
2020 is out!
In a complicated, recessive year, and affected by the COVID-19 pandemic that has changed the world a little more, Kreilabs is born.
In a complicated, recessive year, and affected by the COVID-19 pandemic that has changed the world a little more, Kreilabs is born.
The quality of that dataset is directly related to the organization that manufactures it and how that organization works on data quality.
The sustainability of our model could also be related to the quality of data in the organization.
Beyond data engineering, the accuracy of the model itself, and the performance in its production, if the organization it serves does not have data quality as a priority, probably the model will not be sustained over time.
We apply three different techniques of hyperparameter optimization on a classification model set to compare their accuracy.
We explore platform.ai for image labeling, clustering, and eventually create a model into the tool for the full tag dataset. Link to the full text
Try an image classification model with an unbalanced dataset, and improve its accuracy through data augmentation techniques. Link to the full text in
On December 14, we made the annual balance and projection of the company, defining the strategic plan for the future.
We present our case of success in classifying user tickets by applying different techniques on natural language processing. With our model, we can determine the
In the previous article, we take a metrics guide for regression problems. Now we are going to look at the metrics most used for classification problems
In a series of three articles, we will describe a set of basic statistical concepts and the metrics used for regression and classification Link to
We analyze the impact of classifying movie reviews sentiments based on a language model trained from scratch, or a pre-trained model using the corpus wikitext-103