.

# Sandipan Dey's Blog (48)

### Neural Translation – Machine Translation with Neural Nets with Keras / Python

In this blog, we shall discuss about how to build a neural network to translate from English to German. This problem appeared as the Capstone project for the coursera course “Tensorflow 2: Customising your model“, a part of the specialization “Tensorflow2 for Deep Learning“, by the Imperial College, London. The…

Continue

Added by Sandipan Dey on May 5, 2021 at 3:30am — No Comments

### NLP with Bangla: Generating Text and Sentiment Analysis

In this blog post, I shall discuss on a few NLP techniques with Bangla language, demonstrate how to train a word2vec model with Bangla wiki corpus with tensorflow, and explore how to visualize the semantic similarity between words using t-SNE. Once done, the article will focus on how to train a character / word LSTM on selected Tagore’s…

Continue

Added by Sandipan Dey on January 27, 2021 at 1:00am — No Comments

### Solving a few AI problems with Python: Part 1

In this blog we shall discuss about a few problems in artificial intelligence and their python implementations. The problems discussed here appeared as programming assignments in the edX course CS50’s Introduction to Artificial Intelligence with Python…

Continue

Added by Sandipan Dey on November 6, 2020 at 1:30am — No Comments

## Heat / Diffusion Equation

The following animation shows how the temperature changes on the bar with time (considering only the first 100 terms for the Fourier series for the square wave).…

Continue

Added by Sandipan Dey on July 12, 2020 at 10:00am — No Comments

### Fourier Series and Differential Equations with some applications in R and Python (Part 2)

This is the 2nd part of the article on a few applications of Fourier Series in solving differential equations. All the problems are taken from the edx Course: MITx - 18.03Fx: Differential Equations Fourier Series and Partial Differential Equations. The article…

Continue

Added by Sandipan Dey on July 4, 2020 at 5:00pm — No Comments

### Fourier Series and Differential Equations with some applications in R (Part 1)

In this article, a few applications of Fourier Series in solving differential equations will be described. All the problems are taken from the edx Course: MITx - 18.03Fx: Differential Equations Fourier Series and Partial Differential Equations. The article will be posted in two parts (two separate blongs)

First a basic introduction to the Fourier series will be given and then we shall see how to solve the following ODEs / PDEs using Fourier series:

1. Find…
Continue

Added by Sandipan Dey on July 4, 2020 at 5:00pm — 3 Comments

### Solving Some Image Processing Problems with Python libraries - Part 3

In this article a few more popular image processing problems along with their solutions are going to be discussed. Python image processing libraries are going to be used to solve these problems.

## Histogram Matching with color images

As described here, here is the algorithm:

1. The cumulative histogram is computed for each image dataset, see the figure below.
2. For any…
Continue

Added by Sandipan Dey on August 16, 2018 at 1:30pm — No Comments

### Solving Some Image Processing Problems with Python libraries - Part 2

In this article a few more popular image processing problems along with their solutions are going to be discussed. Python image processing libraries are going to be used to solve these problems.

## Some Applications of DFT

### 0. Fourier Transform of a Gaussian Kernel is another Gaussian Kernel

Also, the spread in the frequency domain  inversely proportional to the spread in the spatial domain. Here is the proof:…

Continue

Added by Sandipan Dey on August 16, 2018 at 1:00pm — No Comments

### Solving Some Image Processing Problems with Python libraries - Part 1

In this article a few popular image processing problems along with their solutions are going to be discussed. Python image processing libraries are going to be used to solve these problems.

## Image Transformations and Warping

### 0. Display RGB image color channels in 3D

1. A gray-scale image can be thought of a 2-D function f(x,y) of the pixel…
Continue

Added by Sandipan Dey on August 16, 2018 at 1:00pm — No Comments

# 1. Back-propagation

This problem also appeared as an assignment problem in the coursera online course Mathematics for Machine Learning: Multivariate Calculus. The description of the problem is taken from the assignment itself.

