Skip to main content

Welcome to the Playground Datasets, where you can explore and use annotated images stored in ReductStore's Time Series Object Store. These datasets are hosted on a remote server accessible via https://play.reduct.store. You can freely access and download these datasets using the access token "reductstore".

Currently, you can browse the following demo datasets:

  • IMDB Ratings: A dataset containing images related to IMDB ratings.
  • Cat Images: A collection of various cat images.
  • MNIST Handwritten Digits: A popular dataset of handwritten digits.

Each of these datasets comes with labels, making them ideal for training and testing deep learning models. You can use these datasets to build and tune your models using frameworks like PyTorch. For more guidance, check out our in-depth blog post on implementing data streaming from a remote database in PyTorch, which walks you through the entire process.

To learn more about the datasets or get the code to access them, you can visit the dataset repository.

ReductStore isn't just for storing images; it's designed to manage blob data that changes over time. This makes it ideal for storing and organizing information by specific moments or in a step-by-step order. That's why the slider below is called Timeline. Just like a timeline in history, it allows you to move through the data in the order it was added, from the first item to the last.

If you have any questions, suggestions, or need additional datasets, feel free to contact us or join our community forum.

    1 - 10