A multi-tenant multi-client scalable product categorising demo stack

Overview

Better Categories 4All: A multi-tenant multi-client product categorising stack

The steps to reproduce training and inference are in the end of this file, sorry for the long explanation.

example workflow

Problem scope

We want to create a full product categorization stack for multiple clients. For each client, and each product we want to find the 5 most suitable categories.

Project structure

The project is split into two layers:

  • ML layer: the python package for training and serving model. It's a pipenv based project. The Pipfile include all required dependencies. The python environment generated by pipenv is used to run the training/inference and run also unit tests. Code is generic for all clients.
  • Orchestration layer: the Airflow DAGs for training and prediction. Each client has its own training DAG and its prediction DAG. These DAGs uses the Airflow BashOperator to execute training and prediction inside the pipenv environment.

img_1.png

Why one DAG per a client instead of a single DAG for all client ?

We could have a single DAG that train all clients. So each client has its own training task inside the same DAG. I chose rather to build a separate DAG for each client. Several reasons motivated my decision:

  • In my past experiences, some individual cients may have problem s with their data and it's more practical to have a DAG per client when it's come to day to day monitoring.
  • New clients may come and other may leave and we may endup with a DAG that keeps constantly adding new Task and loosing others and it's against airflow best practicies.
  • It make sens to have one failed DAG and 99 other successful DAGs rather than a single DAG failing all the time because of one random client training failing each day.

Training

In this part we will train a classification model for each client.

Training package

The package categories_classification include a training function train_model. It takes the following inputs:

  • client_id: the id of the client in training dataset
  • features: a list of features names to use in training
  • model_params: a dict of params to be passed to model python class.
  • training_date: the execution date of training, used to track the training run.

The chosen model is scikit-learn implementation of random forest sklearn.ensemble.RandomForestClassifier. For the sake of simplicity, we didn't fine tune model parameters, but optimal params can be set in config.

In addition to train_model function, a cli binary is created to be able to run training directly from command line. The binary command trainer runs the training:

pipenv run python categories_classification_cli.py trainer --help

Usage: categories_classification_cli.py trainer [OPTIONS]

Options:
  --client_id TEXT     The id of the client.  [required]
  --features TEXT      The list of input features.  [required]
  --model_params TEXT  Params to be passed to model.  [required]
  --training_date TEXT  The training date.  [required]
  --help               Show this message and exit.

Data and model paths

All data are stored in a command base path retrieved from environment variable DATA_PREFIX, default is ./data. Given a client id, training data is loaded from $DATA_PREFIX/train/client_id= /data_train.csv.gz .

Splitting data

Before training, data is split into training set and test set. The train set is used to train the model while the test set is used to evaluate the model after training. Evaluation score is logged.

Model tracking and versioning

The whole training event is tracked in Mlfow as a training run. Each client hash its own experiments and its own model name following the convention " _model". The tracking process saves also metrics and model parameters in the same run metadata.

Finally, the model is saved in Mlflow Registry with name " _model". Saving the model means a new model version is saved in Mlflow, as the same model may have multiple versions.

Prediction

In this part, we will predict product categories using previously trained model.

Prediction package

The package categories_classification include a prediction function predict_categories. It takes the following inputs:

  • client_id: the id of the client in training dataset
  • inference_date: an inference execution date to version output categories

The prediction is done through spark so that it can be done on big datasets. Prediction dataset is loaded in spark DataFrame. We use Mlflow to get the latest model version and load latest model. The model is then broadcasted in Spark in order to be available in Spark workers. To apply the model to the prediction dataset, I use a new Spark 3.0 experimental feature called mapInPandas. This Dataframe method maps an iterator of batches (pandas Dataframe) using a prediction used-defined function that outputs also a pandas Dataframe. This is done thanks to PyArrow efficient data transfer between Spark JVM and python pandas runtime.

Prediction function

The advantage of mapInPandas feature comparing to classic pandas_udf is that we can add more rows than we have as input. Thus for each product, we can output 5 predicted categories with their probabilities and ranked from 0 to 4. The predicted label are then persisted to filesystem as parquet dataset.

Model version retrieval

Before loading the model, we use Mlflow to get the latest version of the model. In production system we probabilities want to push model to staging, verify its metrics or validate it before passing it to production. Let's suppose that we are working the same stage line, we use MlflowClient to connect to Mlflow Registry and get the latest model version. The version is then used to build the latest model uri.

Reproducing training and inference

Pipenv initialization

First you need to check you have pipenv installed locally otherwise you can install it with pip install pipenv.

Then you need to initialize the pipenv environment with the following command:

make init-pipenv

This may take some time as it will install all required dependencies. Once done you can run linter (pylint) and unit tests:

make lint
make unit-tests

Airflow/Mlflow initialization

You need also to initialize the local airflow stack, thus building a custom airflow docker image including the pipenv environment, the mlflow image and initializing the Airflow database.

make init-airflow

Generate DAGs

Airflow dags needs to be generated using config file in conf/clients_config.yaml. It's already created with the 10 clients example datasets. But if you want you can add new clients or change the actual configuration. For each client you must include the list of features and optional model params.

Then, you can generate DAGs using the following command:

make generate-dags

This will can the script scripts/generate_dags.py which will:

  • load training and inference DAG templates from dags_templates, they are jinja2 templates.
  • load conf from conf/clients_config.yaml
  • render DAG for each client and each template

Start local Airflow

You can start local airflow with following command:

make start-airflow

Once all services started, you can go to you browser and visit:

  • Airflow UI in http://localhost:8080
  • Mlflow UI in http://localhost:5000

Run training and inference

In Airflow all DAGs are disabled by default. To run training for a client you can enable the DAG and it will immediately trigger the training.

