OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Infrastructure & Serving

Sales Forecast Mlops At Scale

Sales Forecast MLOps at Scale is an MIT-licensed, cloud-native machine-learning system for ingesting sales data, generating demand forecasts, and operating the associated training and inference pipeline. It combines streaming, scheduled and on-demand retraining, model tracking, monitoring, and a Streamlit forecast interface.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Sales Forecast Mlops At Scale — OpenViST project profile in AI Infrastructure & Serving.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want a self-hosted reference system for moving from streaming sales data to trained models and viewable forecasts, or to study an integrated MLOps deployment.

Who it is for

  • Teams exploring or assembling a self-hosted sales-demand forecasting pipeline.
  • MLOps practitioners who need to examine a system spanning streaming ingestion, model training, inference and monitoring.

Documented uses

  • Ingest sales records from a Kafka stream and process them into PostgreSQL.
  • Schedule weekly Prophet model training and batch forecasts, with on-demand retraining described as a capability.
  • Track models and versions with MLflow and monitor pipeline or system activity with Grafana and Prometheus.
  • Display forecasts by store and product through a Streamlit interface.
  • Set up the system with Docker Compose or Kubernetes/Helm, including a documented GCP deployment path.
Before you choose

Requirements & limits

The repository is public and specifies the MIT license. The project is self-hosted; the README lists Docker, Kubernetes and Helm as setup prerequisites.

  • The system combines Kafka, Spark Streaming, Airflow, Prophet, Ray, MLflow and PostgreSQL, so it spans data engineering and model operations as well as inference.
  • The README documents batch and online inference, weekly scheduled forecasts and on-demand retraining; the described store-and-product interface displays the next seven days of forecasts.
  • Deployment routes include Docker Compose and Kubernetes/Helm, with GCP named as a cloud option.
  • The README describes the stack as large and resource-intensive and cautions that local Kubernetes may be best for setup testing.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Start the stack with Docker Compose

Set up the project services using the README's Docker Compose route.

Before you start

  • Docker
  • The project repository, including its Compose files

Steps

  1. Optionally build the images with docker-compose build.

  2. Start the services with docker-compose -f docker-compose.yml -f docker-compose-airflow.yml up -d.

Expected outcome

The README presents this as the Docker Compose startup path for the system.

Guide limitations

  • The README notes startup may freeze or fail on the first attempt, especially on lower-spec machines.
  • Service restart behavior is not specified for most services, so they will not restart after failures.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
24 October 2026
Original website title
jomariya23156/sales-forecast-mlops-at-scale · Official homepage title · retained in its source language
How we review sourcesSuggest a correction