Data Science Digest

2021-06-12 09:00:16 Tinkering with the Mobile Apps Dataset

In this article, the author demonstrates how you can use an open-source dataset featuring mobile apps data to build your own models. The article includes such steps as choosing a dataset, exploratory data analysis, feature engineering, and predicting with a model. The dataset and the models are available for re-use.

https://bit.ly/3xiMQT0 Open / Comment

2021-06-11 17:00:12 #DataScienceMemes Open / Comment

2021-06-11 12:00:24 Metric Learning Tips & Tricks

In this article, the author presents ways of overcoming the limitations of classification, such as the number of training samples, production integration, and scaling. Specifically, he’ll explain how to train an object matching model with no labeled data and use it in production, to ensure metric learning is more scalable and flexible.

https://bit.ly/3cznb05

Subscribe to our weekly newsletter — https://bit.ly/3gqPUp5 Open / Comment

2021-06-10 13:58:44 DataScience Digest pinned a photo Open / Comment

2021-06-10 12:20:06 ​​Data Science Digest — 10.06.21

The new issue of DataScienceDigest is here! Machine learning in healthcare, the top 10 TED talks on AI, fraud detection in Uber, DatasetGAN, Text-to-Image generation via transformers, and more…

https://bit.ly/2TR87o9

Join @DataScienceDigest Open / Comment

2021-06-09 09:30:08 Introducing Orbit, An Open Source Package for Time Series Inference and Forecasting

Orbit (Object-ORiented BayesIan Time Series) is a general interface for Bayesian time series modeling developed by Uber Engineering. In this article, you’ll learn the ins and outs of Orbit, from the basics and use cases to a tutorial and benchmarks to follow. Uber is going to introduce more dedicated Bayesian time series models, so the project is worth a look.

https://ubr.to/3v6Hbxy Open / Comment

2021-06-08 09:30:06 Long Text Generation by Modeling Sentence-Level and Discourse-Level Coherence

Generating long and coherent text is an important but challenging task. In this paper, the authors propose a long text generation model that represents the prefix sentences at sentence level and discourse level in the decoding process. Extensive experiments show that the model can generate more coherent texts than state-of-the-art baselines.

Paper — https://bit.ly/3uZvvfZ
Code — https://bit.ly/3vXsj61 Open / Comment

2021-06-07 09:00:14 DatasetGAN: Efficient Labeled Data Factory with Minimal Human Effort

DatasetGAN is an automatic procedure to generate massive datasets of high-quality semantically segmented images requiring minimal human effort. Presented by an international team of researchers, it outperforms all semi-supervised baselines and is on par with fully supervised methods using labor-intensive annotations.

Web Page — https://bit.ly/2T8S5WA
Paper — https://bit.ly/2SfcDMP
Code — Coming Soon
#DataScienceDigest #DataScience #MachineLearning #ArtificialIntelligence #AI #ML

Subscribe to our weekly newsletter — https://bit.ly/3fXLuXW Open / Comment

2021-06-06 10:00:13 A Checklist to Track Your Data Science Progress

Progress is fickle. You may think that you are moving forward while, actually, being stuck in the repetition rut. That’s why you need to have a system to track your progress; for example, you can use this awesome checklist by Pascal Janetzky. Get an overview of your progress and find the next goal just by following these steps.

https://bit.ly/3gct7gK Open / Comment

2021-06-05 10:00:09 Airflow and Ray: A Data Science Story

In this article, you’ll learn about a Ray provider for Apache Airflow. Ray is a Python-first cluster computing framework that allows Python code, even with complex libraries or packages, to be distributed and run on clusters of infinite size, enabling fast transformations of Airflow DAGs into scalable machine learning pipelines.

https://bit.ly/3pp5Ufe Open / Comment