Once the model in Mlflow, you can enable the inference DAG and it will immediately trigger a prediction.

Inspect result

To inspect result you run a local jupyter, you do it with:

make run-jupyter

Then visit notebook inspect_inference_result.ipynb and run it to check the prediction output.

A small bot to interact with the reddit API. Get top viewers and update the sidebar widget.

LiveStream_Reddit_Bot Get top twitch and facebook stream viewers for a game and update the sidebar widget and old reddit sidebar to show your communit

Tristan Wise 1 Nov 21, 2021
Volt is yet another discord api wrapper for Python. It supports python 3.8 +

Volt Volt is yet another discord api wrapper for Python. It supports python 3.8 + How to install [Currently Not Supported.] pip install volt.py Speed

Minjun Kim (Lapis0875) 11 Nov 21, 2022
A Python Script to automate searching of available vaccination centers in the city and hence booking

Cowin Vaccine Availability Notifier Cowin Vaccine Availability Notifier takes your City or PIN code as an input and automatically notifies you via ema

Jayesh Padhiar 7 Sep 05, 2021
📦 Opensource Python wrapper for Hiven's REST and WebSocket API

hiven.py 📦 Opensource Python wrapper for Hiven's REST and WebSocket API Installation pip install -U hiven.py Usage hiven.py is currently under devel

Kevin Thomas 3 Sep 03, 2021
Powerful and Advance Telegram Bot with soo many features😋🔥❤

Chat-Bot Reach this bot on Telegram Chat Bot New Features 🔥 ✨ Improved Chat Experience ✨ Removed Some Unnecessary Commands ✨ Added Facility to downlo

Sanila Ranatunga 10 Oct 21, 2022
Official Python client for the MonkeyLearn API. Build and consume machine learning models for language processing from your Python apps.

MonkeyLearn API for Python Official Python client for the MonkeyLearn API. Build and run machine learning models for language processing from your Pyt

MonkeyLearn 157 Nov 22, 2022
Telegram bot using python

Telegram bot using python

Masha Kubyshina 1 Oct 11, 2021
Schedule Twitter updates with easy

coo: schedule Twitter updates with easy Coo is an easy to use Python library for scheduling Twitter updates. To use it, you need to first apply for a

wilfredinni 46 Nov 03, 2022
💻 Discord-Auto-Translate-Bot - If you type in the chat room, it automatically translates.

💻 Discord-Auto-Translate-Bot - If you type in the chat room, it automatically translates.

LeeSooHyung 2 Jan 20, 2022
Linkvertise-bypass - Tools pour bypass les liens Linkvertise

Installation | Important | Discord 🌟 Comme Linkvertise bypass est gratuit, les

GalackQSM 3 Aug 31, 2022
🚀 An asynchronous python API wrapper meant to replace discord.py - Snappy discord api wrapper written with aiohttp & websockets

Pincer An asynchronous python API wrapper meant to replace discord.py ❗ The package is currently within the planning phase 📌 Links |Join the discord

Pincer 125 Dec 26, 2022
Portal Backend for Yuta management

Portal Backend for Yuta management Prerequisites Python 3.10 or above. pip, pdm installed. Quickstart Install the required packages: pdm install Runn

Loc Mai 1 Dec 20, 2021
A script to find the people whom you follow, but they don't follow you back

insta-non-followers A script to find the people whom you follow, but they don't follow you back Dependencies: python3 libraries - instaloader, getpass

Ritvik 5 Jul 03, 2022
E-Commerce Telegram Bot for UCA Students

ucaStudentStore To buy from and sell to other students Features Register the first time, after that you will always be recognised You can login either

Shukur Sabzaliev 5 Jun 26, 2022
News API consisting various sources from Tanzania

Tanzania News API News API consisting various sources from Tanzania. Fork the project Clone the project git clone https://github.com/username/news-a

Innocent Zenda 6 Oct 06, 2022
A Flask & Twilio Secret Santa app.

🎄 ✨ Secret Santa Twilio ✨ 📱 A contactless Secret Santa game built with Python, Flask and Twilio! Prerequisites 📝 A Twilio account. Sign up here ngr

Sangeeta Jadoonanan 5 Dec 23, 2021
Unofficial Medium Python Flask API and SDK

PyMedium - Unofficial Medium API PyMedium is an unofficial Medium API written in python flask. It provides developers to access to user, post list and

Engine Bai 157 Nov 11, 2022
Bot Realm of the Mad God Exalt (ROTMG). (Auto_nexus, Auto_HP, Auto_Loot)

Bot_ROTMG Bot Realm of the Mad God Exalt (ROTMG). (Auto_nexus, Auto_HP, Auto_Loot) *Este projeto visa apenas o aprendizado, quem faz mal uso é o único

Guilherme Silva Uchoa 1 Oct 30, 2021
Framework to collect and process weather data from wttr.in.

Weathercrawler Automatic extraction and processing framework for weather data from wttr.in Installation tested with: Python 3.7.3 Python 3.9.4 git clo

Maurice Günder 0 Jul 26, 2021
Ethereum Gas Fee for the MacBook Pro touchbar (using BetterTouchTool)

Gasbar Ethereum Gas Fee for the MacBook Pro touchbar (using BetterTouchTool) Worried about Ethereum gas fees? Me too. I'd like to keep an eye on them

TSS 51 Nov 14, 2022