Developed MLOps strategy for diamond pricing app

• Day: 2024-04-07
• Time: 15:50 to 16:35
• Project: Dev
• Workspace: WP 1: Strategic / Growth & Development
• Status: In Progress
• Priority: MEDIUM
• Assignee: Matías Nehuen Iglesias
• Tags: Mlops, Machine Learning, Project Management, Diamond Pricing

Description

Session Goal

The goal of this session was to develop a comprehensive MLOps strategy for a diamond price prediction application, focusing on enhancing the Flask application, optimizing Docker usage, planning cloud infrastructure, and establishing documentation best practices.

Key Activities

• Created a detailed agenda outlining objectives related to MLOps, Flask, Docker, and cloud technologies.
• Developed guiding questions to steer the project discussion, covering objectives, data quality, model metrics, infrastructure, user needs, compliance, budget, and timeline.
• Provided a comprehensive project overview, detailing objectives, current status, data sources, model performance, cloud preferences, infrastructure, use cases, security, budget, and timeline.
• Outlined a comprehensive MLOps strategy, detailing the lifecycle from model development to deployment, monitoring, and re-training.
• Structured a 10-day MLOps workflow plan focusing on quick setup, model training, deployment, and documentation.
• Planned integration of a new ‘special’ characteristic into the diamond dataset to enhance UI and model training for unique diamond pricing.

Achievements

• Successfully outlined a comprehensive MLOps strategy and a structured 10-day workflow plan.
• Developed a clear understanding of the project’s objectives, current status, and future direction.

Pending Tasks

• Implement the 10-day MLOps workflow plan.
• Integrate the ‘special’ characteristic into the diamond dataset and adjust the model training accordingly.

Evidence

• source_file=2024-04-07.sessions.jsonl, line_number=0, event_count=0, session_id=6caea3514ec75bd3743404ca69108521bb97239e1786e3f2d1dd74c66c8589a8
• event_ids: []