Examine the effect of augmentations in your browser — When working with image data, practitioners often use augmentations. Augmentations are techniques that artificially and randomly alter the data to increase diversity. Applying such transformations to the training data makes the model more robust. For image data, frequently used candidates are rotating, resizing, or blurring. The effects of the transformations…

A guide to the evolution of generative networks — Introduction In short, the core idea behind generative networks is capturing the underlying distribution of the data. This distribution can not be observed directly, but has to be approximately inferred from the training data. Over the years, many techniques have emerged that aim to generate data similar to the input samples. …

Ask three questions to be better the next time — Comedian Chris Rock comes up with novel ideas in small steps. Before his hugely successful shows, he spends as many evenings as possible in small comedy clubs and tries out new jokes. Of course, most of them fall flat, and the audience laughs at him. But there will inevitably be…

Cal Newport, Greg McKeown, Dan Millman, Malcolm Gladwell — April began with typical — if we dare to use this word given the current circumstances — April weather. Some sun, some rain, a little bit of snow. Thankfully, we readers are flexible: If it rains or snows, we read beside the warm fireplace; if the Sun sends her greetings…

Lessons learned after ten thousand experiments in — Four months is a long time. But, so is ten months. Generally, everything further away than four weeks is in the future. This situation can be a blessing and a curse for machine learning projects. A blessing because we have ample time to finish the project. …

Mark Twain, Adler & van Doren, Stephen Hawking, Barbara Oakley — This March 2022, I finally finished a challenging read, How To Read A Book, by Mortimer J. Adler and Charles van Doren. Besides this, I also traveled to old America by reading Mark Twain’s The Adventures Of Tom Sawyer and The Adventures of Huckleberry Finn. How To Read A Book, by Mortimer J. Adler and Charles van Doren This book has been on…

Randomly rotating images is simple; here’s how to do it — In spirit to a previous post, where the topic was augmenting audio data, today, we focus on augmenting image data. …

TensorFlow’s format might not be easy to start with, but it is great for model training — The TFRecord format is Google’s data format for efficiently storing and reading data. While the underlying idea is simple — bundle objects into boxes —, the first steps are somewhat cryptic. …

Cal Newport, Scott Young, Robert Hassan, Joel Stein, Michael Harris — The choices of this month’s (February 2022) reading list might be seen as a continuation of the overarching theme begun the previous month: How technology affects our lives. Digital Minimalism, by Cal Newport Last month, in January, I had the pleasure to read two books by Calvin Newport, Deep Work and So Good They Can’t…