Best Machine Learning Resources 3
This post is in construction. Subscribe to our paid newsletter to see the latest trend and tutorials in Machine Learning, Deep Learning. Our best, high quality contents, notes from conferences are for paid subscribers only. There is a free subscribe tier as well. Subscribe for free or Subscribe as a premium member today!
Subscribers, please remember to check the website for the latest articles. We avoid sending all articles to your inbox so not spam you. But the website and many articles are updated and created daily!
Machine Learning
High quality visualization of branches of machine learning algorithms
Structure a capstone project, a strong portfolio using Udacity’s capstone project guideline . Perhaps one of the most important concept is performance benchmarking against existing, leading or alternative algorithms. See the section on benchmark.
Deep Learning Frameworks
Machine learning, specifically deep learning frameworks like Tensorflow and Pytorch usually provides model performances and benchmarks. It can in terms of an error metric or time performance. Here’s the top 1 top 5 error rate of pytorch torchvision models on the ImageNet dataset (an industry standard). Here’s a timing benchmark. The github repo is still a work of progress.
Pytorch versus Tensorflow
This month our theme is Pytorch versus Tensorflow
"In a recent survey—AI Adoption in the Enterprise, which drew more than 1,300 respondents—we found significant usage of several machine learning (ML) libraries and frameworks. About half indicated they used TensorFlow or scikit-learn, and a third reported they were using PyTorch or Keras." O'Reilly
Post in construction more content coming soon. Become a subscribe today to see our past resource list! Become a subscribe to get notified about the best machine learning datasets, news about competitions and scholarships. In the subscriber only version of this newsletter we will discuss methodologies of the O’Reilly survey, providing additional insights in the machine learning world.
Data Science
Learning SQL? Try this SQL playground by w3school
Harvard class on Machine Learning (R programming) free online notes
Data Visualization Seaborn visualization using hue, side-by-side vertical bars. Read on our Medium Use incognito mode to read for free. Any paid subscriber can request a copy of the article or link. Email us hi@uniqtech.co from your substack subscriber email.
Feature engineering: undo the effect of a MinMaxScaler. Often after data analysis, we want to examine the original data. Use the inverse_transform function of sklearn of the scaler object to undo a transformation. Some transformations are not loss-less, meaning inverse_transform will not fully restore the original data. PCA can be lossy.
Technical Interviews
Did you know Salesforce has extensive research in Natural Language Processing? NLP Check it out. NLP Is the bread and butter for Salesforce AI. Also a core component of its einstein.ai product.
It is important to look beyond tensorflow and pytorch for your deep learning frameworks. We saw that some job postings still mention MXnet such as this senior data scientist posting from Amazon “Use Deep Learning frameworks like MXNet, Caffe 2, Tensorflow, Theano, CNTK, and Keras to help our customers build DL models” - Amazon job posting.
Coming soon for our paid users: End-to-End Machine Learning Course parts for Jupyter Notebook (under construction) and Python for Data Science.
Software Engineering, Data Engineering for Data Scientists: how much python programming should you know? The Udacity Machine Learning Engineer Nanodegree Syllabus highlights some important skills for data scientists for 2019
You can get the full syllabus on its landing page.
Upload an existing repository to github. If a git history has never being created for this repository you can use this guide here. Two takeaways here: 1. github repositories are usually public. Only premium accounts have private repositories. It is easy for other users to copy your code using a feature called Forking. Watch out. 2. git add remote basically adds your remote repository “URL” in https://.git format to your local repository. Once added local repository knows where to push your local change “uploading” local changes to remote aka github.
Self driving car demo by Waymo (click here). Thinking about getting started with self-driving cars? You can learn the basics using this free Udacity and AWS Deep Racer course.
Free and premium tier available.