Machine Learning Engineer | Avid Reader | Movie Buff

In this blog post, we’ll discuss loss functions, parameter θ and the different types of loss function. I’ve learnt a lot while researching about this topic and hope you’ll feel the same. Without further a due, let’s starts off with loss function.

In simple terms, the objective of a loss function is *to find the difference between or deviation between the actual ground truth of the value and an estimated approximation of the same value.*

In this blog post, we’ll discuss about supervised learning and the class of problems that comes under its umbrella. Before getting into supervised learning, I would highly recommend going through machine learning jargon like what is dataset, target, predictor, model etc. from the previous blog posts.

In Machine Learning, if a label or target is available for an observation then such an observation is called **Supervised Observation**. From technical standpoint, given a set of data points X’s associated to set of labels or outcomes Y’s, we try to build a model that learns to predict *y* from *x.*

This blog post is the continuation of second blog post on concepts of machine learning. Here, we’ll learn, how to build a dataset for the model and basic evaluation metrics to follow after model is trained, and also factors to keep in mind after building machine learning model.

Before applying ML Algorithm, we should perform the analysis our dataset to get the gist of all the variables, its missing values, outliers if any, normalization etc. Once we make our data clean and consumable, we should proceed with developing dataset. This whole process is called as **preprocessing** of data.

Once preprocessing…

This is the second blog post of the machine learning series, here, we’ll discuss the most common concepts of machine learning like features types, missing value, regularization etc

When designing an algorithm for learning features, our goal is to separate the **factors of variation **📊 that explain the observed data.

Features are the components in our dataset which helps in building ML algorithm. There are different types of features based on there value like **Numerical, Ordinal** etc.

Nominal —The variable with nominal values doesn’t posses any order naturally. …

Machine learning is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed. Machine learning focuses on the development of computer programs that can access data and use it learn for themselves.

Machine Learning is essentially a form of applied statistics with increased emphasis on the use of computers to statistically estimate complicated functions and a decreased emphasis on providing confidence intervals around these functions.

“A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if…

In this blog post, we’ll learn about algorithms used on graph or tree data structure. Both the algorithms DFS and BFS are used for searching through the vertices of a tree or graph data structure as efficiently as possible.

It is a recursive algorithm, searching all the vertices in the graph or tree. It puts each vertex into two categories either **visited** or **not visited**. The purpose of the algorithm is to mark each vertex while avoiding cycles. DFS follows pre-order traversal, first visits the root, then left subtree and then right subtree. …

Consider a professional social networking site like Linkedin, where every time we connect with a different person, a link is formed between us and the other person. This small interaction is what is called as **Graph Data Structure**.

Here,** we and the other person** are the **vertex** in the graph and the **connect** request turns into an **edge** between you and the other person. Instead of other person, we can connect with other vertices like **group page, attend an event (temporary node)** etc. Thus, the whole of **Linkedin** is one big collection of vertices and edges.

Graph data structure is…

In this blog post, we’ll discuss about the binary search tree and its implementation with python, as the name suggests, the BST is widely used for **performing search operation** **over the sorted list of numbers.**

Each node in BST has **at most two children** and it performs search with the time complexity of **O(log(n))** where **n** is the **total number of nodes** in the BST and also can be seen from the **height of the tree**.

To generate binary search tree from a sorted list of numbers, we must follow a set of properties:

- All the left nodes are lesser…

In this blog post, we’ll discuss about binary tree and its variants along with Python implementation. In previous post, we learnt about trees and its properties, which is relevant to binary tree as well.

**Binary Tree** is a tree where **each parent has at most two nodes.** There are different binary tree based on the number of leaf nodes or internal child nodes.

In this blog post, we’ll learn about **tree** data structure, related terms and how to traverse through trees.

**Tree is a non-linear hierarchical data structure consisting of vertices (nodes) and edges.** Trees lies on other spectrum of data structure when compared with data structures like stack, queue, linked list etc. In stack, queue and Linked list the data structure is linear and the elements are added in sequence.

Any operation in linear data structure has increasing time complexity as the size of the data increases, which makes it less suitable for the real world problems. **…**