When training machine learning models, we often want to not only obtain a model with high accuracy but also want to know how important individual features for the importance of the model are. There are several reasons why we are interested in this. Let us say that we are building a regression ML model which […]
Author Archives: writer
Generating images with transfer learning
Generating images with transfer learning is really interesting. I’ve been playing with a new technique that has taken deep learning by storm: neural style transfer. Initially created by Leon A. Gatys, Alexander S. Ecker, and Matthias Bethge in 2015, neural style transfer is an optimization technique used to take three images, a content image, a […]
Product categorization and product tagging machine learning solutions
When you enter a physical store, you have tables above aisles that denote the category of products that are being sold in particular section. The online equivalent of this are the product paths that show the main category, subcategory, level 3 category, and so on. In earlier days of internet these categories were set manually […]
Crypto social media analysis
Social media has played an important role in driving the narrative around cryptocurrency sector in recent years. Although initial paper of Satoshi Nakamoto was published on forum posts, see e.g. satoshi nakamoto posts, in the later years the hype about cryptocurrencies was nevertheless substantially driven on social media, especially twitter. It is interesting that in […]
Data Visualization Consulting
There is an old saying that one picture is worth a thousand words and in the modern content marketing this is often true. The most viral posts that one encounters are often the ones where someone produces an interesting presentation of unique data set and its analysis. Another topic that also gets a lot of […]
Geo location of photos using deep learning
Computer vision is part of AI consulting tasks that often involve classification problems where one tries to train a deep learning neural net to classify a given image in one of discrete classes. Typical examples are for example classifying images of animals, food, etc. Classical problem from this set was classifying images as either cat […]