PyTensorStore

PyTensorStore.jl provides a Julia wrapper around the Google TensorStore Python API. TensorStore is a library for efficiently reading and writing large multi-dimensional arrays, supporting various storage backends like Zarr, N5, and Google Cloud Storage.

Contents

• PyTensorStore
• • Contents
  • Key Features
  • Installation
• User Guide
• • Opening a Store
  • Basic Operations
  • Labeled Indexing
  • Transactions
  • Contexts
  • Domain Manipulation
• API Reference
• • Public API
  • Python Module Wrapper

Key Features

• Familiar Interface: Implements much of the Julia AbstractArray interface, including 1-based indexing, size, axes, and eltype.
• Asynchronous Support: Built on TensorStore's asynchronous core, allowing for non-blocking I/O.
• Powerful Indexing: Support for labeled indexing, coordinate translation, and complex domain transformations.
• Transactions: Support for atomic multi-write operations.

Installation

using Pkg
Pkg.add("PyTensorStore")

Note: This package requires a working Python installation with the tensorstore package. CondaPkg.jl is used to manage this dependency automatically.