atiqahammed/Cats-and-Dogs-Breed-Detection

Stars
1
Language
Python
Created about 5 years ago
Updated about 5 years ago

atiqahammed/Cats-and-Dogs-Breed-Detection

atiqahammed

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

We investigate the issue of categorizing fine grained objects to determine the animal breed from an image. In this project we used a data set of cats and dogs, 12 cat breeds and 120 dog breeds. The visual problem is very challenging as these animals, particularly cats, are very deformable and there can be quite subtle dif-ferences between the breeds. We have done a number of tasks here: first we compare two classification approaches:a hierarchical one, in which a pet is first assigned to the cat or dog family and then to a breed, and a flat one, in which the breed is obtained directly. When applied to the task of discrimi-nating the 132 different breeds of pets, the models obtain an average accuracy of about 59%, a very encouraging result considering the difficulty of the problem.

Control-Flow-Graph

In computer science, a control flow graph (CFG) is the graphical representation of control flow or computation during the execution of programs or applications. In this project I am reading java code and making control flow graph for all function from the given java code.

Bengali-Braille-to-Text-Translator

Bengali braille to text translator is a desktop application to translate Bengali braille code into corresponding Bengali text. More information about it is on the website line given here.

B-plus-tree-dictionary-searching

to implement dynamic multilevel indexing, B-tree and B+ tree are generally employed. The drawback of B-tree used for indexing. To implement a simple dictionary I used b+ tree algorithm for indexing.

6th-semester

This repository is created managing the files of sixth semester. It contains all books, slides, necessary document of all courses of sixth semester. Besides all short assignments, documents of those assignments and their presentation slides are stored here.

Point-of-Sale

Point of sale is a Desktop application developed on java with database connection. In this system a user can create an user account an purchase product. The admin can add, update products. It was a group assignment in SE-306 course.

Mood-Metrics

In the field of software development, de- spite the central role played by software de- velopment in the production and implementa- tion of IT managers the focus is progressively on process improvement. The demand has prompted a number of new and/or enhanced software development approaches with perhaps the most prominent being object-orientation (00).For testing the OO based softwares, a number of metrics are suggested. MOOD and CK metrics are well-known metrics used to measure some Object-Oriented programs char- acteristics. A system based on the MOOD and CK was developed for evaluating and grading C programs in this research. Ten C based sys- tems were chosen for this and the metrics were used for evaluating the programs.

Routine-de-Classe

This is an android application for maintaining class routine. This application will provide it's user to create any class routine, maintain the routine, invite other users to be added to the routine and join to a specific groups to have their class schedule.

BMKS-web

this is a website design for Bohail Manok Kalyan Songstha(BMKS). Its a non academic project for supporting BMKS. The main perpose for this project is to provide a website and increase my experience.

File-System

File system is a desktop app for managing files. In this system user can create an user account, add new files, copy, open delete and share existing files.

Draw

This is a project in which I am trying to implement a painter for android stdio. Initially I will draw a single line in a layout, gradually circle, oval, square, rectangle will be added.

FourConnect

This a android based OS game. Where user will play with intelligent agent. first the user will perform his/her turn, the computer will perform followed by. When ever the agent or user can arrange their colors in number 4 in row or column or diagonal then the game will be finished and last performer will win.

Books

I like to collect and read books in my leisure time. Although I am fond of reading book in hard copy, but sometimes I collect some softcopy of them. For making those books available for all I create this rep. I also have a online library named Atiq's Library that is created by using google site. https://sites.google.com/iit.du.ac.bd/atiqslibrary

Problem-Solving

In this repository I am keeping all the solution that I tried to solve problems. I try to solve online problem in various sites. But I prefer Codeforces most. Here is my Codeforces profile link http://codeforces.com/profile/_Scare_Crow_ .

puppeteer-vs-selenium

An automated testing tool is capable of replaying pre-recorded and predefined activi- ties, comparing the outcomes with the antic- ipated conduct and reporting to a test en- gineer the achievement or failure of these manual trials.Once automated tests are pro- duced, they can be readily repeated and ex- panded with manual testing to execute duties that are impossible. Selenium and Puppeteer both are web testing tools. This work focuses on an empirical study on both of the tools features and experimented on them.

Feature-Selection

In this age of information, data is increasing in both number of instances and features. This huge amount of features cause serious problem with respect to learning algorithms and accuracy. Because, in high dimensional data, the number of irrelevant and redundant data might be also huge. So, machine learning algorithms face additional complexity. Unfortunately, as the amount of machine readable information increases, the ability to understand and make use of it does not keep pace with it’s growth. Machine learning provides tools by which large quantities of data can be automatically analyzed. Fundamental to machine learning is feature selection. Feature selection refers to the fact, selecting attributes or variables. In other words, feature selection is an algorithm for finding a optimum subset of feature having minimum classification error or maximum accuracy.