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…

Getting Started

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.

…

Finally: After countless hours, you finished this project. Time after time, you traversed GitHub repositories, scrolled through StackOverflow posts, and tried to decipher TensorFlow’s cryptic error messages.

From one moment to another, the project is done; you’ve done your part. Where before you would spend each day hands down in…

TensorFlow’s TFRecord format is handy. However, for the first few times, it’s inconvenient to handle. While many datasets come readily from TensorFlow Datasets, some must be parsed manually. The ESC-50 dataset, short for Environmental Sound Classification, is one of them.

Dataset description

The ESC50 dataset consists of 2000 individual audio files, and…

About five weeks ago, I switched my main computer hand. Typically, as a right-handed user, I’ve got my computer mouse or trackpad to the right. But, partly out of interest, and partly because my left arm was always stuck with hovering over the keyboard, I decided to change that:

Trackpad…

It’s a paradox situation: You go to school, which prepares you for college. You then study towards a degree. With this degree, you then get a job. Once you’re settled, you work a) to not get fired, or b) to be promoted, or c) to get a raise so you…

Without a doubt, the ImageNet dataset has been a critical factor in developing advanced Machine Learning algorithms. Its sheer size and many classes have been challenging to handle. These problems led to better data-handling tools and novel Neural Networks architectures.

TensorFlow Datasets is such a dataset handling tool. With its…

A recent study proposed a Tiffany Mood Forecast Model (TMFM) that allows users to predict the mood of people named Tiffany. In this work, I study this model and point out its major weakness. From that, I show how simple multiplication with zero surprisingly makes ground for a generalized formulation…