This post will describe and explain maybe the most critical part of predictive modelling: how to correctly estimate the performance of a machine learning model. This is performed by setting up a trusted cross-validation framework. It’s crucial to get this right, otherwise your model performance estimates will not reflect the true model performance.
Last race of the season, last race for Fernando Alonso! Also last race for Kimi in a Ferrari and last race for Vandoorne, Ocon and possibly for some others… Both championships may have already been decided but the upcoming Abu Dhabi Grand Prix will be one to remember forever. One could write pages about Alonso’s career, its ups and downs, his wrong choices but also about his relentless passion, his consistency, his dominance! My model thinks he is one of…
Lewis Hamilton may have already clinched the driver’s championship however there’s still a lot to play for in the final two races of the season. The most important battle is the one for the positions 4 to 9 in the constructors championship where each team might get a lot of prize money if they are able to get higher in the standings.
“F***ing finally”! That’s what Kimi said after his surprising win in Austin, Texas. He had last won in 2013 with Renault while his last win with Ferrari was back in 2009 during his first stint with the team. Kimi also became the oldest Grand Prix winner since Nigel Mansell in 1994. In contrast, Vettel did yet another mistake and has virtually no chances to win the championship.
Lewis Hamilton could clinch the title in the United States grand prix this weekend if he outscores Vettel by eight points. Anyway, it is now a matter of when, not if – despite Mercedes’ Toto Wolff saying the opposite. In COTA, both drivers have won but Lewis has taken by far more wins than Seb. A fun fact is that all races in this circuit have been won from the front row of the grid.
The biggest news coming into the Japanese GP were about Raikkonen’s haikus. In case you don’t know what they are (like I did), please have a look at the link. Joking aside, Lewis Hamilton and Mercedes have played well the team game called ‘F1’ and are on their way to winning both championships again.
Russian GP started with the announcement that Antonio Giovinazzi will replace Ericsson in Sauber next season and will partner Kimi Raikkonen. Tomorrow Torro Rosso is also expected to present Daniil Kvyat as one of its drivers for next year while Brendon’s future is uncertain. Besides all this, the battle for the driver’s championship is still on. If Vettel wins all remainining races, he still gets the title.
The big news this week is that Kimi Raikkonen is moving to Sauber with a 2-year contract and Leclerc is taking his position in Ferrari for the next season. Although the move for Kimi was unexpected, it’s been rumored that he has bought part of the Sauber team; after all, it’s the team he started in F1 with and it’s base is right next to his home in Switzerland. It’s a smart move from Ferrari who want to have a…
You know what is the most geeky way to ‘prove’ that Alonso is the best F1 driver ever? It is Neural Embeddings! In this post I will try to give an intuitive explanation of what neural embeddings are, how they can be calculated and show some examples of how they capture semantic information about the objects they represent. In the end of this post, you will understand how all this relates to Alonso’s driving skills.
Monza, the temple of speed, is the next destination of F1; a place where Ferrari has not tasted the home win since Alonso took the victory in 2010. The latest engine upgrade showed some impressive signs and the Prancing Horse is hoping to capitalize on that in front of thousands of tifosi.