Machine Learning

I wrote an article about Microsoft integrating Python in Excel. I love the idea and I think it’s going finally get machine learning into the hands of Excel users. Check out this tutorial on how to use SciKit-Learn in Excel from Anaconda. The new Python in Excel integration by Microsoft and Anaconda grants access to the entire Python ecosystem for data science and machine learning. Thanks to its direct connection to Anaconda Distribution, we can leverage built-in functionality with packages like NumPy, pandas, Seaborn, and scikit-learn directly within our Excel workbooks. ...

An older link from April 2017 that I believe became AutoGluon. AutoGluon is fantastic for time series and a host of other AutoML use cases. So, what models do we include in our ensemble? Pretty much any reasonable model we can get our hands on! Specific models include variants on many well-known approaches, such as the Bass Diffusion Model, the Theta Model, Logistic models, bsts, STL, Holt-Winters and other Exponential Smoothing models, Seasonal and other other ARIMA-based models, Year-over-Year growth models, custom models, and more. Indeed, model diversity is a specific objective in creating our ensemble as it is essential to the success of model averaging. Our aspiration is that the models will produce something akin to a representative and not overly repetitive covering of the space of reasonable models. Further, by using well-known, well-vetted models, we attempt to create not merely a “wisdom of crowds” but a “wisdom of crowds of experts” scenario, in the spirit of Mannes [6]. ...

The following changelog captures the internal changes the gold and silver, Bitcoin and dogecoin, and other forecasting model updates or deprecation. This information is for reader and subscriber updates and deployed to markets.neuralmarkettrends.com Version 0.8 September 12, 2025 - Refactored Gold & Silver Price forecast scripts, to be launched on September 13, 2025. Heating Oil and Corn futures forecast prices have been dropped and scrubbed from site. Optimized stock volatility, market regime, and seasonal performance script and redeploying forecasts. Version 0.7 July 25, 2025 - Added pricing forecast for Heating Oil, and Corn Futures. Building a real time scoring of Gold and Silver prices API. Version 0.6 January 22, 2025 - Consolidated Market Regime, Volatility, Seasonal Performance, and Seasonal Returns into simple URLs January 1, 2025 - Restarted Gold, Silver, and Gold to Silver Ratio autoposts Version 0.5.1 March 27, 2024 - Fixed search function to use site data only, no more Google Search January 29, 2024 - Deployed 1 week ahead gold and silver price forecast posts. Deprecated 1 day ahead gold and silver forecasts. January 31, 2024 - Began migration of old markets.neuralmarkettrends.com posts and added Disqus for commenting system. February 7, 2024 - Updated gold and silver price forecasts to included volatility and swing price targets. February 12, 2024 - Added weekly gold to silver ratio forecasts. Version 0.5 January 22, 2024 - markets.neuralmarkettrends.com to be deprecated by February 29, 2024. Version 0.4.1 September 5, 2023: All gold and silver price forecasts have been moved back to markets.neuralmarkettrends.com Version 0.4 June 15, 2023: Deployed 1 day ahead price forecast for silver and gold prices to neuralmarkettrends.com. All 5 day, 4 week, and 6 month price forecast models to remain on markets.neuralmarkettrends.com Version 0.3.1 May 3, 2023: Updated 5 day, 4 week, and 6 month forecast charts with bokeh generated images. Added in auto support and resistance lines to charts. May 3, 2023: Redeployed Bitcoin and Dogecoin forecast models and charts. Version 0.2.1 April 25, 2023: Deployed 5-day forecast and 4-week forecast for gold, silver, and gold to silver ratio models on markets.neuralmarkettrends.com. Forecasts are recalibrated and redeployed daily. Version 0.2.0 April 23, 2023: Deployed long-range gold and silver prices forecast models. Additionally deployed a high-performance and high-accuracy silver prices forecast challenger model to the models in production. April 12, 2023: Added bids and asks volatility spread predictions for gold and silver—recalibrated models with new market information. Version 0.1.0 April 2, 2023: Initial Gold and Silver price forecasting model deployed.

Yet, the Market call is very hard to ignore. She sings of easy money and it’s tempting.

In this week’s Data Science and Machine Learning link round-up, we’ll share some links that caught our (my) eye during the past few weeks.

A flowchart/gist of Resnet 512.

Stacking models is a great technique for squeezing out more performance. First, let me describe what I mean by stacking. The idea is to divide the training set into several pieces like you would do in k-folds cross validation. For each fold, the rest of the folds are used to obtain a predictions using all the models 1…M. Via Burak Himmetoglu’s Blog

All the stuff you post about yourself and what you like in Facebook or some other social network is a marketer’s wet dream. Data mining companies are now capitalizing on the free information you post about yourself, mining it, and then selling statistically significant data relationships to marketers via the social network’s APIs. A company called Colligent mines social networks for data that it sells to record labels to help them decide which demographics or individual fans might like a particular artist, and those are just the very first nuggets marketers pull out of profiles. ...

** There are NEW livestream videos about RapidMiner! Visit my Channel here ** In Lesson 3, I introduced the Sliding Window Validation operator to test how well we can forecast a trend in a time series. Our initial results are very poor, we were able to forecast the trend with an average accuracy of 55.5%. This is fractionally better than a simple coin flip! In this updated lesson I will introduce the ability of Parameter Optimization in RapidMiner to see if we can forecast the trend better. ...

** There are NEW livestream videos about RapidMiner! Visit my Channel here ** In Lesson 2, I went over the concept of MultiObjective Feature Selection (MOFS). In this lesson we’ll build on MOFS for our model but we’ll forecast the trend and measure it’s accuracy. Revisiting MOFS We learned in lesson 2 that RapidMiner can simultaneously select the best features in your data set while maximizing the performance. We ran the process and the best features were selected below. ...