MAADSBML Usage

Installation

To use MAADSBML, first install docker engine in your computer:

STEP 1: Install Linux

You will need to have Linux OS installed

In Windows –you can install WSL (windows subsystem for Linux). Open Powershell or command prompt in Administrator mode and type:

wsl --install

Once wsl is installed then update the Linux distro:

sudo apt update & sudo apt upgrade

In Mac –Use Terminal

Or get a VM running with Linux Ubuntu installed

STEP 2: Install Docker

Install Docker: You can install Docker Desktop (Windows/Mac) Or in linux run:

sudo apt install docker.io

Give docker socket access to your network:

sudo chmod 666 /var/run/docker.sock

Checking if docker is properly installed do:

docker ps

STEP 3: Pull MAADSBML Docker Container

Pull the maadsbml docker container for Windows/Linux (AMD64):

docker pull maadsdocker/maads-batch-automl-otics

STEP 3b: Pull MAADSBML Docker Container (MAC/Linux Arm64)

Pull the maadsbml docker container for MAC/Linux (ARM64):

docker pull maadsdocker/maads-batch-automl-otics-arm64

STEP 4: Install MAADSBML Python Library

Install the MAADSBML Python library:

pip install maadsbml

Note

Furthe Setup and Configurations can be found here: MAADSBML Setup and Configurations

Running the MAADSBML Docker Container

Step 1: Create Local Folders

a. {YOUR LOCAL FOLDER PATH}/csvuploads

b. {YOUR LOCAL FOLDER PATH}/pdfreports

c. {YOUR LOCAL FOLDER PATH}/autofeatures

d. {YOUR LOCAL FOLDER PATH}/outliers

e. {YOUR LOCAL FOLDER PATH}/sqlloads

f. {YOUR LOCAL FOLDER PATH}/networktemp

g. {YOUR LOCAL FOLDER PATH}/networks

h. {YOUR LOCAL FOLDER PATH}/exception

i. {YOUR LOCAL FOLDER PATH}/staging

j. {YOUR LOCAL FOLDER PATH}/backup

{YOUR LOCAL FOLDER PATH} is the root folder path on your local machine: i.e. c:/maadsbml

Important

Once you created local folder - then use the Docker -v to map your local folders to the Docker container folders. If you do not do volume mappings, all data will be stored ONLY in the docker container and NOT ACCESSIBLE outside of the container. If you STOP the container ALL YOUR MAADSBML OUTPUT DATA WILL BE LOST.

MAADSBML Folder Explanation

MAADSBML Folder	Description
csvuploads	THIS IS WHERE YOU STORE YOUR OWN FILE FOR MAADSBML TRAINING
pdfreports	THIS IS WHERE YOU WILL FIND THE MAADSBML PDF REPORT
autofeatures	THIS IS WHERE YOU WILL FIND THE AUTOFEATURES
outliers	THIS IS WHERE YOU WILL FIND OUTLIERS
sqlloads	THIS IS A SYSTEM FOLDER
networktemp	THIS IS A SYSTEM FOLDER
networks	THIS IS WHERE THE ALGORITHMS ARE STORED
exception	THIS IS THE JSON FILE FOR THE ALGORITHM OUTPUT
staging	THIS IS A SYSTEM FOLDER
backup	THIS IS WHERE ALL YOUR BACKUP REPORTS ARE SAVED LOCALLY

MAADSBML Docker Run Command

You need to configure and run this command to start the MAADSBML container solution.

