Publications by T. Moudiki
Super Mario Bros © in the browser using PyScript
**Disclaimer: ** I have no affiliation with Anaconda Inc.If you’re a Pythonista, you’ve probably been hearing about Anaconda’s PyScript in the past few days. Since it’s been introduced at PyCon US 2022, its growth in terms of GitHub stars has been spectacular (to those who are downloading/using my packages thousands of times each month, I...
1782 sym Python (668 sym/1 pcs) 2 img
A Machine Learning workflow using Techtonique
Contents0 – Import packages that will be used in the demo1 – Data-wrangling (using the querier)2 – Modeling/Hyperparameter tuning (using mlsauce and GPopt)3 – Explain model’s decisions (using the-teller)0 – Import packages!pip install querier # A query language for Python Data Frames (part of Techtonique) !pip install mlsauce # Miscel...
906 sym Python (14340 sym/38 pcs) 4 img
Prediction intervals (not only) for Boosted Configuration Networks in Python
In this post, I use the following Python packages:BCN: for adjusting Boosted Configuration Networks regression (see post 1 and post 2 for more details on BCNs, and this notebook for classification examples) to sklearn’s diabetes dataset.the-teller: for interpreting BCNs, and obtaining prediction intervals. So far, as of october 2022, the-telle...
1829 sym Python (5038 sym/18 pcs) 4 img
A web application for forecasting in Python, R, Ruby, C#, JavaScript, PHP, Go, Rust, Java, MATLAB, etc.
Content 0 – Intro 1 – Create an account 2 – Get a token for authentication 3 – Requests for forecasts 4 – On model calibration and cross-validation0 – IntroIn this post, I’ll describe an (work-in-progress) Application Programming Interface (API) for time series forecasting. An API is a system that can receive requests from your comp...
5641 sym Python (3555 sym/9 pcs) 4 img
Simple interfaces to the forecasting API
A few weeks ago, I introduced a forecasting API (Application Programming Interface). The application can be found here: https://techtonique2.herokuapp.com/ So far, as of 2022-11-23, this API contains four methods for univariate time series forecasting (with prediction intervals): mean a (not so naïve) benchmark method, whose prediction is the...
2192 sym R (2077 sym/8 pcs) 2 img
Boosted Configuration (neural) Networks Pt. 2
A few weeks ago, I introduced Boosted Configuration (neural) Networks (BCNs), with some examples of classification on toy datasets. Since then, I’ve implemented BCN for regression (continuous responses) in R, and released a Python version (built on top of the R version) of the package on PyPi. What are BCNs? Statistical/Machine Learning (ML) m...
5336 sym R (5434 sym/6 pcs) 16 img
News from ESGtoolkit, ycinterextra, and nnetsauce
In this post, I introduce new versions of ESGtoolkit, ycinterextra, and nnetsauce. ESGtoolkit (for R) is a toolkit for Monte Carlo Simulation in Finance, Economics, Insurance, Physics, etc. ycinterextra (for R) is used for yield curve interpolation and extrapolation nnetsauce (for Python and R) does supervised Statistical/Machine Learning using...
2923 sym R (1462 sym/9 pcs) 2 img
Boosted Configuration (_neural_) Networks for classification
A few years ago in 2018, I discussed Boosted Configuration (neural) Networks (BCN for multivariate time series forecasting) in this document. Unlike Stochastic Configuration Networks from which they are inspired, BCNs aren’t randomized. Rather, they are closer to Gradient Boosting Machines and Matching Pursuit algorithms; with base learners bei...
3206 sym R (8345 sym/9 pcs) 1 tbl
Model calibration with `crossval`
Model calibration in the context of this post, is about finding optimal hyperparameters for Statistical/Machine learning (ML) models. Optimal in the sense that they minimize a given criterion such as model’s accuracy on unseen data, model’s precision, Root Mean Squared Error (RMSE), etc. What are ML models’ hyperparameters? Let’s take the...
5461 sym R (1748 sym/8 pcs) 6 img
Using R in Python for statistical learning/data science
As someone who’s been using the statistical computing language R for 15 years now, I’d been wondering if there was a way to call my favorite R functions from Python. If you’ve been asking yourself too, yes, there is a way: using rpy2. rpy2 can be installed from the command line as: pip install rpy2 In order to install the R packages neces...
2214 sym Python (1998 sym/7 pcs) 2 img