harish3110 Goto Github PK

airline_tweets_sentiment_analysis

This repository contains code for building a Sentiment Classifier on Kaggle's Twitter Airlines Tweets by using the famous ULMFiT approach and decoding the methodologies used in the paper.

bagging-classifier-analysis

The following iPython notebook is to analyze the importance of bootstrap aggregations (bagging) by comparing base learners with bagged models on various classification ML models.

boosting-classifier-analysis

carnd-semantic-segmentation

Identified pixel-wise navigable road area in car dash cam images using TensorFlow and a Fully Convolutional Network (FCN) based on the VGG-16 image classifier architecture (trained and tested on the KITTI data set). Achieved near perfect identification of road area on holdout test dataset, as a qualitative measure.

deep-learning-v2-pytorch

Projects and exercises for the latest Deep Learning ND program https://www.udacity.com/course/deep-learning-nanodegree--nd101

dlwpt-code

Code for the book Deep Learning with PyTorch by Eli Stevens, Luca Antiga, and Thomas Viehmann.

document-ai

dogs-vs-cats-classifier

Aim to create an algorithm that successfully classifies whether images contain either a dog or a cat successfully using convolution neural networks

fastai-v3

Starter app for fastai v3 model deployment on Render

fastpages

An easy to use blogging platform, with enhanced support for Jupyter Notebooks.

kinship-detection-using-vggface

Project based on the Kaggle competition “Recognizing Faces in the Wild” which involved extracting the facial feature set from the FIW dataset using VGGFace in order to develop a Siamese CNN network to check for kinship between images of people. The model thus deveoped acheived an ROC score of 0.802 on the leaderboard which ranked our team in the top 200.

mlnd-p2-customer-segmentation

Customer Segmentation/ Unsupervised Learning module project in Udacity's Machine Learning Nanodegree Program

mlnd-p3-finding_donors

my_fastai_notebooks

A collection of all my fastai notebooks during Part 1 of Fastai v4 'Practical Deep Learning for Coders'

my_hf_notebooks

Repo consisting of notebooks pertaining to training NLP models using HuggingFace Transformer models

nlp

A collection of all my NLP projects.

nlp-with-python

Scikit-Learn, NLTK, Spacy, Gensim, Textblob and more

ntentional

The ntentional blog - a machine learning journey

patent_scraping

predicting-pl-2016-17-results

This is the code for Predicting Premier League Winners for a 38 match 2016-2017 season using multiple machine learning models based on training data from 2000 - 2016 PL Season data and finally choosing the most optimum model and optimizing it further.

principal-component-analysis

Introduction to PCA using Sklearn on Wines dataset and comparing the results of PCA vs normal logistic regression.

project-8--kidnapped-vehicle-using-particle-filter

Implemented a two-dimensional particle filter in C++ capable of localizing a vehicle within desired accuracy and time. Combined known map waypoints with simulated LiDAR data and vehicle pose transformations to achieve accurate localization.

pytorch-tutorial

A collection of notebooks for using PyTorch for various Deep Learning tasks.

sdcnd-capstone_project-path_planning

Built a path planner in C++ that autonomously navigates a vehicle through traffic on a highway. Employed environmental prediction, behavioral planning, and trajectory generation to achieve over 20 miles of smooth, incident-free driving.

sdcnd-p1-lane_line_detection

Utilised OpenCV in Jupyter notebook to create a simple image processing pipeline for detecting and identifying highway lane lines in an image or video using traditional Computer Vision techniques. Achieved lane line recognition across all frames of two different vehicle dash cam videos.

sdcnd-p10-model_predictive_control

Implemented Model Predictive Control to drive a vehicle around a track using IPOPT and CPPAD packages to determine optimal trajectory. Achieved reliable performance even at high speeds and with additional latency between commands.

sdcnd-p2-traffic_sign_classifier

Utilized TensorFlow deep learning framework and OpenCV in Python to train a classifier for the GTSRB traffic sign dataset. Implemented data augmentation and image jitter to achieve 95.6% accuracy on hold-out test data set.

sdcnd-p3-behavioral_cloning

Utilized Keras deep learning framework and OpenCV computer vision framework in Python to train a car to drive in a simulator. Achieved full performance in the training environment, as well as a previously unseen environment, through intricate data selection/augmentation strategy and neural network tuning. Video Link:

sdcnd-p4-advanced_lane_finding

Utilized OpenCV in Python Jupyter notebook to create a robust image processing pipeline for detecting, recognizing, and identifying the current highway lane in an image or video. Additionally, calculated car position within lane and lane radius of curvature based on coefficients of polynomial fit. Achieved lane line recognition across all frames of two different vehicle dash cam videos.

sdcnd-p5-vehicle_detection

Created a vehicle detection and tracking pipeline with OpenCV, histogram of oriented gradients (HOG), and support vector machines (SVM). Optimized and evaluated the model on video data from an automotive camera taken during highway driving.