Why RAG Chatbots Struggle in Production
โOur RAG chatbot worked perfectly in the POC.But once we scaled to 50,000 documentsโฆ accuracy dropped to 60%.โ If youโve worked with enterprise RAG systems, youโve probably heard this story.…
What is Stacking of Models in Machine Learning?
The last Ensemble method we will discuss in this series is called stacking (short for stacked generalization). It is based on a simple idea: instead of using trivial functions (such…
Ensemble Learning: A Comprehensive Guide to AdaBoost and Gradient Boosting
Introduction: In the realm of machine learning, ensemble learning techniques such as AdaBoost and Gradient Boosting have revolutionized the way we approach classification and regression tasks. These powerful algorithms harness…
Introduction to Dimensionality Reduction
The text discusses the curse of dimensionality in machine learning, highlighting challenges in high-dimensional spaces. It suggests reducing features to improve training efficiency and visualization, while addressing potential information loss…
Sending Data in Unstructured File Form
Unstructured data files consist of a series of bits. The file doesnโt separate the bits from each other in any way. You canโt simply look into the file and see…
Random Forests | Machine Learning from Scratch
As we have discussed, a Random Forest is an ensemble of Decision Trees, generally trained via the bagging method (or sometimes pasting), typically with max_samples set to the size of…
Accessing Data in Structured Flat-File Form
In many cases, the data you need to work with wonโt appear within a library, such as the toy datasets in the Scikit-learn library. Real-world data usually appears in a…
Gini Impurity or Entropy? How to decide the root node in decision tree?
By default, the Gini impurity measure is used, but you can select the entropy impurity measure instead by setting the criterion hyperparameter to “entropy”. The concept of entropy originated in…
Decision Trees | Machine Learning from Scratch
Like SVMs, Decision Trees are versatile Machine Learning algorithms that can perform both classification and regression tasks, and even multioutput tasks. They are very powerful algorithms, capable of fitting complex…
How can A linear model learn non-linear/discrete patterns?
Introduction During model development, one of the techniques that many donโt experiment with is feature discretization. The core idea is to transform a continuous feature into discrete features, mostly one-hot…
Support Vector Machines (SVM) Algorithms
A Support Vector Machine (SVM) is a very powerful and versatile Machine Learning model, capable of performing linear or nonlinear classification, regression, and even outlier detection. It is one of…
What is early stopping? | Machine Learning from Scratch
Machine learning models, particularly those trained iteratively using algorithms like Gradient Descent, face the risk of overfitting the training data. One powerful and elegant solution to this challenge is known…
Information Gain in Machine Learning
Information Gain (IG) is critical in machine learning and decision tree algorithms, particularly in data classification and ๐๐๐๐ญ๐ฎ๐ซ๐ ๐ฌ๐๐ฅ๐๐๐ญ๐ข๐จ๐ง. Information Gain Information Gain is a concept used in the field…
What is Lasso Regression? | Machine Learning from Scratch
Least Absolute Shrinkage and Selection Operator Regression (simply called Lasso Regression) is another regularized version of Linear Regression: just like Ridge Regression, it adds a regularization term to the cost…
Regularized Linear Models(Ridge Regression) | Machine Learning from Scratch
As we saw in previous posts, a good way to reduce overfitting is to regularize the model (i.e., to constrain it): the fewer degrees of freedom it has, the harder…
Learning Curves | Machine Learning from Scratch
Till now, We have read about Gradient Descent,Min-Batch Gradient Descent,Stochastic Gradient Descent and other type of Gradient Descents and Polynomial Regression. In this post we will learn about Learning Curves…