Kasım 19, 2019 —
When the TensorFlow YouTube channel launched in 2018, we had a vision to inform and inspire developers around the world about what was possible with Machine Learning. With series like Coding TensorFlow showing how you can use it, and Made with TensorFlow showing inspirational stories about what people have done with TensorFlow and much more, the channel has grown greatly. But we learned an import…
Portuguese versions of ML Zero to Hero are now available
Kasım 19, 2019
When the TensorFlow YouTube channel launched in 2018, we had a vision to inform and inspire developers around the world about what was possible with Machine Learning. With series like Coding TensorFlow showing how you can use it, and Made with TensorFlow showing inspirational stories about what people have done with TensorFlow and much more, the channel has grown greatly. But we learned an important lesson: it’s a global phenomenon, and to reach the world effectively, we should provide some of our best content in multiple languages with native speakers presenting. So today we’re delighted to launch the first wave of this -- the popular Zero to Hero series in Portuguese!
Machine Learning com TensorFlow: De Zero a 100.
Parece que hoje em dia é impossível abrir um navegador, jornal ou livro sem ver algo sobre Machine Learning ou IA. Há muita informação e muita publicidade. Com isso em mente, Laurence Moroney, da equipe TensorFlow, queria produzir uma série de vídeos em quatro partes, da perspectiva de um desenvolvedor, sobre o que realmente é o Machine Learning. Baseado em uma palestra bastante popular do Google IO 2019 chamada "Machine Learning: Zero to Hero with TensorFlow", a série de vídeos está disponível em português!
No vídeo 1, você aprenderá que o Machine Learning representa um novo paradigma na programação. Em vez de programar regras explícitas em uma linguagem como Java ou C ++, você constrói um sistema treinado em dados para inferir as próprias regras. Mas como é realmente o ML? Aqui você encontrará um exemplo básico do Hello World de construção de um modelo de ML, apresentando idéias que aplicaremos nos episódios posteriores a um problema mais interessante: a visão por computador.
No vídeo 2, você conhecerá a visão computacional básica com o aprendizado de máquina, ensinando um computador a ver e reconhecer objetos diferentes. Você também pode trabalhar com um exemplo aqui: https://goo.gle/34cHkDk
No vídeo 3, vamos discutir as redes neurais convolucionais e por que elas são tão poderosas nos cenários de visão computacional. Uma convolução é um filtro que passa por uma imagem, a processa e extrai recursos que mostram uma semelhança na imagem. Neste vídeo, você verá como eles funcionam, processando uma imagem para ver se você pode extrair recursos dela! Você também pode tentar um codelab: http://bit.ly/2lGoC5f
No vídeo 4, você verá como criar um classificador de imagens para pedra, papel e tesoura. No episódio um, mostramos um cenário de pedra, papel e tesoura e discutimos o quão difícil pode ser escrever código para detectá-lo e classificá-lo. À medida que os episódios progridem no aprendizado de máquina, aprendemos como criar redes neurais, desde a detecção de padrões em pixels brutos, até sua classificação e detecção de recursos usando convoluções. Neste episódio, colocamos todas as informações das três primeiras partes da série em uma. Caderno Colab: http://bit.ly/2lXXdw5. Conjunto de dados de pedra, papel e tesoura: http://bit.ly/2kbV92O
Esperamos que vocês gostem desta série e informe-nos se quiser ver mais!
This blog post is also available in English:
It seems you can’t open a browser, newspaper or book without seeing something about Machine Learning or AI. There’s a lot of information and a lot of hype. With that in mind, Laurence Moroney from the TensorFlow team wanted to produce a 4-part series of videos, from a developer’s perspective, about what Machine Learning actually is. It’s based on his popular talk from Google IO 2019, and called “Machine Learning: From Zero to Hero with TensorFlow”
Here’s video 1 where you’ll learn that Machine Learning represents a new paradigm in programming, where instead of programming explicit rules in a language such as Java or C++, you build a system which is trained on data to infer the rules itself. But what does ML actually look like? Here you’ll walk through a basic Hello World example of building an ML model, introducing ideas which we'll apply in later episodes to a more interesting problem: computer vision.
Here’s video 2 where you’ll walk through basic computer vision with machine learning by teaching a computer how to see and recognize different objects. You can also work through an example yourself here: https://goo.gle/34cHkDk
Here’s video 3 where we discuss convolutional neural networks and why they are so powerful in Computer vision scenarios. A convolution is a filter that passes over an image, processes it, and extracts features that show a commonality in the image. In this video you'll see how they work, by processing an image to see if you can extract features from it! You can also try a codelab!: http://bit.ly/2lGoC5f
Here’s video 4 where you’ll see how to build an image classifier for rock, paper, and scissors. In episode one, we showed a scenario of rock, paper, and scissors; and discussed how difficult it might be to write code to detect and classify these. As the episodes have progressed into machine learning, we’ve learned how to build neural networks from detecting patterns in raw pixels, to classifying them, to detecting features using convolutions. In this episode, we have put all the information from the first three parts of the series into one. Colab notebook: http://bit.ly/2lXXdw5. Rock, paper, scissors dataset: http://bit.ly/2kbV92O
We hope you enjoy this series, and please let us know if you want to see more!
Next post
Portuguese·
Portuguese versions of ML Zero to Hero are now available
Kasım 19, 2019
—
When the TensorFlow YouTube channel launched in 2018, we had a vision to inform and inspire developers around the world about what was possible with Machine Learning. With series like Coding TensorFlow showing how you can use it, and Made with TensorFlow showing inspirational stories about what people have done with TensorFlow and much more, the channel has grown greatly. But we learned an import…