• 👋 Hi, I’m @NamanDhavalDesai
• 👀 I’m interested in Problem Solving.
• 🌱 I’m currently learning Advanced Machine Learning.
• 📫 How to reach me: https://www.linkedin.com/in/naman-desai-a61985204/

This contains 8 repositories.

1. Prism:

Players research and information searching model.

Improvised on a previously existing static database which stores the information of all the different versions of football players in an online game (FIFA Mobile). This already existing static database does not show the user all the parameters at once, hence to compare between the statistics of players is almost impossible. Implement the storage of different accounts of different users making the database comparatively more secure and personalized too. Implement a better way of visualizing and comparing the differences between different versions of players. Include and view personalized and detailed information of each player in the database. Languages to be used: JAVA and SQL.

The contents of this repsitory are as follows:

1. Folder NRX07, this folder conatains the entire java code which includes the GUI (pictures included) within the src sub folder. It is in a java application blueprint.
2. Folder SQL contains the SQL code (needed to be run before the java program) and the data extracted from the internet in .csv format in the Data folder. You can copy and paste this data in the structure created by the sql code if you do not want to extract data from the internet (sraping takes 3+ hours).
3. Folder Documents contains the working paper, EER diagram, flowchart diagram and a powerpoint presentation on the working of the application.
4. Folder Test contains the java code used to test the mysql connector and the working of the jsoup package.

2. HTML5andCSS3:

A repository to store all HTML5 and CSS3 codes.

This repository consists of 2 folders:
Introduction: This folder contains a html file, a css file and a png file which is inserted in the html file.
Omnifood: This folder conatins mutiple sub folders and files.

1. Resources: This folder contains mutiple sub folders and files.
   i) CSS: This folder conatins the css code and a sub folder to store all the pictires used in the CSS program.
   ii) Img: This folder contains all the images added into the html file directly.
   iii) js: This folder contains all the javascript programs (scripts).
   iv) Favicons: This folder conatains all the favicons.
   v) PHP: This folder contains all the php codes used to make the form of the website usable.
2. Vendors: This folder contains mutiple sub folders and files.
   i) CSS: This folder conatins 3 CSS files, one to set up a grid, one to normalize the webpage on all devices, one to get the animation features and one to insert all the icons used in the webpage.
   ii) Fonts: This folder conatins the fonts used in the webpage which are imported into the html file.
   iii) js: This folder contains all the javascript programs (scripts) taken from other vendors.
3. HTML file: The index.html file produces a stunning web page for an imaginary company omnifood.

Omnifood is a company which provides people with healthy food and its delivery so users have a healthier and a better way to have their meals as per their deitary requirements. The webpage includes the companies description, their most popular meals, some pictures of their dishes, its working, locations avaible for delivery, customer's review and finally the packages and the form to sign up at the bottom. Included animations and multiple beautifications to make it look like a modern webpage which is a must to capture and involve more users.

3. R-Programming:

Notes and programs from the Johns Hopkins Data Science Specialization. This repository contains files, such as:

1. Folder Nodes conatins files Codes 1 to Codes 8 are all complied into one file named Notes (notes of every command shown in the 9 courses of the specialization), and Introductory folder and a notes folder for R-Markdown documents.
2. Folder FIFA contains files (different versions of a program) to web scrape the entire fifa mobile database from fifarenderz.com and a excel file containing the output of the program.
3. Practice programs folder conatins 2 files: a) Ppattern file contains a code that creates a pattern to solve for n numbers where each intger i in the range from 1 to n is an input and o is the output. The condition (equation being), o to the power of o to the power of o (i number of times) gives i. b) Arrangement of data is an R code to jumble and then arrange a dataset randomly. This algorithm could be used arrange values quicker.
4. Folder library has 9 sub folders wherein each folder has the assignment(s) solution(s) for every course.
5. The study material folder conatains study material for R programming (books and text doccuments).

4. Games:

This repository is to store the games programmed in java.
This repository contains one file named Games.java.
(My first hands on project).
(Created in 2015 using BlueJ at the age of 14).
This repository conatains one program (.java) that consists of 9 games:

1. Hangman.
2. X and 0.
3. Jumble words.
4. Cricket.
5. Chess
6. Guess the code.
7. Battleships.
8. Connect 4.
9. Othello.

All in one program. Used different functions (one or more for each game) within the same class and called a particular function (depending on teh choice made by the user) in the main function using conditional branching. This programs contains over 34 functions and 8500 lines of program. The program does not contain any GUI and only the backend code and basic printing using the print statements to view the output and play the game.

5. Java-Programming:

A repository to store all java programs. This repository contains 3 folders.

1. Applicable: To store all the programs that can have real world applications.
2. OutOfTextbook: To store the programs which are not present in the textbook.
3. Textbook: To store the programs whose questions are present in the textbook. (Textbook - Java Applications by Shailesh Manjrekar).

6. Python-Programming:

A repository to store all python programs.

