top of page

MLP BLOG
​
​
​
GET UP TO DATE WITH A NEW PERSPECTIVE ON MACHINE LEARNING, DATA SCIENCE, AND 3D PRINTING
​
​
Ara


Alejandro Romero
16 Åžub 20244 dakikada okunur
The Hyperspace is not Always Euclidean: f1 Scores in Positive/Negative Curvature Using Eigenvectors and Sklearn - Part VI
f1 score from eigenvectors assuming positive curvature outperforms euclidean metrics, which gives a hint of reality in a curved space or non
1 görüntüleme0 yorum

Alejandro Romero
16 Åžub 20241 dakikada okunur
The Hyperspace is not Always Euclidean: UPDATED - Study of f1 Score with Sklearn Metrics - Part V (Colab)
Study of f1 Score with Sklearn Metrics in Curved Surfaces. Non Euclidean Geometry.
1 görüntüleme0 yorum


Alejandro Romero
9 Eki 20232 dakikada okunur
Data is Nor 2D Neither 3D but 4D... or Higher
Data is not 2D, 3D or 4D, it is more complicated and difficult to visualize
3 görüntüleme0 yorum


Alejandro Romero
2 Eyl 20232 dakikada okunur
The Hyperspace is not Always Euclidean: f1 Score in Positive Curvature Fits Best Data - Part IV
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
2 görüntüleme0 yorum


Alejandro Romero
31 Tem 20232 dakikada okunur
The Hyperspace is not Always Euclidean: Study of f1 Scores with Sklearn Metrics - Part II
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
4 görüntüleme0 yorum


Alejandro Romero
2 May 20233 dakikada okunur
The Hyperspace is not Always Euclidean: Study of f1 Score with Sklearn Metrics - Part I
...this sparse matrix that contains all classes with the most influence, is compared against the 'y_true' that needs also to be...
6 görüntüleme0 yorum
bottom of page