Skip to main content

Tracto solutions guide

Welcome to this guide on working with Tracto!

This notebook serves as a guide to working with Tracto. It contains links to other notebooks that demonstrate how to solve typical tasks related to YTsaurus. Each notebook offers step-by-step examples to help you navigate and make the most of YTsaurus.

Additionally, a small Index section is included to address common questions and provide quick references.

How to use examples​

  1. Copy notebook to your home directory //home/\<login\> or to //tmp
  2. Open your copy.
  3. Activate Run all button.

Screenshot 2024-09-20 at 14.10.44.png

Introduction​

  1. Jupyter Notebooks
  2. python sdk

Data processing​

  1. Basic data preparation using the Central Park Squirrel dataset. Would squirrels be more active in colder temperatures?

Managing data​

The notebooks below demonstrate how to work with data and metadata on YTsaurus.

  1. Files
  2. Documents
  3. Tables

Upload data​

  1. Upload small table

Tractorun​

  1. Basic torch mnist example
  2. Advanced torch example

Inference

  1. Offile inference

Jupyter Notebooks​

  1. Demo

SPYT​

  1. SPYT in Jupyter

Administration and management​

  1. Cookbook

Index​

How to work with table schemas?​

  1. Schema object
  2. Schema and yt_dataclass
  3. Mapreduce operations on tables with schema

How to work with datetime fields in schema?​

  1. Date
  2. Datetime

How to join tables?​

  1. By reduce operation

How to upload pytorch dataset?​

  1. MNIST

How to run SPYT standalone?​

  1. spark-launch-yt

How to use pyspark?​

  1. In Jupyter

How to create a new user?​

  1. cli

How to create a new group?​

  1. cli

How to set user's password?​

  1. cli