1. Learning Programs: To store all python programs which were used for learing the basics of Python.
2. OCR:
   i) Dowload poppler and tesseract from the sites below or from the attached files:
       a) from https://poppler.freedesktop.org/
       b) from https://github.com/UB-Mannheim/tesseract/wiki
   ii) Ensure that the files are in the location mentioned below:
       a) in C:/Program Files/poppler-0.68.0/
       b) in C:/Program Files/Tesseract-OCR/
   iii) Filing the input and output file names with their path in pdftoimagetopdf("","") which is the last line of code in the datamodification.py file.
   iv) Run the .py files in the order mentioned below:
       a) datamodification.py
       b) postdatamodification.py
   v) The code takes an input non searchable pdf file and converts it into a searchable pdf file.

• Note - Remove the data from file geckodriver.log before the execution.
• Note - The other files in this folder perform tasks as thier file name suggests.

7. NamanDhavalDesai.github.io:

A repository to host the website (Omnifood).
Check out the websites description in the HTML5andCSS3 repository's description under Omnifood.

8. FR and QR Application

Face Recognition and QR Code Application.

Software part of a solution for creating a system within professional institutions wherein each individual who is a part of that institution could automate their work at some level by using ID cards which contain a QR code, this ID card coupled with the facial recognition technology could provide a unique identity to individuals.

The contents of this repsitory are as follows:

1. Folder Source, this folder conatains the entire python code:
   i) Subfolder pycache contains the cache data, to improve the speed of the application.
   ii) Subfolder cascades contains the cascade files for the facial recognizer.
   iii) Subfolder gui_images contains the images used for the graphical user interface (GUI).
   iv) Subfolders, qrcode, images, downloads and person contain 2 files namely, Delete_1 and Delete_2, these files need to removed before execution. These folders are used by the application for various tasks, see the documentation for more details.
   v) File app.py contains the entire python program.
   vi) File labels.pickle is used to label the categorized images.
   vii) File trainner.yml contains the data required to ren the application.
2. Folder Documents contains the working paper, logbook, EER diagram, flowchart diagram and a powerpoint presentation on the working of the application.

9. ShopKart

Shopkart is an Application where you can create and log into an account and select and add items to the cart and get the grand total of the cost of the products in the cart.

10. Chest-X-Ray-Classifier

The following program uses a CNN model to classify if a chest x-ray image is healthy or not. This function will be able to differentiate multiclass categories where label names are present: normal, virus, and bacteria. To solve this problem CNN modelling (Convolutional Neural Network) is used, due to its excellent image-splitting ability.

Dataset used:

• The dataset used in this project could be found and installed from the Kaggle link mentioned below, https://www.kaggle.com/paultimothymooney/chest-xray-pneumonia.

• The entirety of the dataset is approximately 1.3 GB consisting of three sub folders. The constituents and description of these sub folder is mentioned below:
  This is the main folder which contains 3 main sub folders named test, train and val (validation).

  • The training dataset is generally the largest dataset as the model needs to be created from the data it provides.
  • The testing dataset is generally smaller to the training dataset but larger than the validation dataset as this generally provides the accuracy of the model by testing the model.
  • The validation dataset generally is the smallest dataset and provides the validation to the model.

• The training dataset comprises Normal or healthy lungs (1341 images) with the existence of a total of 766 people’s x-rays and Pneumonia or people having unhealthy lungs (3875 images), among which 1345 and 2530 samples for virus and bacteria respectively. There exists singular or multiple x-rays images of individual people. In the second case, there are in total 1945 people’s x-rays. Whereas, the test dataset comprises two output classes such as healthy lungs (234 images) and people having unhealthy lungs (390 images).

• The data is read by the program and then depending on the images’ characteristics the model is made which can now classify if a pair of lungs are healthy or unhealthy and if they are unhealthy then does the person have virus or a bacterial infection. The contents of the validation dataset are used to validate the classifier after it has been made. It could be considered as a part of the training dataset. It usually is not a large dataset. Usually out of a hundred percent 70 percent of the data belongs to the training dataset while the remaining 30 percent usually belongs to the validation set. This sub folder contains two more sub-sub folders such as Normal (8 images) and Pneumonia (8 images). All the images in this sub-sub folder are x-rays of people having healthy and unhealthy lungs (bacteria or virus). It prevents overfitting of the classifier and helps to tune the parameters of the model.

Results:

• Accuracy, Misclassification rate, Precision, Sensitivity (Recall), Specificity and F1-score can be calculated using the formulae mentioned bellow:
  • Accuracy: (TP + TN) / (TP + TN + FP + FN):
  • Misclassification rate: (FP + FN) / (TP + TN + FP + FN) or (1 - Accuracy):
  • Precision: (TP) / (TP + FP):
  • Sensitivity: (TP) / (TP + FN):
  • Specificity: (TN) / (TN + FP):
  • F1-score: (2 x Precision x Sensitivity) / (Precision + Sensitivity) or (2 x TP) / ((2 x TP) + FP + FN):