Pinned Repositories
Bank-Marketing-Classification
Bank Marketing Classification using scikit-learn library to train and validate classification models like Logistic Regression, Decision Tree, Random Forest, Naïve Bayes, Neural Network and Support Vector Machine. Also conducted comparative study on the above models when applied on different feature sets obtained via feature selection (Chi-Square Test), feature transformation (Principal Component Analysis) and feature elimination.
Batch-Normalization
Batch Normalization using tf.layers and tf.nn libraries. Also comparing the performances with and without batch normalization.
Clustering-Analysis-on-customers-of-a-wholesale-distributor
Project in which unsupervised learning techniques are applied on product spending data collected for customers of a wholesale distributor in Lisbon, Portugal to identify customer segments hidden in the data.
Credit-Card-Fraud-Detection
Credit Card Fraud Detection using various Anomaly Detection Algorithms and Techniques
Finding-Similar-Images-using-Locality-Sensitive-Hashing
Finding Similar Images using Locality Sensitive Hashing (using Min Hashing and Random Hyperplanes technique)
General-Adversarial-Network-MNIST
Building our own GAN and train it on MNIST dataset. Built the generator and discriminator networks using Tensorflow.
Tweet-analysis-to-find-new-gym-clients
Finding local tweets (that are restricted to any geographic location) and search for people who require and/or are interested in fitness and gym activities
Vehicle-Detection-using-Faster-R-CNN
Vehicle Detection using state-of-the-art architecture Faster R-CNN with Pretrained models like VGG16 and ResNet50
Year-Prediction-using-Regression
Predicting Release Year of a Song using different Regression algorithms implemented from scratch using numpy
prakhardogra921's Repositories
prakhardogra921/Vehicle-Detection-using-Faster-R-CNN
Vehicle Detection using state-of-the-art architecture Faster R-CNN with Pretrained models like VGG16 and ResNet50
prakhardogra921/Bank-Marketing-Classification
Bank Marketing Classification using scikit-learn library to train and validate classification models like Logistic Regression, Decision Tree, Random Forest, Naïve Bayes, Neural Network and Support Vector Machine. Also conducted comparative study on the above models when applied on different feature sets obtained via feature selection (Chi-Square Test), feature transformation (Principal Component Analysis) and feature elimination.
prakhardogra921/Batch-Normalization
Batch Normalization using tf.layers and tf.nn libraries. Also comparing the performances with and without batch normalization.
prakhardogra921/Finding-Similar-Images-using-Locality-Sensitive-Hashing
Finding Similar Images using Locality Sensitive Hashing (using Min Hashing and Random Hyperplanes technique)
prakhardogra921/General-Adversarial-Network-MNIST
Building our own GAN and train it on MNIST dataset. Built the generator and discriminator networks using Tensorflow.
prakhardogra921/Deep-Q-Learning
Tensorflow implementation of a deep neural network that can learn to play games via reinforcement learning. For this notebook the game used is the Cart-Pole game that is available in the OpenAI Gym library. Also the deep neural network can be used for other games as well. In order to run this on your system you will also need to clone the OpenAI gym repository.
prakhardogra921/Monte-Carlo-Methods-for-Reinforcement-Learning
Implementations of many Monte Carlo (MC) algorithms for updating policies of an environment using action values, greedy and epsilon-greedy procedures. Environment used for this notebook is the BlackJack Environment (can be seen in the OpenAI Gym library) and these functions can be used for other environments as well.
prakhardogra921/Year-Prediction-using-Regression
Predicting Release Year of a Song using different Regression algorithms implemented from scratch using numpy
prakhardogra921/Autoencoders-using-NN-and-CNN
Autoencoders using simple Neural Networks and Convolutional Neural Networks. Also demonstrating that CNNs can be used for image denoising.
prakhardogra921/Character-wise-prediction-using-LSTM
Character-wise prediction using LSTM. Implemented using Tensorflow.
prakhardogra921/Deep-Convolutional-Generative-Adversarial-Network
Implement the Deep Convolutional GAN model to generate full color images. Trained the DCGAN model on the Street View House Numbers dataset.
prakhardogra921/Dog-Breed-Classifier
Dog Breed Classifier
prakhardogra921/Dynamic-Programming-for-Reinforcement-Learning
Implementations of many classical dynamic programming algorithms for updating policies of an environment via policy improvement and value iteration procedures. The environment used for this notebook is the Frozen Lake Environment (can be seen in the OpenAI Gym library) and the functions can be used for other environments as well.
prakhardogra921/Face-Generation
Using generative adversarial networks to generate new images of faces. Also tested the GAN with MNIST dataset.
prakhardogra921/Generating-TV-Scripts-using-LSTM
Generating our own Simpsons TV scripts using LSTM. Used part of the Simpsons dataset of scripts from 27 seasons. The Neural Network built will generate a new TV script for a scene at Moe's Tavern. Implemented using Tensorflow.
prakhardogra921/HandWritten-Digit-Recognition-using-MLP
Hand-Written Digit Recognition using MLP (Multi-Layer Perceptron). Implemented using Keras library.
prakhardogra921/Image-Augmentation-on-CIFAR-10-Dataset
Image Augmentation on CIFAR-10 Dataset to improve classification accuracy
prakhardogra921/Image-Classification-on-CIFAR-10-Dataset
Image Classification on CIFAR-10 Dataset using CNN (Convolutional Neural Networks) and MLP (Multi-Layer Perceptron)
prakhardogra921/Image-Styling
Style Transfer of one image to another
prakhardogra921/Neural-Network-using-Tensorflow
Using Neural Network on notMNIST dataset using Tensorflow
prakhardogra921/Practising-Keras
This repository contains 2 notebooks: Predicting Student Admission and Analyzing IMDB data using Keras
prakhardogra921/Regression-using-Neural-Networks
Built a neural network from scratch to predict number of bikeshare users on a given day
prakhardogra921/Sentiment-Classification-using-LSTM
Sentiment Classification using LSTM. Implemented using Tensorflow.
prakhardogra921/Sentiment-Classification-using-Neural-Network
Sentiment Classification using Neural Network
prakhardogra921/Skip-Gram-word2vec
Implemented the word2vec algorithm using the skip-gram architecture. Implemented using tensorflow.
prakhardogra921/Smart-Cab-training-using-Q-Learning
Train a self-driving agent (Smart Cab) using Q-Learning algorithm
prakhardogra921/Social-Media-Complaint-Workflow-Automation-Tool-Using-Sentiment-Intelligence
Automating the complaint classification and forwarding mechanism for bank posts on Facebook web pages.
prakhardogra921/TD-Learning
Implementations of many Temporal-Difference (TD) methods like Sarsa, expected Sarsa and Q-learning control algorithms. Environment used for this notebook is the Cliff Walking Environment (can be seen in the OpenAI Gym library) and these functions can be used for other environments as well.
prakhardogra921/Teach-a-Quadcopter-to-fly
Designed an agent to fly a quadcopter, and then train it using Deep Q-Learning algorithm.
prakhardogra921/Transfer-Learning-using-VGGNet
Transfer Learning on Flower Images using VGGNet