Skip to main content

An automated AI/ML solution from Marlabs

Project description

mAdvisor AutoML by Marlabs

Python

mAdvisor AutoML is an automated AI/ML solution from Marlabs that translates data into meaningful insights & predictions without any manual intervention. AutoML gives you the power of cognitive technologies like machine learning, machine reasoning, deep learning, natural language generation, natural language processing and expert rules systems with your limited knowledge of AI/ML, thereby enabling enterprises to identify revenue streams, enhance customer experience and productivity. This solution is designed for application and machine experts, so that machine learning models can be created with no help from a data scientist. A 30-day trial license for mAdvisor AutoML® is issued on activation.

The solution includes the following features:

  1. Ability to comprehend and monetize Big Data
  2. Rapid time to insights
  3. No dependency on data scientists & analysts to create briefs
  4. Rapid development of predictive apps
  5. Expandable and Scalable to the adoption of new use cases

Installation

pip

$ pip install mAdvisor

Usage

  • A Licence Key is required to use this package
  • Register yourself to activate the 30 days free trial
  • Connect with Marlabs mAdvisor team to purchase a paid licence.

User Registration and Plan Subscription

Sign Up

import requests
base_url = 'https://madvisor-dbc.marlabsai.com/automl'
url = base_url + '/account/sign-up'
data= {'email': '<email id>',
 'username':'<user name>',
 'password':'<password>'
}
response = requests.post(url, data=data)
if response.status_code != 200:
    print('Failed response code {}'.format(response.status_code))
print('Output result: {}'.format(response.json()))

Subscribe to the 30 days trial plan

url = base_url+'/subscription/subscribe-plan'
data= {'username':'<user name>',
 'password':'<password>',
 'subscription_type': 'TRIAL',
 'plan': 'FREE'
 }

response = requests.post(url, data=data)
if response.status_code != 200:
    print('Failed response code {}'.format(response.status_code))
print('Output result: {}'.format(response.json()))

Check Your Active Subscriptions

url = base_url+'/subscription/active-plans'
data= {'username':'<user name>',
 'password':'<password>'
 }

response = requests.post(url, data=data)
if response.status_code != 200:
    print('Failed response code {}'.format(response.status_code))
print('Output result: {}'.format(response.json()))

Start Using the Library

AutoML model training

This class is used to initiate AutoML training job, user has to use the slug value returned from here while doing prediction.

from mAdvisor import train
model = train(train_data="<train dataset pass either path to the file or dataframe>",
               target="<Target Variable Name>",
               token = "<Licence Key>")
model_slug, _ = model.fit()

AutoML prediction

Train data prediction can be using this class

from mAdvisor import score
model = score(test_data="<test dataset pass either path to the file or dataframe>",
               training_slug="<model slug value received from train output>",
               token = "<Licence Key>")
prediction_rules, predicted_data, _ = model.fit()

AutoML model Training and Prediction in a single go

Used to initiate automl job, both model training and scoring can be done using this class.

from mAdvisor import automl
model = automl(train_data="<train dataset pass either path to the file or dataframe>",
               test_data="<test dataset pass either path to the file or dataframe>",
               target="<Target Variable Name>",
               token = "<Licence Key>")
model_slug, prediction_rules, predicted_data, _ = model.fit()

Automated Data Preprocessing, Feature engineering and Feature Selection for train data

Feature engineering module will return two dataframes, one for linear algorithms & second one for tree based algorithms, and a slug value which is to be used while preparing test data.

from mAdvisor import AutoFE
preprocess = AutoFE(target="<Target Variable Name>",
                    train_df = "train dataframe to be used",
                    token = "<Licence Key>")
linear_df, tree_df, fe_slug, _ = preprocess.fit()

Automated Data Preprocessing, Feature engineering and Feature Selection for test data

Feature engineering module will return two dataframes, one for linear algorithms and second one for tree based algorithms.

from mAdvisor import AutoFE_Test
preprocess_test = AutoFE_Test(test_df = "<test dataframe to be used>",
			 fe_slug="<Unique slug value received from train data feature engineering>",
                   	 token = "<Licence Key>")
linear_df_test, tree_df_test, _ = preprocess_test.fit()

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distributions

mAdvisor-0.3.6-cp38-cp38-win_amd64.whl (7.3 MB view details)

Uploaded CPython 3.8Windows x86-64

mAdvisor-0.3.6-cp38-cp38-manylinux2014_x86_64.whl (89.6 MB view details)

Uploaded CPython 3.8

mAdvisor-0.3.6-cp37-cp37m-win_amd64.whl (6.7 MB view details)

Uploaded CPython 3.7mWindows x86-64

mAdvisor-0.3.6-cp37-cp37m-manylinux2014_x86_64.whl (34.3 MB view details)

Uploaded CPython 3.7m

mAdvisor-0.3.6-cp36-cp36m-win_amd64.whl (6.6 MB view details)

Uploaded CPython 3.6mWindows x86-64

mAdvisor-0.3.6-cp36-cp36m-manylinux2014_x86_64.whl (34.6 MB view details)

Uploaded CPython 3.6m

File details

Details for the file mAdvisor-0.3.6-cp38-cp38-win_amd64.whl.

