Does feature Engineering help DL Models — The classic ML model suffered when it is challenged by unstructured data. A special case in point was the ImageNet Challenge, where initially the state-of-the-art would use some intelligent hand-crafted features and then a strong classifier like SVM. Later, these were surpassed by the CNN models. Now this derivation of…

the two animals to tame Like human beings, machine learning models make errors. We need to work to minimize the error. But before that, we need to be more familiar with the error. The following figure summarizes the different components of error.

A step by step tutorial to GAN — A generative model can potentially do magic, if trained properly it may write poetry, generate music, draw images like an expert. The objective of GAN is to generate synthetic samples after which are very realistic. GAN models learn the trick in an adversarial setting. There are two multi-layer neural networks…

Handling some of the Design issues of LSTM — Solar energy is one of the most important constituents of alternative sources of clean and renewable energy. Forecasting of Solar Energy Generation is critical for downstream application and integration with the conventional power grids. Rather than measuring the photo-voltaic output of the solar cells, often the radiation received from the…

Achieving class balance with few lines of python codes — Class Imbalance is a quite frequently occurring problem manifested in fraud detection, intrusion detection, Suspicious activity detection to name a few. In the context of binary classification, the less frequently occurring class is called the minority class, and the more frequently occurring class is called the majority class. You can…

One of the superstars of the ensemble family — Here is an attempt to unpack one of the most important and little difficult ensemble algorithm named gradient boosting. Ensemble, Bagging, and Boosing: Ensemble learning happens when you use multiple models. …

The lesser the features, the easier to interpret the model — Let’s approach this problem of feature selection using Chi-Square a question and answer style. If you are a video guy, you may check out our youtube lecture on the same. Question 1: What is a feature? For any ML or DL problem, the data is arranged in rows and…

Clearing the fog of confusion around semi-supervised learning — We all have come across semi-supervised learning as a type of machine learning problem. But it is a concept not understood really well. It’s best to understand this by getting our hands dirty and precisely that’s what we are bringing on. The use-case is simple, we have some data that…

Class Imbalance can put our algorithm off balance — It was the late 1990s when Niesh V Chawla (The main brain behind SMOTE), then a graduate student at the University of South Florida was working on a binary classification problem. He was dealing with mammography images and his task was to build a classifier that will take a pixel…