Machine-learning

Developments in AI are changing fast, RAG and CAG are the latest developments that need to be understood. Each has their benefits and differ in where they can be applied.

LLMs are being enhanced by RAG and AI Agents, which are improving accuracy and limiting hallucinations.

After initial overfitting on small datasets, models can eventually achieve exceptional generalisation through extended training and optimisation.

Maintaining a good in-production model requires understanding the understanding of model and concept drift.

The typical data science E2E workflow and pipeline.

SHAP and Synergy matrix will be an important part of a machine learning pipeline.

Hyperparameter Tuning

Performing a Statistic tests

Second part in understanding ML Models.

First part in understanding ML models.

This blog covers the wisdom of the crowds effect and how we can apply it when developing machine learning models.

Understanding the Decision Tree training process is key to being able to select whether your data requires the model to be trained using Entropy or Gini impurity Criterions.