docker run --net=host -d  \
-v {YOUR LOCAL FOLDER PATH}/csvuploads:/maads/agentfilesdocker/dist/maadsweb/csvuploads:z \
-v {YOUR LOCAL FOLDER PATH}/pdfreports:/maads/agentfilesdocker/dist/maadsweb/pdfreports:z \
-v {YOUR LOCAL FOLDER PATH}/autofeatures:/maads/agentfilesdocker/dist/maadsweb/autofeatures:z \
-v {YOUR LOCAL FOLDER PATH}/outliers:/maads/agentfilesdocker/dist/maadsweb/outliers:z \
-v {YOUR LOCAL FOLDER PATH}/sqlloads:/maads/agentfilesdocker/dist/maadsweb/sqlloads:z \
-v {YOUR LOCAL FOLDER PATH}/networktemp:/maads/agentfilesdocker/dist/maadsweb/networktemp:z \
-v {YOUR LOCAL FOLDER PATH}/networks:/maads/agentfilesdocker/networks:z \
-v {YOUR LOCAL FOLDER PATH}/exception:/maads/agentfilesdocker/dist/maadsweb/exception:z \
-v {YOUR LOCAL FOLDER PATH}/staging:/maads/agentfilesdocker/dist/staging:z \
-v {YOUR LOCAL FOLDER PATH}/backup:/Viperviz/viperviz/views/backup:z \
-p 5595:5595 \
-p 5495:5495 \
-p 10000:10000 \
--env TRAININGPORT=5595 \
--env PREDICTIONPORT=5495 \
--env ABORTPORT=10000 \
--env COMPANYNAME='Your Company' \
--env MAXRUNTIME=120 \
--env ACCEL=0 \
--env MAINHOST=127.0.0.1 \
--env CHIP=AMD64 \
--env VIPERLOGNAME=bmllogs \
--env VIPERVIZPORT=9090 \
--env BROKERHOSTPORT=127.0.0.1:9092 \
--env KAFKACLOUDUSERNAME= \
--env KAFKACLOUDPASSWORD= \
maadsdocker/maads-batch-automl-otics

Important

Do not modify the right-hand side of the colon in the volume mapping. For example {YOUR LOCAL FOLDER PATH}/csvuploads :/maads/agentfilesdocker/dist/maadsweb/csvuploads:z

MAADSBML Docker Run Parameters Explained

Docker Run Parameter	Description
-d	runs the container in detached mode
-v	start of docker volume mapping
-p	port forwarding. For example, -p 5595:5595 means to forward your HOST port 5595 to the CONTAINER port 5595. Left-hand side of the colon is the HOST port, and right hand side of the colon is the container port: HOST:CONTAINER
--env	this is docker’s environment variable setting
TRAININGPORT	this is the port that MAADSBML listens on for training on your dataset
PREDICTIONPORT	this is the port that MAADSBML listens on for predictions after you have trained an algorithm on your dataset
ABORTPORT	this is the port that MAADSBML listens on for ABORTING the training run.
COMPANYNAME	You can specify your company name. This will appear on the MAADSBML pdf report output.
MAXRUNTIME	You can specify the maximum number of minutes to train on your dataset before MAADSBML aborts.
ACCEL	This is useful for training on VERY LARGE datasets. Set this to 1 if training on very large datasets (200K+ rows), otherwise leave at 0.
MAINHOST	This is the IP Address MAADSBML will listen on for connections.
CHIP	Set to AMD64 for Windows/Linux or ARM64 for Mac
VIPERLOGNAME	All MAADSBML logs are stored in this Kafka Topic
VIPERVIZPORT	This is the port for the MAADSBML dashboard. Refer to MAADSBML Real-Time Dashboard
BROKERHOSTPORT	This is the boker host and port for Kafka For on-premise Kafka use: 127.0.0.1:9092 For Cloud Kafka you need the cluster broker url You get this from your Clound vendor: AWS or Confluent
KAFKACLOUDUSERNAME	If using Kafka Cloud you must specify a cloud username This usually the API Key
KAFKACLOUDPASSWORD	If using Kafka Cloud you must specify the cloud password. This is usually the API Secret
maadsdocker/maads-batch-automl-otics	MAADSBML Docker container for Windows/Linux users
maadsdocker/maads-batch-automl-otics-arm64	MAADSBML Docker container for MAC/Linux users

Important

Port forwarding is needed to access MAADSBML container from Jupyter notebook or any other exteral application. MAADSBML is REST API compliant.

If Docker Successfully Setup

Go Inside the Container