AI related Blogs

What are the differences between micro, macro, and weighted averages and why should you care?

In multı-class classıfıcatıon problems, models often compute a metrıc for each class.

What is an API, and how does it work?

API stands for application programming interface, which is a set of definitions and protocols for building and integrating application software.

Take a moment to understand what a neural network is.

Virtual “neurons” that are joined in layers make form a neural network.

What are Diffusion Models? 🤔

In this article, we will explore the concept of diffusion models and how they can be used to enhance the quality of samples generated by VAEs (and GANs).

Things to Know About the Feature Pyramid Network (FPN) Design

The Feature Pyramid Network (FPN) design is a novel approach to neural architecture search and feature fusing.

Summary: NIN — Network In Network (Image Classification)

This blog provides a quick overview of Network In Network (NIN), a collaboration between the National University of Singapore and the Graduate School for Integrative Sciences and Engineering.

Create a Subclass of Callbacks to Control Learning Rate and Print Training Results for Each Epoch in a Different way

Custom callbacks to control learning rate and print training results for each epoch in a different way

Don’t BLINDLY reuse pre-trained models from a hub.

WeightWatcher is an open-source, diagnostic tool for analyzing Deep Neural Networks (DNN), without needing access to training or even test data.

Summary: Maxout Network (Image Classification)

This blog provides a summary of Maxout Network from the Université de Montréal. The initial author and creator of the Generative Adversarial Network is Ian J. GoodFellow (GAN).

You only have a 16 GB GPU; What’s the best way to fit an L / XL model & fine-tune?

Summary: Dropout — A Simple Way to Prevent Neural Networks from Overfitting (Image Classification)

A Very Famous Regularization Approach to Prevents Co-Adaptation so as to Reduce Overfitting.

How to Create Directory Tree for Projects

First we have to install a python package called directory-tree-generator. Installing process has been shown in the image below.

Summary: Rectified Linear Units Improve Restricted Boltzmann Machines

Rectified Linear Units(ReLU) Improve Restricted Boltzmann Machines, by University of Toronto in 2010 ICML, has around 20000 Citations which outperforms Sigmoid.

Summary: AlexNet, CaffeNet — Winner of ILSVRC 2012 (Image Classification)

AlexNet and CaffeNet are discussed in this article. ILSVRC (ImageNet Large Scale Visual Recognition Competition), an image classification competition, has named AlexNet the winner for 2012.

Summary: LeNet-1, LeNet-4, LeNet-5, Boosted LeNet-4 (Image Classification)

The LeNet, a traditional deep learning convolutional neural network for image classification, has previously been the subject of plenty of literature review.

Keyboard Shortcuts for Google Docs

Here is a collection of keyboard Google Docs shortcuts that will save time while working on any document. I have separated these shortcuts them into different sections — Text Formatting, Paragraph Formatting, Alignment, Text Selection and Bonus

Deep Learning Specialization

Here is a compilation of numerous topics, organized from fundamental to advanced knowledge, that are required for deep learning specialization.

Speed up your scikit-learn modeling by 10–100X with just one line of code

How to open a program with a keyboard shortcut in windows 11

In this article, we will learn how to to open a program with a keyboard shortcut in windows 10.

Most Used Colab Commands

This blog is a collection of some important colab commands I need to use for training ML or DL models.

Tensorflow-GPU Installation

Installing tensorflow-gpu is a hectic task because of matching different versions of dependencies. So I decided to make this simple for everyone.

11 most useful Windows 11 keyboard shortcuts

Today we’re gonna talk about some excellent windows 11 keyboard shortcuts. These shortcuts can save tons of our time which is important because you know time is money.

Publish a Jupyter Notebook as a Blog Post on Github Pages

This blog is about publishing jupyter notebook as a blog post on github pages.

A Paper Rocked The Deep Learning World in December 2015

A paper released in December 2015 sent shockwaves across the deep learning community. This publication is recognized as one of the most important in contemporary deep learning.