Retinal Vasculature Segmentation with a U-Net Architecture

From: https://towardsdatascience.com/retinal-vasculature-segmentation-with-a-u-net-architecture-d927674cf57b The structures exhibited by the retinal vasculature infer critical information about a wide range of retinal pathologies such as Prematurity (RoP), Diabetic Retinopathy(DR), Glaucoma, hypertension, and Age-related Macular Degeneration(AMD). These pathologies are amongst the leading causes of blindness. Accurate segmentation of retinal vasculature is important for various ophthalmologic diagnostic and therapeutic procedures. A … Continue reading Retinal Vasculature Segmentation with a U-Net Architecture →

LSTM-FCN for cardiology

From: https://towardsdatascience.com/lstm-fcn-for-cardiology-22af6bbfc27b We will work on the application of the algorithm on a database containing electrocardiograms (ECG) of 2 different types, and see if this new model can help detect patients at a high risk of sudden death. LSTM-FCN architecture LSTM-FCN architecture (source: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8141873) This algorithm consists of 2 parts: a LSTM block and a FCN … Continue reading LSTM-FCN for cardiology →

Bayesian Inference Algorithms: MCMC and VI

From: https://towardsdatascience.com/bayesian-inference-algorithms-mcmc-and-vi-a8dad51ad5f5 Unlike other areas of machine learning (ML), Bayesian ML requires us to know when an output is not trustworthy. When you train a regression or xgboost model, the model can be taken at face value given the settings and data. With Bayesian ML, the output is not guaranteed to be correct. Bayesian workflow … Continue reading Bayesian Inference Algorithms: MCMC and VI →

PyTorch layer dimensions: what size and why?

From: https://towardsdatascience.com/pytorch-layer-dimensions-what-sizes-should-they-be-and-why-4265a41e01fd Preface This article covers defining tensors, and properly initializing neural network layers in PyTorch, and more! Introduction You might be asking: “How do I initialize my layer dimensions in PyTorch without getting yelled at?” Is it all just trial and error? No, really… What are they supposed to be? For starters, did you … Continue reading PyTorch layer dimensions: what size and why? →

Data Science Project Flow for Startups

From: https://towardsdatascience.com/data-science-project-flow-for-startups-282a93d4508d I was recently asked by a startup I’m consulting (BigPanda) to give my opinion about the structure and flow of data science projects, which made me think about what makes them unique. Both managers and the different teams in a startup might find the differences between a data science project and a software … Continue reading Data Science Project Flow for Startups →

Deploying R Shiny apps using ShinyProxy on Windows 10

From: https://www.databentobox.com/2019/11/05/deploy-r-app-with-shinyproxy/ Background R Shiny is a powerful tool for building data products, from data visualisations to predictive models. Built by RStudio, this package is highly integrated with the RStudio IDE, making it the primary choice for production. Although it is relatively easy to build a Shiny app and make it run on our local machines, … Continue reading Deploying R Shiny apps using ShinyProxy on Windows 10 →

Build Data Pipelines with Apache Airflow

From: https://towardsdatascience.com/https-medium-com-xinran-waibel-build-data-pipelines-with-apache-airflow-808a4de79047 Originally created at Airbnb in 2014, Airflow is an open-source data orchestration framework that allows developers to programmatically author, schedule, and monitor data pipelines. Airflow experience is one of the most in-demand technical skills for Data Engineering (another one is Oozie) as it is listed as a skill requirement in many Data Engineer job postings. In … Continue reading Build Data Pipelines with Apache Airflow →

Denosing Lung CT Scans using Neural Networks with Interactive Code — Part 4, Convolutional Residual Neural Networks

From: https://towardsdatascience.com/denosing-lung-ct-scans-using-neural-networks-with-interactive-code-part-4-convolutional-resnet-74335714a4ae Another attempt to denoise CT Scan of lungs, this time we are going to use more sophisticated Convolutional ResNet Architecture. Specifically, we are going to use the architecture proposed in this paper, “Deep Residual Learning for Image Recognition”. Also, as usual lets do manual back propagation to compare our results. Network Architecture (Image Form) Image … Continue reading Denosing Lung CT Scans using Neural Networks with Interactive Code — Part 4, Convolutional Residual Neural Networks →

Denosing Lung CT Scans using Neural Networks with Interactive Code — Part 3, Convolutional Residual Neural Networks

From: https://towardsdatascience.com/denosing-lung-ct-scans-using-neural-networks-with-interactive-code-part-3-convolutional-residual-6dbb36b28be So since I will be using a lot of image data, I will move on to Tensorflow to harness the power of GPU however, no worries, we will implement all of our back propagation. (Also compare the final results with auto differentiation). Now due to midterms I wasn’t able to do much, so … Continue reading Denosing Lung CT Scans using Neural Networks with Interactive Code — Part 3, Convolutional Residual Neural Networks →

Bayesian Nonparametric Models

From: https://www.datasciencecentral.com/profiles/blogs/6448529:BlogPost:635167 Bayesian Nonparametrics is a class of models with a potentially infinite number of parameters. High flexibility and expressive power of this approach enables better data modelling compared to parametric methods. Bayesian Nonparametrics is used in problems where a dimension of interest grows with data, for example, in problems where the number of features … Continue reading Bayesian Nonparametric Models →