File metadata

  • Download URL: mAdvisor-0.3.6-cp38-cp38-win_amd64.whl
  • Upload date:
  • Size: 7.3 MB
  • Tags: CPython 3.8, Windows x86-64
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.7.0 requests/2.25.1 setuptools/47.1.1.post20200604 requests-toolbelt/0.9.1 tqdm/4.61.1 CPython/3.6.10

File hashes

Hashes for mAdvisor-0.3.6-cp38-cp38-win_amd64.whl
Algorithm Hash digest
SHA256 203ff559d83d3578cff6df6f8b509c3f2c2f6fb2e4a54e661f4e2615d5a8f539
MD5 ef02c8d372c3998ebdf8d3659b38815f
BLAKE2b-256 cced182e5db2fea651f8b4a159f78e210336f814bf87cddfcac99b1824af0002

See more details on using hashes here.

File details

Details for the file mAdvisor-0.3.6-cp38-cp38-manylinux2014_x86_64.whl.

File metadata

  • Download URL: mAdvisor-0.3.6-cp38-cp38-manylinux2014_x86_64.whl
  • Upload date:
  • Size: 89.6 MB
  • Tags: CPython 3.8
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.7.0 requests/2.25.1 setuptools/47.1.1.post20200604 requests-toolbelt/0.9.1 tqdm/4.61.1 CPython/3.6.10

File hashes

Hashes for mAdvisor-0.3.6-cp38-cp38-manylinux2014_x86_64.whl
Algorithm Hash digest
SHA256 d637fd65c428226cbb90dcb0faf3afdd63e0666964f28ee309be3130372057d7
MD5 1db1d1a0f8de20ccd06e45bd01298034
BLAKE2b-256 f7499a70c03afcccd2dae2659ca5cab6ef6564c376e941a682b41fa856d3197a

See more details on using hashes here.

File details

Details for the file mAdvisor-0.3.6-cp37-cp37m-win_amd64.whl.

File metadata

  • Download URL: mAdvisor-0.3.6-cp37-cp37m-win_amd64.whl
  • Upload date:
  • Size: 6.7 MB
  • Tags: CPython 3.7m, Windows x86-64
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.7.0 requests/2.25.1 setuptools/47.1.1.post20200604 requests-toolbelt/0.9.1 tqdm/4.61.1 CPython/3.6.10

File hashes

Hashes for mAdvisor-0.3.6-cp37-cp37m-win_amd64.whl
Algorithm Hash digest
SHA256 d440673a986aefc9f277c9212c288e4e814f0f6ad26f3c2f0f3b0c6962b1cca2
MD5 a1786aca9e88bf472b7e71fb3d9216f0
BLAKE2b-256 671f0e0f936d6af3bbe78579bd22379f274023d7834d4568b77daec9b937ea21

See more details on using hashes here.

File details

Details for the file mAdvisor-0.3.6-cp37-cp37m-manylinux2014_x86_64.whl.

File metadata

  • Download URL: mAdvisor-0.3.6-cp37-cp37m-manylinux2014_x86_64.whl
  • Upload date:
  • Size: 34.3 MB
  • Tags: CPython 3.7m
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.7.0 requests/2.25.1 setuptools/47.1.1.post20200604 requests-toolbelt/0.9.1 tqdm/4.61.1 CPython/3.6.10

File hashes

Hashes for mAdvisor-0.3.6-cp37-cp37m-manylinux2014_x86_64.whl
Algorithm Hash digest
SHA256 7aa12c167053e042b519eac251b5756140d5f154c83a248d886e14f31322d9d4
MD5 4827dec8e87de45336d6e2cc7c6a1ee8
BLAKE2b-256 0a1788384d3ae7de2ccdf6562329c68fe9619b027d182ab8d1ab2a8cd2b8c8de

See more details on using hashes here.

File details

Details for the file mAdvisor-0.3.6-cp36-cp36m-win_amd64.whl.

File metadata

  • Download URL: mAdvisor-0.3.6-cp36-cp36m-win_amd64.whl
  • Upload date:
  • Size: 6.6 MB
  • Tags: CPython 3.6m, Windows x86-64
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.7.0 requests/2.25.1 setuptools/47.1.1.post20200604 requests-toolbelt/0.9.1 tqdm/4.61.1 CPython/3.6.10

File hashes

Hashes for mAdvisor-0.3.6-cp36-cp36m-win_amd64.whl
Algorithm Hash digest
SHA256 23908b873cfc6da54f4686c0de888619fb819c76bcadbd9a405da6e0e1bd4c40
MD5 c3581ca3a82106fc3d55237594e5962c
BLAKE2b-256 e3250a2dcb3c67bdc067cc9e5c88deb1509a9cc9def9c6b4daea531e65021b6c

See more details on using hashes here.

File details

Details for the file mAdvisor-0.3.6-cp36-cp36m-manylinux2014_x86_64.whl.

File metadata

  • Download URL: mAdvisor-0.3.6-cp36-cp36m-manylinux2014_x86_64.whl
  • Upload date:
  • Size: 34.6 MB
  • Tags: CPython 3.6m
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.3.0 pkginfo/1.7.0 requests/2.25.1 setuptools/47.1.1.post20200604 requests-toolbelt/0.9.1 tqdm/4.61.1 CPython/3.6.10

File hashes

Hashes for mAdvisor-0.3.6-cp36-cp36m-manylinux2014_x86_64.whl
Algorithm Hash digest
SHA256 2d39b64ee5be41fb7a90b960fe880d082c39b3660489d7cf20248550e0aff298
MD5 12ac16d6d2375affbc609c22d880b4a6
BLAKE2b-256 db74f8cec448e95f6d9df148ca274c43739a2e3e610a97b0069cb7684190246a

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page