Marlon Misra's repositories
advanced-lane-finding
advanced lane finding pipeline. used distortion correction, various color spaces, Sobel operators, perspective transforms, and polynomial line fitting.
lane-finding
pipeline that annotates front-facing car video with lane lines. used color transforms, edge detection, Hough feature extraction, and line segment analysis.
vehicle-tracking
pipeline that detects other vehicles in front-facing car video footage. used various feature extractors, svm's, conv. neural nets, and a window search technique.
behavioral-cloning
convolutional net that determines optimal steering angles based on dashcam footage.
CarND-Behavioral-Cloning-P3
starting files for the Udacity behavioral cloning project
custom-app-base
This is the base repo for building custom apps that can be forked to create a new app.
face-keypoints-lasagne
playing around with the lasagne framework for face keypoint detection
lane-finding-beginner
Pipeline that annotates dash cam footage with lane lines. used edge detection, gough transforms, filters masks.
langchain-ts-starter
Langchain.js template to get started quickly
marlon-misra-2
personal portfolio website
marlonmisra
minimalist personal website
marlonmisra2.github.io
personal page
predicting-boston-housing-prices
boston housing price prediction model. udacity project focused on model evaluation.
Realtime_Multi-Person_Pose_Estimation
Code repo for realtime multi-person pose estimation in CVPR'17 (Oral)
student-intervention
ml project on student success. cleaned data, processed features, and determined accuracy metrics. implemented SVM, NB, and DT models. tuned params and justified best model.
titanic-survival-exploration
simple exploratory project with python and pandas
traffic-sign-recognition
deep cnn's that classifies traffic signs. utilized conv layers, pooling layers, dropout, r2 regularization.
traffic-signs-old
deep cnn's that classifies traffic signs. utilized conv layers, pooling layers, dropout, r2 regularization.
traffic-signs-with-keras
after using tensorflow to build a traffic sign classification model, used keras to do the same in fewer steps.
transfer-learning-with-alexnet
project on transfer learning. started off with alexnet and did feature extraction and training on new model.