This project is a Sample of professional work. Here I show apps which are live and sample projects developed to showcase my skills.
Project is created with:
- Python
- OpenCV
- Jupyter Notebook
- Kotlin
- Java
- C#
- Unity3D
Image Classifier with Convolutional Neural Networks - click here for more
A PyTorch Image classifier based on Convolutional Neural Network and Supervised Learning.
- Only one file needed - ai_image_classifier.py -
- you can use it to train a model in three lines of code.
- The model uses a ResNet backbone.
See example implementing this classifier to distinguish between oranges and pears:
Deep RL Single Agent Bananas collector with Deep Q networks - Link to repo
A training an agent using Deep Q networks to navigate (and collect bananas!) in a large, square world.
A reward of +1 is provided for collecting a yellow banana, and a reward of -1 is provided for collecting a blue banana. Thus, the goal of your agent is to collect as many yellow bananas as possible while avoiding blue bananas.
The task is episodic, and in order to solve the environment, your agent must get an average score of +13 over 100 consecutive episodes.
This project has been solved on a Jupyter notebook
Deep RL Single Agent Double-joint arm - Link to repo
Reinforcement Learning Agent for Joint Arm using Deep Deterministic Policy Gradient (DDPG)
- Training average rewards:
Deep RL Multi-agent tennis collaboration/competition - Link to repo
-
This repository contains the implementation of a Multi-Agent collaboration using Multi-Agent Deep Deterministic Policy Gradient (MADDPG).
-
This has been part of Udacity's Deep Reinforcement Learning Nanodegree program
Mouse control with hand gestures - Link to repo
Finger detection and tracking with OpenCV. Detects and counts fingers. With five fingers it controls the cursor.
Single Pop - A Unity3D Game for Android - Link to repository
Developed in Unity3D, single pop is an android minimalistic game
- This repository contains some of the key scripts used in single pop.
- All scripts were developed on C#.
Scraping Flight data from Kayak - Link to repo
- Run example.py for testing code. It will fetch flight info from Milan to Madrid for next day.
- Run scraper_kayak.py for adding your input. You can choose the origin and destination of the scraping.
- chromedriver.exe must be in the same folder as the python script.
- chrome browser must be installed.
- When running the script, Chrome will open and will look like this:
Web scraping BOOKING.COM - Link to repo
Web scraping hotel rooms data from booking.com with Python.
- Run example.py for testing code. It will fetch hotel data from London for next day for duration of 1 night and 2 adults.
- This script uses selectorlib. Therefore Chromedriver or selenium are not required.
- After the script runs a CSV file will be created named "example_output.csv" in the same folder where the script runs. The csv looks like this on excel:
Android apps live on GooglePlay
Android App which provide train journey data for Belgium
-
Search train routes and connections across, from and to Belgium.
-
Fetch train times for all Trains stations and companies within Belgium. Use this app for any rail route query. Find the fastest route from Origin to Destination in Belgium.
-
Find all the times and connections from NMBS / SNCB (National rail service of Belgium)
Train Catcher - click here for more
-
UK Train station boards. Departure and arrival boards for any train station in the UK.
-
Train catcher allows you to visualise the departure and arrival boards for any train station in the United Kingdom. It also includes all the calling points between origin and destination. Just select which station you which to see and whether you wish a departure or arrival screen.
Mock exams and simulations for the UK driving theory test
Android app for studying for the UK driving theory test.