Your Accuracy Is a Lie — Here's How to Fix It (The Architect's Guide to Robust Model Validation)

Boosting - scikit-learn Professional Course

Bagging - scikit-learn Professional Course

Decision Trees - scikit-learn Professional Course

Linear Models Regularization - scikit-learn Professional Course

Linear Models Intuitions - scikit-learn Professional Course

Introducing the data accessor of the EstimatorReport

Learn open source machine learning with Skolar

Introducing the search feature

The variable thresholds trick

Introducing the feature_importance accessor of the EstimatorReport

Automatically tuning thresholds

Metadata routing in scikit-learn

Accuracy rarely matters

How to remedy a badly calibrated machine learning model

Introducing the ComparisonReport

When calibration beats metrics

The difference between metrics and scorers

Introducing the CrossValidationReport

Remembering the order of the train_test_split

The ups and downs of stratified cross-validation

Playing with the classification report

Introducing the EstimatorReport

Time for some (extreme) distillation with Thomas van Dongen - founder of the Minish Lab

Dumb models can be very smart

What the official scikit-learn certification looks like

When precision equals recall

Precision, recall and F1-score

Imbalanced-learn: regrets and onwards - with Guillaume Lemaitre, core-maintainer

The optimisers curse

Why the MinHashEncoder is great for boosted trees

How the HashingVectorizer works

You want to be in control of your own Copilot with Ty Dunn - founder of Continue.dev

What it is like to maintain the scikit-learn docs with David Arturo Amor Quiroz, docs maintainer

Sqlite can totally do embeddings now with Alex Garcia, creator of sqlite-vec

How to rethink the notebook with Akshay Agrawal, co-creator of Marimo

Topics vs. embeddings

How the GapEncoder works

PCA as an embedding technique

Feature engineering for overlapping categories

You're always (always!) dealing with many (many!) tables - with Madelon Hulsebos

Data checks for estimators

Improving models via subsets

How Narwhals has many end users ... that never use it directly. - Marco Gorelli

Decayed estimators for timeseries

More flexible models via sample weights

Why ridge regression typically beats linear regression

Understanding how the KernelDensityEstimator works

Pragmatic data science checklists with Peter Bull

Use-cases for inverted PCA

Don't worry too much about missing data

Model safety, that's a pickle! with Adrin Jalali - scikit-learn maintainer

Boosting vs. semi-supervised learning

Benchmarking boosted trees against overfitting

Monotonic, and better, boosting

Moving towards KDearestNeighbors with Leland McInnes - creator of UMAP

Histograms for faster boosting

Getting deeper into trees

Why tree gradients give you a boost

Talk like a DataFrame, run like SQL with Phillip Cloud - core-committer on Ibis