Into the Unknown
Machine Learning in Production deeplearning.ai Designing an end-to-end production machine learning system is a challenge. This is rarely done by any single individual, but a well-collaborated team is required to achieve the goal and make the system keep running. The lifecycle is pretty much like the one employed for software project, however obviously there are…
Generative Pre-trained Transformers (GPT) University of Glasgow Scotland has been on my must-visit wish list for a long time. On my mind, it is full of myths, folklore and legends. When I found this course on Coursera, I enrolled in the blink of my eyes. This course is succinct, but informative. Should you have some…
GPT stands for generative pretrained transformers. It is a type of language model which has been trained on a very large amount of text. It is capable of generating highly fluent text, making it applicable to a variety of tasks. “Generative” means GPT can generate something for you. “Pretained” refers to the fact that the…
DevOps, DataOps, MLOps Duke University If you are kinda fed up with training models and would like to try some new perspectives and build some new skills, MLOps is a very good point to start with. I personal felt like falling down to a rabbit hole when I took this course. Nowadays, the ecosystem for…
You probably have known the DevOps methodology which is very commonly used in the software industry, which basically deploy your infrastructure programmatically, do continuous integration, test your code automatically, deploy the software continuously in production. Overall the software is constantly being updated with new features and then you have the ability to make changes in…
Python Essentials for MLOps Duke University You will learn a lot from this course: Python language itself, PyTest , Pandas, Numpy, a few Frameworks and Cloud APIs. I can not agree more that some many skills together are needed to succeed in an MLOps role. The course first covers basics of programming languages (like data…
Machine Learning: Algorithms in the Real World Alberta Machine Intelligence Institute It is really an exciting journey! This specialization shows you a lot that you probably never imaged as an engineer. A successful machine learning project is far from the machine learning itself. You need to see farther, think bigger and experiment more. Machine learning…
Optimizing Machine Learning Performance Alberta Machine Intelligence Institute In the real world, the journey to a successful machine learning project does not start with cool ideas and technology. At least, there are thing way more important then “letting your servers crunch data and get results”. The first question is to ask: “As an organization, where…
Integrating Machine Learning Systems Issues arise from putting your machine learning model into a real life system. To actually make use of your model, you need a way to interact with it, which creates complications you can’t address simply by testing the model in isolation. Machine Learning is all about data. Each model in production…
Although many organizations recognize the importance of machine learning for making the most of their businesses, very few actually know how to get started. It is also required to develop a machine learning strategy for the business, moving beyond individual successes into a long-term growth plan. Machine learning impacts entire business from processes and staff…
Probabilistic Deep Learning with TensorFlow 2Imperial College London The focus of this course is the TensorFlow Probability library. Spoiler alert! Probability distributions are important factors you need to consider. From now on, building model is not only as simple as adding layers and squeezing your GPU to calculate various weights. This is challenging course. From…
The Variational Autoencoder (VAE) is an algorithm for inference and learning in a latent variable generative model. In it’s simplest form, it’s an unsupervised learning algorithm and like normalizing flows, the generative model can be used to create new examples similar to the data set. However, unlike normalizing flows, the generative model is not invertible…
Data for Machine LearningAlberta Machine Intelligence Institute Do you agree that the skills of manipulating data is more important than building fancy models? I do believe so. Raw data is almost never ready to use, and they are only valuable only if being extracted and transformed properly. “Garbage in, garbage out”. Moreover live data expose…
There are many ways that data can go wrong, sometimes through no fault of its own. Imbalanced Data A dataset with skewed class proportions where the vast majority of your examples come from one class is called an imbalanced dataset. Not surprisingly, having imbalanced classes in your learning data impacts the model that results. You…
Having a deep understanding of data is an essential prerequisite for doing EDA (Exploratory Data Analysis) as well as feature engineering. Very often, only certain types of feature engineering techniques are valid for certain types of data. Why do we need feature engineering? In order to visualize it better and to use it as features…