omegaml

MNIST Digits Recognition using Machine Learning This is a live implementation of the MNIST digits recognition problem using a model trained & deployed by omega|ml

The model is adopted from this scikit-learn example (https://scikit-learn.org/stable/auto_examples/classification/plot_digits_classification.html) The model has been trained with and deployed by omega|ml with just one line of code.

How to use:

1. in Jupyter NB copy/paste/run the code from the scikit learn example https://scikit-learn.org/stable/auto_examples/classification/plot_digits_classification.html

omegaml

# Using an omegaml dataset as a source and sink
#
# * No need to have a MQTT or Kafka broker
# * Use the omegaml REST API for both input and output to support any client
from minibatch.contrib.omegaml import DatasetSource, DatasetSink
import minibatch as mb

# cleanup (not strictly required, i.e. can be kept ongoing)
om.datasets.drop('stream-sink', force=True)
om.datasets.drop('stream-sample', force=True)

omegaml

How to develop an omega|ml plugin

omega|ml plugins come in two forms:

• Mixins - extend the functionality of an existing component, e.g. om.datasets, om.modelsor om.runtime
• Backends - add new capabilities to store and process a certain type of model (framework) or data object

If you want to add support for a new machine learning framework that omega|ml does not support yet, implement a backend. If you want to add pre- and post-processing to some method already provided by omega|ml, implement a mixin.

omegaml

"""
omegaml write and read large CSV files from/to remote or local file paths
Supports s3, hdfs, http/s, sftp, scp, ssh

This is sample code for and provided for convenience. Adopt to your specific
requirement as needed.

Installation:
    !pip install -q getgist
    !rm -f *omx_iotools.py && getgist -y omegaml omx_iotools.py

omegaml

omegaml

# version of this plugin
version = '0.1.1'

"""
Pre-release of omegaml parallel processing for MDataFrame

Usage:
    def myproc(df):
        # do something with df, a in-memory Pandas DataFrame
        # return None to store the updated df, return a DataFrame or Series

omegaml

from omegaml.backends.basedata import BaseDataBackend
from base64 import b64encode, b64decode
from sqlalchemy import create_engine

import pandas as pd

# version of this plugin
version = '0.1.3'

class SnowflakeDataBackend(BaseDataBackend):

omegaml

omegaml

omegaml

istrue = lambda v: (
    (v.lower() in ('yes', '1', 'y', 'true', 't'))
    if isinstance(v, str) else bool(v)
)
isfalse = lambda v: not istrue(v)