Search Results for "computations"

This repository gathers Google’s practical tooling for Fully Homomorphic Encryption (FHE), focused on making it possible to run computations on encrypted data without ever decrypting it. At its core is a “transpiler” that converts ordinary functions (typically written in a restricted subset of C++ or similar) into FHE circuits, plus backends that execute those circuits with different FHE libraries. The workflow usually mirrors normal software development: write and test a cleartext implementation, run a simulator to validate logic and performance characteristics, then compile to an encrypted form to run with real FHE parameters. ...

...Each of these is composable and avoid the addition of “special” extensions by being as general as possible. Koka tracks the (side) effects of every function in its type, where pure and effectful computations are distinguished. The precise effect typing gives Koka rock-solid semantics backed by well-studied category theory, which makes Koka particularly easy to reason about for both humans and compilers. Effect handlers let you define advanced control abstractions, like exceptions, async/await, or probabilistic programs, as a user library in a typed and composable way. ...

...It is a statically typed, data-parallel, and purely functional array language in the ML family, and comes with a heavily optimizing ahead-of-time compiler that presently generates either GPU code via CUDA and OpenCL, or multi-threaded CPU code. Futhark is not designed for graphics programming, but can instead use the compute power of the GPU to accelerate data-parallel array computations. The language supports regular nested data-parallelism, as well as a form of imperative-style in-place modification of arrays, while still preserving the purity of the language via the use of a uniqueness type system. While the Futhark language and compiler is an ongoing research project, it is quite usable for real programming and can compile nontrivial programs which then run on real machines at high speed.

Apache TVM is an open source machine learning compiler framework for CPUs, GPUs, and machine learning accelerators. It aims to enable machine learning engineers to optimize and run computations efficiently on any hardware backend. The vision of the Apache TVM Project is to host a diverse community of experts and practitioners in machine learning, compilers, and systems architecture to build an accessible, extensible, and automated open-source framework that optimizes current and emerging machine learning models for any hardware platform. ...

...It is meant to take the best features of multiple paradigms and combine them all into a single, easy-to-use language, suitable for professionals and hobbyists alike. (The basic OOP paradigm, with metaprogramming features, will be implemented first.) Its primary goal is to allow very heavy computations to be performed efficiently and easily, even by scientists from other fields than computer science. For more information about the language, see the Wiki.

...This model introduces two key concepts -- places and asynchronous tasks -- and a few mechanisms for coordination. With these, APGAS can express both regular and irregular parallelism, message-passing-style and active-message-style computations, fork-join and bulk-synchronous parallelism. Both its modern, type-safe sequential core and simple programming model for concurrency and distribution contribute to making X10 a high-productivity language in the HPC and Big Data spaces. User productivity is further enhanced by providing tools such as an Eclipse-based IDE (X10DT). ...

NimTorch is a deep learning library for the Nim programming language, providing bindings to PyTorch for efficient tensor computations and neural network functionalities.

A domain-specific compiler that facilitates the generation of high-performance parallel programs for a class of scientific computations encountered in chemistry and physics.