In this assignment, we shall train a neural network to draw a curve. The curve takes one input variable, the…

Continue

Added by Sandipan Dey on May 31, 2018 at 10:00pm — No Comments

### Implementing PEGASOS: Primal Estimated sub-GrAdient SOlver for SVM, Logistic Regression and Application in Sentiment Classification (in Python)

Although a support vector machine model (binary classifier) is more commonly built by solving a quadratic programming problem in the dual space,  it can be built fast by solving the primal optimization problem also. In this article a Support Vector Machine implementation is going to be described by solving the primal optimization…

Continue

Added by Sandipan Dey on April 28, 2018 at 3:30pm — No Comments

### Implementing a Soft-Margin Kernelized Support Vector Machine Binary Classifier with Quadratic Programming in R and Python

In this article, couple of implementations of the support vector machine binary classifier with quadratic programming libraries (in R and python respectively) and application on a few datasets are going to be discussed.

The next figure describes the basics of Soft-Margin SVM (without kernels).

SVM in a nutshell

• Given a (training) dataset consisting of positive and negative class instances.
• Objective is to find…
Continue

Added by Sandipan Dey on April 23, 2018 at 9:30am — No Comments

### Some Reinforcement Learning: The Greedy and Explore-Exploit Algorithms for the Multi-Armed Bandit Framework in Python

In this article the multi-armed bandit framework problem and a few algorithms to solve the problem is going to be discussed. This problem appeared as a lab assignment in the edX course DAT257x: Reinforcement Learning Explained by Microsoft. The problem description is taken from the assignment itself.

## The Problem Statement and Some Theory

Given a set of  actions with some unknown reward distributions, …

Continue

Added by Sandipan Dey on April 4, 2018 at 9:30am — No Comments

### Learning Distributed Word Representations with Neural Network: an implementation from scratch in Octave

In this article, the problem of learning word representations with neural network from scratch is going to be described. This problem appeared as an assignment in the Coursera course Neural Networks for Machine Learning, taught by  Prof.  Geoffrey Hinton from the University of Toronto in 2012.

## Problem Statement

In this article…

Continue

Added by Sandipan Dey on March 18, 2018 at 12:00am — No Comments

### Autonomous Driving – Car detection with YOLO Model with Keras in Python

In this article, object detection using the very powerful YOLO model will be described, particularly in the context of car detection for autonomous driving. This problem appeared as an assignment in the coursera course Convolution Networks which is a part of the Deep Learning Specialization (taught by Prof. Andrew Ng.,  from Stanford…

Continue

Added by Sandipan Dey on March 11, 2018 at 11:30pm — 3 Comments

### Implementing Lucas-Kanade Optical Flow algorithm in Python

In this article an implementation of the Lucas-Kanade optical flow algorithm is going to be described. This problem appeared as an assignment in  a computer vision course from UCSD. The inputs will be sequences of images (subsequent frames from a video) and the algorithm will output an optical flow field (u, v) and trace the motion of the moving objects. The problem description is taken from the assignment itself.

## Problem…

Continue

Added by Sandipan Dey on February 28, 2018 at 11:30am — No Comments

### Graph-Based Image Segmentation in Python

In this article, an implementation of an efficient graph-based image segmentation technique will be described, this algorithm was proposed by Felzenszwalb et. al. from MIT.  The slides on this paper can be found from Stanford Vision Lab.. The algorithm is closely related…

Continue

Added by Sandipan Dey on February 28, 2018 at 11:30am — No Comments

### Image Colorization Using Optimization in Python

This article is inspired by this SIGGRAPH paper by Levin et. al, for which they took this patent , the paper was referred to in the course CS1114 from Cornell.  This method is also discussed in the coursera online image processing course by NorthWestern University. Some part of the problem description is taken from the paper itself. Also, one can refer to the implementation provided by the authors in matlab, the following …

Continue

Added by Sandipan Dey on February 13, 2018 at 1:00pm — No Comments

### Interactive Image Segmentation with Graph-Cut in Python

In this article, interactive image segmentation with graph-cut is going to be discussed. and it will be used to segment the source object from the background in an image. This segmentation technique was proposed by Boycov and Jolli in this paper

## Problem Statement: Interactive graph-cut segmentation

Let’s implement…

Continue

Added by Sandipan Dey on February 13, 2018 at 10:30am — No Comments

### Recursive Graphics, Bi/Tri-linear Interpolation, Anti-aliasing and Image Transformation in Python

The following problem appeared in an assignment in the Princeton course COS 126 . The problem description is taken from the course itself.

## Recursive Graphics

Write a program that plots a Sierpinski triangle, as illustrated below. Then develop a program that plots a recursive patterns of your own design.…

Continue

Added by Sandipan Dey on January 24, 2018 at 10:00am — No Comments

2021

2020

2018

2017