Paras Naren's Projects
PSQE - Probability, Statistics and Queueing for Engineers
Simple air hockey simulation with two movable pads and a continously moving ball.
Awesome AI Learning with +100 AI Cheat-Sheets, Free online Books, Top Courses, Best Videos and Lectures, Papers, Tutorials, +99 Researchers, Premium Websites, +121 Datasets, Conferences, Frameworks, Tools
Developed a completely Remote and Automized Scanning and Storage system, that implements OCR to extract text from a scanned RC Card and stores the structured data into a Database.
A program that performs named entity and category recognition on English sentences
One of the most popular starter data sets in data science, the Titanic data set. This is a data set that records various attributes of passengers on the Titanic, including who survived and who didn’t. Here I have detected some missing value, replace the missing values and also create new values added to the dataset. There are two csv files, first one is titanic_original.csv and second one is tatanic_clean.csv. Second csv is generated from the R code, called 'titanic.r' here. Have fun.
Cheat Sheets
Keras Implementation of popular Deep RL Algorithms (A3C, DDQN, DDPG, Dueling DDQN)
A NLP Model that detects cues (words that cause negation) and scope (negated part of the sentence). The model employs NSlack SVM for binary classification of cues and Conditional Random Fields for scope detection.
A handwritten digit classifier that classifies digits from the MNIST database using a Convolution Neural Network.
A image detection model that is used to differentiate between regular vehicles and emergency vehicles.
TensorFlow examples
Part of Orient Bell Data Innovation Challenge, where a normalised RMSE leaderboard score of 99.968 was achieved.
Rotation of a grayscale image by 90° , 180° and 360°
Synthesis of images, primarily of birds from user vocal descriptions of images
Developed Phrase Embeddings of Medical data, using articles obtained from the PubMed FTP server. Medical terms that are semantically similar were clustered together, thereby helping to identify the inconsistencies
A data analytical approach to predict the best performing 11 players of IPL 2019, using player data from the past 10 seasons.
Given a snapshot of a social network, we infer which new interactions among its members are likely to occur in the near future?
Participated in the Phillips Datascience Hackathon that provided a dataset of mobile specifications with a target variable of price range to be predicted.
Predicting discount to be provided by a pharmaceutical company to its customers using Mutli-class classification.
A poem generation model, that generates poems modeled on the famous Christopher Poindexter's poems and quotes.
📚 A practical approach to machine learning.
A Random Forest Regression approach to predict whether the given individual with a past history of drug use will resort back to it.