Python Projects
PythonNotes on good ideas for organizing and developing a Python project. […]
Continue reading [View source ]
Reinforcement Learning
CartPole-v0
Reinforcement Learning
Solve the CartPole-v0 from
OpenAI gym using Q-learning with experience resampling.
[…]
The Unsupervised Learning Trick
AlgorithmIn this git repository I show the unsupervised learning trick on the mnist data using a Random Forest. It works sorta ok. […]
Continue reading [View source ]
Correlation Clustering
TechniqueAn example of how to cluster numerical features using their correlation. […]
Continue reading [View source ]
Articles and Blogposts
ReferencesA list of references to blog posts, articles, books etc. I find nice. […]
Continue reading [View source ]
Bokeh Example
PythonAn app used to display real estate data in Denmark. […]
Continue reading [View source ]
Docker Container
DockerCommands to run a minimal docker container. […]
Continue reading [View source ]
Haversine
Distance MetricThe haversine distance can be used to calculate the distance between two points on a sphere. Here is my derivation of it. […]
Continue reading [View source ]
Tqdm
Pythontqdm is a nice Python module
for creating a progressbar when processing a list of items. […]
Operationel Range
ThoughtsWhile working at Maersk there was a power-to-speed model used for planning, basically a polynomial inferred from a regression, taking in some extra inputs besides power and outputting speed. For this model there was defined valid input power levels. In this post I will talk about my thoughts on why it seems useful to define the operational range AND data type of an inferred function in an operational setting. […]
Continue reading [View source ]
Hodrick Prescott Filtering
AlgorithmHodrick-Prescott filtering is a filtering method. A filtering method can be used to find \(x_t\) given \(x_1,\dots,x_t\). In this post we will go through a simple filtering example with Hodrick-Prescott filtering. […]
Continue reading [View source ]
Machine Learning Yearning
TechnologyA presentation I did on the book Machine Learning Yearning by Andrew Ng using remark.js. […]
Continue reading [View source ]
Prime Numbers And Strangeness Of Independent Random Variables
ThoughtsRandom variables have an intuitive definition, but might not fully cover the intuitive concept of independence. […]
Continue reading [View source ]
Difference Between The Pyspark Pipeline And The Sklearn Pipeline
ProgrammingThe pipeline object in sklearn makes it easy to sequentially apply a list of transforms and a final estimator [1]. This makes it easy to organize models. PySpark has a similar pipeline API but there are some differences. […]
Continue reading [View source ]
Lazy Proof Of Monty Hall
RiddleThe Monty Hall problem is an awesome brain teaser […]
Continue reading [View source ]
Pydata London
TechnologyWhile working at Maersk I went to PyData 2019. I found two presentations very useful and I have used learnings from both in my day job. […]
Continue reading [View source ]
Pyspark Tricks
ProgrammingSome useful patterns / tricks I find myself using at work. […]
Continue reading [View source ]
Broadcast Variable Pyspark Example
ProgrammingBroadcasting a variable is useful for repeatedly used read-only variables in an application […]
Continue reading [View source ]
Gradient Descent
AlgorithmThe gradient descent algorithm and its variants are some of the most widely used optimization algorithms in machine learning today. In this post a super simple example of gradient descent will be implemented. […]
Continue reading [View source ]
Leave One Out Encoding
EncodingLeave one out encoding is a new feature encoding technique used by a Kaggle Grandmaster […]
Continue reading [View source ]
Model Evaluation
Statistical LearningGiving an estimate of generalization error in Machine Learning is vital. […]
Continue reading [View source ]
XGBoost
AlgorithmIn this post we will go through an example application of the XGBoost algorithm […]
Continue reading [View source ]
Adaboost The Original Boosting Algorithm
AlgorithmIn this post the famous AdaBoost algoritm will be implemented in R […]
Continue reading [View source ]
Blagging
AlgorithmTwo sampling methods often applied in practice are: oversampling and undersampling. […]
Continue reading [View source ]
Perceptron
AlgorithmThe Perceptron by Rosenblat is one of the earliest machine learning algorithms. It was proposed as a model of an artificial neuron, and can be used for classification. […]
Continue reading [View source ]