07-11, 12:00–12:30 (Europe/Amsterdam), If (1.1)
Last year I attended PyData Eindhoven for the first time. I got inspired and now I’m back to present my first neural network, a network that was trained to predict the Spring Classics of cycling! With this neural network, I’m attempting to beat my friends, and myself, in a well-known fantasy cycling game.
In this talk, I will elaborate on the process of building a model from scratch. This will include data collection, model training and finetuning, and of course a discussion of the predicted results. The predictions will also be compared to an existing cycling prediction platform that I use as benchmark. Lastly, I’ll try to provide some insights into the model using SHAP values.
None
I'm working as a software engineer at Royal HaskoningDHV, a Dutch consulting and engineering firm. Both professionally and as a hobby, I have been delving into some AI-related subjects. I'm happy to give my first lecture at PyData Eindhoven, to share about my deep dive into Machine Learning, combined with my passion for cycling.