P-01
Building a Desktop Tool Your Colleagues Can Use with Python's tkinter
Turn a data script into a small tkinter app: ttk widgets and grid, variable-bound state, a responsive main loop and logic kept out of the GUI.
Home
Writing · 15 posts
Writing
Notes on machine learning, NLP, Python, BI and data work, newest first.
15 posts
-
-
P-02
How to Choose a Machine Learning Algorithm for Tabular Data
A short sequence of questions for picking a model on tabular data: define the target, consider how it is used, start with a baseline, compare fairly.
-
P-03
Quantum Computing and Data Science: What May Change, and What Already Needs Attention
A sourced look at quantum hardware limits, optimisation and Monte Carlo research, harvest-now-decrypt-later risk, and the NIST post-quantum standards.
-
P-04
BERT in Practice: Tokens, Embeddings and Fine-Tuning a Text Classifier
What happens to a sentence inside BERT, how to try it in a few lines of Python, and the choices that decide whether fine-tuning works.
-
P-05
Managing Large Datasets Well: Six Habits from Raw Files to Reports
Six practical habits for large, messy datasets: land raw data, validate early, load incrementally, model for queries and keep history.
-
P-06
Free Python and Data Science Courses Worth Your Time
A short, checked list of free Python, pandas, SQL and machine learning courses from Harvard, Kaggle, Google and others, and who each suits.
-
P-07
Why BERT Mattered: From Static Word Vectors to Contextual Language Models
How BERT moved NLP from one vector per word to context-aware representations, what it built on, and why it mattered for financial text.
-
P-08
Reading a Confusion Matrix When the Positive Class Is Rare
A worked example of the confusion matrix and scikit-learn's classification_report on imbalanced data, and how to pick a decision threshold.
-
P-09
Machine Learning Without the Mystique: Supervised, Unsupervised and Overfitting
What machine learning is, how supervised and unsupervised learning differ, and why train/test splits matter, with a small scikit-learn example.
-
P-10
How Readers See Charts: Perception Principles for Reporting Visuals
Why position beats angle and area, when bars need a zero baseline, and how to use colour and titles so report charts are read correctly.
-
P-11
Power BI vs Tableau: A Practical Comparison of Data Engines, Calculations and Analytics
A fair comparison of Power BI and Tableau covering data storage, DAX versus table calculations, built-in analytics and sharing, based on vendor docs.
-
P-12
Web Scraping Techniques: From HTML Parsing to Headless Browsers
A practical guide to scraping in Python: regex vs DOM parsing, headless browsers for JavaScript pages, crawling at scale, and scraping responsibly.
-
P-13
Principal Component Analysis Explained, with a Python Example
What PCA does, how to run it in scikit-learn on correlated data, how to choose the number of components, and the mistakes to avoid.
-
P-14
How Linear Regression Works: Least Squares, Normal Equations and Gradient Descent
A walk-through of ordinary least squares: residuals, the cost function, normal equations, gradient descent and the assumptions behind the fit.
-
P-15
Boosting Algorithms for Phishing Detection: Features, Tuning and Evaluation
How gradient boosting is applied to phishing email detection, and what matters most: features, imbalance-aware metrics, tuning and stacking.
No posts match. Clear the search or pick another topic.