I have worked with Python, Deep Learning and Computer Vision for almost 5 years, in both academic research and industrial settings. I had started with Jupyter Notebook but quickly switched to Jupyter Lab because of several reasons. And I loved using it until I started to use VS Code for programming. I am a big fan of writing proper CLI programs than creating notebooks for most tasks. However, for data analysis and quick visualizations, nothing beats the interactive nature of the Jupyter cells!

The thing about machine learning is that almost anyone can do it successfully. It doesn't take much to get a Jupyter notebook and work on scraped data. You can get into the Jupyter lab environment to do almost anything you want in the R programming language. However, how do you actually transition from a Jupyter notebook to the real world? The question of operationalizing machine learning is typically where most engineers get stuck.

JupyterLab is fundamentally intended to be an extendable environment. Any component of JupyterLab can be enhanced or customized using JupyterLab extensions. New themes, file viewers and editors, or renderers enabling rich outputs in notebooks are some of the things they can offer. Keyboard shortcuts, settings in the system, and items to the menu or command panel can all be added via extensions. Extensions can depend on other extensions and offer an API for use by other extensions.

If you are looking for a fast and quick introduction to python machine learning, then this course is for you. It is designed to give beginners a quick practical introduction to machine learning by doing hands-on labs using python and JupyterLab. I know some beginners just want to know what machine learning is without too much dry theory and wasting time on data cleaning. So, in this course, we will skip data cleaning. All datasets is highly simplified already cleaned, so that you can just jump to machine learning directly. Machine learning (ML) is a type of artificial intelligence (AI) that allows software applications to become more accurate at predicting outcomes without being explicitly programmed to do so.

In this short tutorial, we are going to look at the very cool and interesting image-based 3D human shape estimation model. We are going to implement a pipeline on Spell workplace with a jupyter lab and run the pre-trained model to turn our custom image into a 3D model. Signup to the Spell if you haven't and you can get 10$ with of GPU time on T4, P100, k80, and V100 GPUs for free. We are going to use Facebook research's pifuhd model which contains a PyTorch implementation of "Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization". Note: At least 8GB GPU memory is recommended to run the PIFuHD model. We can use any GPU from Spell.

Imagine that you are working on a project, with a team of 10 people. All members of this team, have to work from home now, because of the ongoing pandemic, so all of them have different laptops, different system specifications, different operating systems, etc. Now one fine day, a team member pushes a new change to GitHub, that adds some new functionality to your project. Unfortunately, these new changes do not work for some people, maybe because of different versions of the software installed on the different computers. So you have a very common problem, that many teams often face. "It works for him, but not for me" Docker was made specifically to solve this problem.