option
run_predict_string_string string string (json) string double string (json)
run_predict_string_bytes string string (json) bytes double string (json)
run_predict_string_arrint string string (json) int[M] double[M] string (json)
run_predict_string_arrfloat string string (json) double[M] double[M] string (json)
run_predict_string_arrstring string string (json) string[M] double[M] string (json)
run_predict_bytes_string bytes string (json) string double string (json)
run_predict_bytes_bytes bytes string (json) bytes double string (json)
run_predict_bytes_arrint bytes string (json) int[M] double[M] string (json)
run_predict_bytes_arrfloat bytes string (json) double[M] double[M] string (json)
run_predict_bytes_arrstring bytes string (json) string[M] double[M] string (json)
run_predict_arrint_string int[V] string (json) string double string (json)
run_predict_arrint_bytes int[V] string (json) bytes double string (json)
run_predict_arrint_arrint int[V] string (json) int[M] double[M] string (json)
run_predict_arrint_arrfloat int[V] string (json) double[M] double[M] string (json)
run_predict_arrint_arrstring int[V] string (json) string[M] double[M] string (json)
run_predict_arrfloat_string double[V] string (json) string double string (json)
run_predict_arrfloat_bytes double[V] string (json) bytes double string (json)
run_predict_arrfloat_arrint double[V] string (json) int[M] double[M] string (json)
run_predict_arrfloat_arrfloat double[V] string (json) double[M] double[M] string (json)
run_predict_arrfloat_arrstring double[V] string (json) string[M] double[M] string (json)
run_predict_arrstring_string string[V] string (json) string double string (json)
run_predict_arrstring_bytes string[V] string (json) bytes double string (json)
run_predict_arrstring_arrint string[V] string (json) int[M] double[M] string (json)
run_predict_arrstring_arrfloat string[V] string (json) double[M] double[M] string (json)
run_predict_arrstring_arrstring string[V] string (json) string[M] double[M] string (json)

The input "option" field needs to be a json format. Any style is Ok but we have some reserved fields below.

Field Type Description
suppress_log_input bool True: NOT print the input and output to the log message.
False (default): Print the input and output to the log message.