Pinned Repositories
Deep-Learning
Implemented the deep learning techniques using Google Tensorflow that cover deep neural networks with a fully connected network using SGD and ReLUs; Regularization with a multi-layer neural network using ReLUs, L2-regularization, and dropout, to prevent overfitting; Convolutional Neural Networks (CNNs) with learning rate decay and dropout; and Recurrent Neural Networks (RNNs) for text and sequences with Long Short-Term Memory (LSTM) networks.
feature-selection-for-machine-learning
feature_engine
Feature engineering package with sklearn like functionality
HeteroArchGen4M2S
HeteroArchGen4M2S: An automatic software for configuring and running heterogeneous CPU-GPU architectures on Multi2Sim simulator. This tool is built on top of M2S simulator, it allows us to configure various heterogeneous CPU-GPU architectures (e.g., number of CPU cores, GPU cores, L1$, L2$, memory (size and latency (via CACTI 6.5)), network topologies (currently support 2D-Mesh, customized 2D-Mesh, and Torus networks)...). The output files include the results of network throughput and latency, caches/memory access time, and dynamic power of the cores (can be collected after running McPAT).
Machine-Learning
Machine learning techniques, such as Linear Regression, Logistic Regression, Neural Networks (feedforward propagation, backpropagation algorithms), Diagnosing Bias/Variance, Evaluating a Hypothesis, Learning Curves, Error Analysis, Support Vector Machines, K-Means Clustering, PCA, Anomaly Detection System, and Recommender System.
RLE-NOC
SDC-term1-Advanced-Lane-Finding
Detected highway lane boundaries on a video stream with OpenCV image analysis techniques, including camera calibration matrix, distortion correction, color transforms, gradients, etc., to create a thresholded binary image, a perspective transform to rectify binary image ("birds-eye view"). Detected lane pixels and fit to find the lane boundary, determined the curvature of the lane and vehicle position with respect to center. Warped the detected lane boundaries back onto the original image.
SDC-term1-Behavioral-Cloning
Built and trained a convolutional neural network to drive the car itself autonomously in a simulator using Tensorflow (backend) and Keras. Experimented with a modified Nvidia architecture. Performed image processing with brightness, shadow augmentation, and flipped images. Used dropout and Adam optimizer to generalize the network for driving multiple tracks. The datasets are used via Udacity's source for training the model. Trained the model on Amazon AWS EC2 platform with GPU instances.
SDC-term1-Traffic-Sign-Classifier
Built and trained a deep neural network to classify traffic signs, using TensorFlow. Experimented with different network architectures. Performed image pre-processing and validation to guard against overfitting. The datasets are collected from the German Traffic Sign for training and random traffic signs downloaded from internet for testing.
Statistical-Learning
ttungl's Repositories
ttungl/Deep-Learning
Implemented the deep learning techniques using Google Tensorflow that cover deep neural networks with a fully connected network using SGD and ReLUs; Regularization with a multi-layer neural network using ReLUs, L2-regularization, and dropout, to prevent overfitting; Convolutional Neural Networks (CNNs) with learning rate decay and dropout; and Recurrent Neural Networks (RNNs) for text and sequences with Long Short-Term Memory (LSTM) networks.
ttungl/Machine-Learning
Machine learning techniques, such as Linear Regression, Logistic Regression, Neural Networks (feedforward propagation, backpropagation algorithms), Diagnosing Bias/Variance, Evaluating a Hypothesis, Learning Curves, Error Analysis, Support Vector Machines, K-Means Clustering, PCA, Anomaly Detection System, and Recommender System.
ttungl/SDC-term1-Advanced-Lane-Finding
Detected highway lane boundaries on a video stream with OpenCV image analysis techniques, including camera calibration matrix, distortion correction, color transforms, gradients, etc., to create a thresholded binary image, a perspective transform to rectify binary image ("birds-eye view"). Detected lane pixels and fit to find the lane boundary, determined the curvature of the lane and vehicle position with respect to center. Warped the detected lane boundaries back onto the original image.
ttungl/basic_reinforcement_learning
An introductory series to Reinforcement Learning (RL) with comprehensive step-by-step tutorials.
ttungl/data-engineering-gcp
Data Engineering on Google Cloud Platform
ttungl/SDC-term1-Traffic-Sign-Classifier
Built and trained a deep neural network to classify traffic signs, using TensorFlow. Experimented with different network architectures. Performed image pre-processing and validation to guard against overfitting. The datasets are collected from the German Traffic Sign for training and random traffic signs downloaded from internet for testing.
ttungl/SDC-term2-Model-Predictive-Control
Implemented Model Predictive Control to drive the vehicle around the track (even with additional latency between commands).
ttungl/120-DS-Interview-Questions
My Answer to 120 Data Science Interview Questions
ttungl/aws-lex-web-ui
Sample Amazon Lex chat bot web interface
ttungl/beam-example
Apache Beam example
ttungl/clusterone-sdc-test
ttungl/clusterone-word2vec
ttungl/clusterone-word2vec-1
ttungl/Coding-Interview-Challenge
Python concise solutions to the Leetcode problems.
ttungl/Competitive-Data-Science
ttungl/competitive-data-science-1
How to Win a Data Science Competition: Learn from Top Kagglers
ttungl/Data-Science-cs109
ttungl/DataflowJavaSDK
Google Cloud Dataflow provides a simple, powerful model for building both batch and streaming parallel data processing pipelines.
ttungl/Does-Twitter-Hate-Cats
ttungl/donation-analytics
As a data engineer working for political consultants whose clients are cash-strapped political candidates. They've asked for help analyzing loyalty trends in campaign contributions, namely identifying areas of repeat donors and calculating how much they're spending. Identify areas (zip codes) that could be sources of repeat campaign contributions.
ttungl/examples
A set of examples around pytorch in Vision, Text, Reinforcement Learning, etc.
ttungl/FreeML
Data Science Resources (Mostly Free)
ttungl/kaggle-past-solutions
A searchable compilation of Kaggle past solutions
ttungl/models
Models and examples built with TensorFlow
ttungl/object-detection
Detect a Phone or not a Phone in the images. Use Logistic Regression to classify the images. Prediction accuracy is achieved at 96.15%.
ttungl/Pinterest-coding-challenge
ttungl/python-reference
Python Reference (The Right Way)
ttungl/reinforcejs
Reinforcement Learning Agents in Javascript (Dynamic Programming, Temporal Difference, Deep Q-Learning, Stochastic/Deterministic Policy Gradients)
ttungl/self-driving-demo
Distributed TensorFlow implementation of steering angle prediction model for self-driving car. Includes integration to run on ClusterOne.
ttungl/spatial-transformer-network
A Tensorflow Implementation of Spatial Transformer Networks