We would be seeing different kinds of Convolutional Neural Networks and how they differ from each other in this article. These are some groundbreaking CNN architectures that were proposed to achieve a better accuracy and to reduce the computational cost .

1. LeNet-5

This is also known as the Classic Neural Network that was designed by Yann LeCun, Leon Bottou, Yosuha Bengio and Patrick Haffner for handwritten and machine-printed character recognition in 1990’s which they called LeNet-5. The architecture was designed to identify handwritten digits in the MNIST data-set. The architecture is pretty straightforward and simple to understand. The input images were…


In this article, I would be covering how we can deploy a custom deep learning container algorithm on Amazon Sagemaker. Sagemaker provides 2 options wherein the first option is to use built-in algorithms that Sagemaker offers that includes KNN, Xgboost, Linear Learner, etc. while the other option is to use your custom docker container from ECR(Elastic Container Registry). In this article, we would see how we can deploy our custom docker container and host it on Sagemaker. For data purposes, I have used the famous Titanic data set from Kaggle. …


There are several drawbacks of using a sliding window for object localization such as selecting appropriate kernel size, stride etc. which leads to high computational cost.

Introduction

YOLO (You only look once) was introduced by Joseph Redmon, Santosh Divvala, Ross Girshick and Ali Farhadi in 2015. The best things about YOLO is that it achieves high accuracy while also being able to run in real time. The approach of YOLO is to frame object detection as a regression problem and passing it to fully a connected neural network . The system divides the input image into an S×S grid and each…


I encountered an issue, where i wanted to attach two commands/functions to a Tkinter button. One function was designed to train a machine learning model and other was a Determinate Progress Bar. The whole intent was to display a progress bar on the UI that shows model training which is running in the back-end.

Using Lambda to bind multiple functions on a button

The image illustrates a button which is attached to two independent functions

The use of lambda provides the functionality to attach many functions to the button but they are executed sequentially. Hence, the progress bar would be completed before the model training has even started or vice-versa.

Using Multi-threading to bind functions

To encounter the sequential execution of the functions, they can be…


Introduction:

Convolutional Neural Networks are deep neural networks that were designed typically to handle image datasets. When we are dealing with pixels, generalization becomes extremely difficult if we feed all the pixels directly to a fully connected network after flattening. Imagine, if we have a data-set with images each of size (360*360) in R.G.B, then we would have 388800 pixels(input vector) of a single image to feed to a multi layer Perceptron. Hence, there is need to highlight the features, get rid of noisy pixels and perform dimensionality Reduction.

Dealing with Images directly with Ordinary Neural Nets

For a better understanding, lets begin training our model without convolutions. Here…

Bhavesh Singh Bisht

I am passionate about data science and have a profound history of working in an AI/ML firm

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store