Hermann Kekeli GOLO

Je viens de lancer la BETA de la plateforme🧪 J'ai besoin de vos avis et contributions teapaw.org/gausoft_/status/205173… cc @hlabsdev @Scarabea_Yeye

What's the "Hello World" of machine learning? Here are 4 problems I'd recommend if you want to start from the beginning: Problem 1 The Titanic Challenge in Kaggle is among the most popular problems for beginners. This is a classification problem that deals with structured data. It will show you the impact of feature engineering. Problem 2 For a regression problem, solve the House Pricing challenge in Kaggle. The cool thing about these problems is that thousands of people have solved them. You'll never be stuck! Problem 3 Handwritten Digit Recognition using the MNIST dataset. This is a classic, and it's a nice transition to start dealing with images and learning about neural networks. You can solve this problem with a simple, fully connected network. You can also use a convolutional neural network. Problem 4 To start with NLP, look at the Chinese Word Segmentation problem. Your job is to split a sequence of Chinese characters into words. There are many problems out there. When starting, look for popular problems with many solutions and explanations. They will give you the best chance to learn.

Discord has announced that it has scaled Elixir to 5 million concurrent users. The platform, which is primarily used for gaming, chat and VoIP, says that the figure represents a "significant milestone" for the company blog.discordapp.com/scaling-…