One function — many possibilities. How to use linear algebra for deep learning in a clear and simple way. — Linear algebra plays a fundamental role in the field of deep learning. It is always about shapes, transpose, etc. Libraries like PyTorch, Numpy, and Tensorflow offer a lot of functions for this. But you may forget one or the other or confuse a function with one from another library. Even…

A short glance at promising developments of NLPs state-of-the-art architecture — It is 2017, and the paper “Attention is all you need” provides NLP a new promising architecture: the Transformer. Four years later – an eternity in the field of deep learning – the Transformer is still a state-of-the-art architecture and is used for various NLP tasks. But the world did…

Data cleaning and data augmentation (not only) for text simplification — Introduction This is the second article about my project of text simplification using transformers. If you have not read the first one, feel free to do so, but it is not mandatory to understand this article. Feel free to check out the GitHub repo for this project. Data preprocessing is not…

Learning deep learning using transformers for text simplification — Language is surrounding us every minute of every day. One prerequisite of it is its understandability. So without understanding each other, language loses its purpose. It is not a coincidence that clean and clear speech and language are highly reputable. Only with that we can share knowledge and ideas over…