Introduction

C++ implementations of Deep Lake optimized for faster data fetching and computations

Overview of Deep Lake's High-Performance Compute Engine built in C++

Compute Engine offers high-performance implementations of compute-heavy Deep Lake features, such as distributed dataloading, large queries, and indexing.

The engine is built in C++, the user interface is still in Python, and it is installed from compiled C++ binaries using: pip install "deeplake[enterprise]",

In order to use Compute Engine, Deep Lake data must be stored in Deep Lake Storage, or in the user's cloud while being connected to Deep Lake using Managed Credentials.

Features Optimized in the Compute Engine:

pagePerformant DataloaderpageTensor Query Language (TQL)pageIndex for ANN SearchpageManaged Tensor Database