-
Decision Trees, Random Forests and Gradient Boosting in Python (Data Science Series, Part 12)
A single decision tree scored a perfect 1.0000 on its training data and 0.6222 on held out folds. Here is how bagging and boosting repair that, and why on our churn data the ensembles barely beat plain logistic regression.
-
Model Evaluation in Python: Metrics, Cross Validation and Data Leakage (Data Science Series, Part 11)
A single train test split moved our churn AUC by nine points depending on the seed. Here is how to pick a metric that matches the decision, read a cross validation spread honestly, and catch the leakage that manufactures scores you should not believe.
-
Logistic Regression in Python and Your First Classifier (Data Science Series, Part 10)
Build a churn classifier with logistic regression, read its coefficients as odds ratios, and learn why 73.77 percent accuracy can mean your model never predicted a single churn.
-
Linear Regression in Python From the Inside Out (Data Science Series, Part 9)
Least squares is three lines of numpy once you see the geometry. This part fits, checks and defends a linear model, then shows the collinearity failure that produces nonsense coefficients without ever raising an error.
-
Statistical Inference for Machine Learning: Sampling, Confidence Intervals and What a p Value Is Not (Data Science Series, Part 8)
Every metric you report is one draw from a distribution. Here is how to put a confidence interval on it, when to bootstrap, and the three readings of a p value that quietly wreck model selection.
-
Probability and Distributions a Modeller Actually Needs (Data Science Series, Part 7)
Probability is what separates a model that ranks customers from a model you can attach money to. Here is the working subset a modeller needs, with a churn worked example showing why a well ranked model can still lose cash.
-
Feature Engineering in Python: Where Model Accuracy Actually Comes From (Data Science Series, Part 6)
Encoding, ratio features and cross fitted target encoding on the churn table, with a measured leakage demo that turns a column of random noise into an AUC of 0.800.
-
Getting Data Into Python: APIs, Files, SQL Pulls and Formats That Bite (Data Science Series, Part 5)
Most model errors enter at the moment data is read. Here is how I pull from paginated APIs, run SQL into pandas safely, and pick a file format, with measured numbers from the churn project.
-
NumPy and pandas Past the Basics: Vectorisation, Merges, Reshaping and Memory (Data Science Series, Part 4)
Most pandas mistakes do not raise an error. They cost you memory, hours of runtime, and rows you did not know you gained. Here is what changed my day to day handling of the churn dataset.
-
Python Setup for Data Science That Will Not Break: Virtual Environments, Notebooks and Reproducibility (Data Science Series, Part 3)
A working Python setup for data science that survives a laptop change, a colleague, and a rerun six months later. Virtual environments, tool choice, notebook discipline, pinned versions and seeds, with real errors and the fixes.
-
The Data Science Lifecycle End to End, From Business Question to Retired Model (Data Science Series, Part 2)
Modelling was 11 percent of my last churn project. Here is the eight stage lifecycle that accounts for the other 89 percent, including the retirement stage almost nobody plans for.
-
What a Data Scientist Actually Does, vs Analyst, ML Engineer and Researcher (Data Science Series, Part 1)
Four job titles get used interchangeably and they are not the same job. Here is what a data scientist actually does all day, how the role differs from analyst, ML engineer and research scientist, and what the labour numbers say about the path.
Architect’s Toolkit
PJ’s Tools
VMware Cloud Foundation
- VCF Documentation
- VCF 9 Planning & Preparation Workbook
- VCF Bill of Materials (BoM)
- VMware Compatibility Guide
- VMware Interoperability Matrix
- VMware Configuration Maximums
- VMware Ports & Protocols
- VMware Hands-on Labs
- RVTools Download
Nutanix
AI & Cloud-Native Platform
- NVIDIA Build (Model Catalog)
- NVIDIA AI Enterprise Reference Architecture
- NVIDIA NIM Performance Benchmarking
- NVIDIA NGC Catalog
- NeMo Microservices Helm Chart
- Helm Charts Repository
- Hugging Face Models
Architecture & Design
About the Author
Dr Pranay Jha
Dr. Pranay Jha is a Cloud and AI Consultant with 18+ years of experience in hybrid cloud, virtualization, and enterprise infrastructure transformation. He specializes in VMware technologies, multi-cloud strategy, and Generative AI solutions. He holds a PhD in Computer Applications with research focused on Cloud and AI, has published multiple research papers, and has been a VMware vExpert since 2016 and a VMUG Community Leader.
Recommendations & Endorsements
Pranay has very deep technical knowledge and exceptional capabilities. He has strong ability to well-manage complex situations and deliver the customer expectations with high standards of satisfaction. We worked together in one of most complex migration projects and Pranay delivered flawlessly.
Pranay was my VMware Architect on one of our financial accounts at IBM. He was instrumental in documenting and recommending our reference architecture. As a Chief Architect, you are only as good as your team that supports you. Pranay was top notch.
Pranay is proactive, result oriented, responsible and technically sound. Always ready to put all his energy to get the job done. Exceptional troubleshooting skills in Windows Server, virtualization and VMware — a great asset to any company.
Pranay is one of the most talented and dedicated persons I have ever worked with. His vast knowledge from different domains enables him to discuss and help on almost every topic in IT. Needless to say, he is expert of experts in VMware.
DrJha