In this post, we’ll try to forecast my weight using Forecast and Facebook’s Prophet packages. We’ll see what is the performance from Facebook’s method in a simple case of forecast.

Usually Yann LeCun’s MNIST database is used to explore Artificial Neural Network architectures for image recognition problem.

In the last post the use of a ANN (LeNet architecture) implemented using mxnet to resolve this classification problem.

But in this post, we’ll see that the MNIST problem isn’t a difficult one, only resolved by ANNs, analyzing the data set we can see that is possible, with a high degree of precision, resolving this classification problem with a simple k-nearest neighbors algorithm.

In this R Notebook I implement an Convolutional Neural Network (CNN) using the MNIST Database for handwritten digits recognition using mxnet framework for R.

In this quick post, we’ll take one MTB ride, tracked by FitBit in a TXC File, and generate a animated gif. Using gganimate Package and using the same code we learned, we can animate the map with few words.

Collinearity refers to the situation in which two or more predictor variables collinearity are closely related to one another. The presence of collinearity can pose problems in the regression context, since it can be difficult to separate out the individual effects of collinear variables on the response.

This R Notebook seeks to ilustrate some of the difficulties that can be result from a collinearity.

In this RNotebook we’ll read a TCX and GPX files, used to track physical training and exercises evolving GPS and paths used by some workout Mobile Apps and Devices. Particularly we’ll will process one TCX file containing a MTB ride mine and transforming the a useful R data.frame ploting the ride track over a map.