MLP BLOG

f1 score from eigenvectors assuming positive curvature outperforms euclidean metrics, which gives a hint of reality in a curved space or non

Data analysis with Sklearn for stock prices, new features for the new approach.

Data is not 2D, 3D or 4D, it is more complicated and difficult to visualize

F1 score get a boost of up to 79% for the most important class in a support vector machines classification problem. Positive curvature of no

In part I it was explained that a first sight at the f1 scores -considering features as in a classic 3D space against a non-euclidean one- g

The Hyperspace is not Always Euclidean: Study of f1 Score with Sklearn Metrics, Grid Search - Part III (Google Colab)

...this sparse matrix that contains all classes with the most influence, is compared against the 'y_true' that needs also to be...

Each frequency bin is treated as coordinate feature, then, the hyperplane has as many dimensions as classes are.

In the prior post it was proven that, as a stochastic variable, a stock quote is impossible to be predicted using regression. It is based on