sayakdhar Goto Github PK

-self-playing-atari-game-using-reinforcement-learning-natural-language-processing

We introduce the first deep reinforcement learning agent that learns to beat Atari games with the aid of natural language instructions. The agent uses a multimodal embedding between environment observations and natural language to self-monitor progress through a list of English instructions, granting itself additional reward for completing instructions in addition to increasing the game score. Our agent significantly outperforms Deep-Q Networks, Asynchronous Advantage Actor-Critic (A3C) agents, and the best agents posted to Open-AI Gym on what is often considered the hardest Atari 2600 environment: MONTEZUMA’S REVENGE.

automation-menu

The whole project is developed for automating all the near possible trending technologies in one go. We are providing an user-friendly environment to our project for the succourment to any type of user. According to the user's desideratum , just a single click will create an obscuring environment in the background which is connected with ansible for the creation, setting up and configuration of the desired output and this will be viewed in the website(successfull or not).The whole project is authenticated by a stellar authentication page that will be unlocked only by the user's faceprint with the use of LBPH algorithm.

aws

AWS Practices & Learnings

business-model-for-a-led-tv-brand

We saw the television products that are currently available in the market, became aware that the technology available to the people is at very high tag price. Hence we came up with our brand that’ll provide people the luxury of Technology at an Affordable Price.

directory-structure-implementation

1. Create folders 2. Back or Up button 3. Navigate easily between parent folders and subfolders in UI 4. Show current pwd (Present Working Directory) in top This directory implementation is similar to an operating system directory structure, where you can create directory and browse through them at various levels.

duplexa

The proposed application aims at converting sign language into human readable format (as subtitles) and projects it directly to our eyes/spectacles. This enables a person, who’s otherwise not acquainted with sign lexicons to understand a dumb person. As promised for duplex connectivity, the person who wishes to communicate with a mute person, can type in just the exact sentences which will be translated into the relevant signs and made available through the display The application can be scaled by making it capable of deciphering Braille as well, which can serve as a respite for normal persons understand the script as well

esca-leviosa-ar-menu-

Digital Marketing, Augmented Reality, and Mobile Computing are three technology trends dominating the commerce world. In this project, we are proposing an Android app based on Augmented Reality, which is going to help restaurants to provide better service to their customers. This project presents an application on mobile phones that represents menus into a 3D model of the dishes present on it. The system runs on an android platform mobile phone. It uses image recognition to trace and register image markers. Then, it processed the input into a pre-processed feature database of menu items using Features from Accelerated Segment Test (FAST) implemented by Vuforia Library. The system will check the database, and after the matching it will generate the 3D model of the food using the Unity 3D game engine. It is able to trace and render up to six 3D models of the dishes in real-time. This app has various new features like real-time tagging and food identifier. Firstly, the presented Android Application is opened. The camera will start automatically which require to scan the menu with the dishes. When you scan the menu, an augmented 3D models of all the items present on that menu will pop up. Now you can see 3D dishes on your table through the android device. Calories for every item is shown alongside the 3D model. Along with this we have an animated video which will let other restaurants share their menu cards with us and also work along with us.

flight-status-for-airline-employees

This project contains total functions based on current Flight Statuses which can be viewed by airline employees .The total project is based on Php.

interview

online-banking-management

This project contains total functions based on online banking system.The total project is based on Php.

python_codings

Python related files

quality-assurance_-sayak-dhar-leadsquared-technologies-

Answers to the given three questions for Quality Assurance profile

react-rest-api-with-aws

student-dbms

This project contains total functions based on Student Database Management system.The total project is based on C.

test

text-generator-using-unidirectional-bidirectional-lstm-rnn-

Text generation is a subfield of natural language processing. It leverages knowledge in computational linguistics and artificial intelligence to automatically generate natural language texts, which can satisfy certain communicative requirements. The purpose of this project is to discuss about text generation, using machine learning approaches, especially Recurrent Neural Networks (RNN).

vit-gamedev

weekly-work

All my weekly small projects comprising of python codes are attached to this repository