The competition had commenced, and I was the tenth on the list.

In 2013, I decided to try a new goal. I have a bucket list, and after another loss, I am completing my evolving adventure list. I always wanted to participate in a bodyball competition. I approached my daughter and said, "How about Eddie, do you want to join the fitness contest," she said. Trophy "Although the pressure is great, I named her" a small coach "and talked to her what I can eat, what I can not eat and let her participate in every step .

Over the next few years I participated in several games. I never succeeded in spending as much time as possible, but each game was a wonderful learning experience. One of the goals I set in 2017 was to choose the Kaggle competition and concentrate on winning my first gold medal completely. The competition I chose was the planet: Learn Amazon from outer space. I already have some experience with satellite imagery, and the use case seems interesting. This is a wonderful choice and I found that it will bring more results in the coming months.

For those unfamiliar with the planet's competition, the purpose of the game is to track footprints of people in the Amazon rainforest. Satellite images are provided by Planet at a resolution of 3 meters - meaning that each pixel resembles 3 meters. This allows small changes such as selective recording to be detected. The contest is set to a multilabel classification problem and the metric used is the F2 score. Labels include selective logging, agriculture, major tropical rainforests, clouds, roads and so on.