Search Results for "beam"

Beam is especially useful for Embarrassingly Parallel data processing tasks, and caters to the different needs and backgrounds of end users, SDK writers and runner writers.

Phoenix is a high-performance, productive web development framework written in Elixir. It runs on the Erlang VM (BEAM). It is designed to support both traditional request/response web applications. It also supports real-time, soft-real-time applications via WebSockets, channels, PubSub, and presence features. Phoenix emphasizes fault tolerance, scalability, and developer productivity. It provides tools like code generators, LiveView integration, templating, routing, and a flexible plug pipeline. ...

Mix tasks to simplify use of Dialyzer in Elixir projects. Elixir 1.6 is required, to support the new pretty printing feature. If your project is not yet on 1.6, continue to specify 0.5 in your mix deps. Warning messages have been greatly improved, but are filtered through the legacy formatter to support your existing ignore files. You can optionally use the new Elixir term format for ignore files. You may want to use the --format short argument in your CI pipelines. There are several...

BentoML simplifies ML model deployment and serves your models at a production scale. Support multiple ML frameworks natively: Tensorflow, PyTorch, XGBoost, Scikit-Learn and many more! Define custom serving pipeline with pre-processing, post-processing and ensemble models. Standard .bento format for packaging code, models and dependencies for easy versioning and deployment. Integrate with any training pipeline or ML experimentation platform. Parallelize compute-intense model inference...

membrane_core is the foundation of the Membrane multimedia framework for Elixir, providing the abstractions and runtime needed to build real-time audio and video pipelines. It models media processing as a graph of lightweight, supervised OTP processes—elements connected by links—so work is isolated, fault-tolerant, and easy to scale or reconfigure at runtime. The core defines a clear lifecycle and callback API for elements, plus concepts like buffers, events, and capabilities/format...

desktop enables building cross-platform desktop applications with Elixir by pairing a Phoenix/LiveView UI with a native webview shell. The approach keeps application logic on the BEAM—supervised, fault-tolerant, and hot-reload-friendly—while rendering an HTML/CSS/JS interface inside the system’s embedded browser engine. It offers conveniences for packaging and distribution on Windows, macOS, and Linux, including app metadata, icons, and startup integration. The library exposes desktop-specific affordances such as system tray menus, window management, and notifications, so applications feel native rather than like generic web wrappers. ...

seq2seq is an early, influential TensorFlow reference implementation for sequence-to-sequence learning with attention, covering tasks like neural machine translation, summarization, and dialogue. It packaged encoders, decoders, attention mechanisms, and beam search into a modular training and inference framework. The codebase showcased best practices for batching, bucketing by sequence length, and handling variable-length sequences efficiently on GPUs. Researchers used it as a baseline to reproduce classic results and to prototype new attention variants and training tricks. It also offered scripts for data preprocessing, evaluation, and exporting models for serving. ...

The X Beam ("the cross beam") is a cross platform C++ 3D world simulation engine with powerful customization support. It is currently being developed for Linux, Mac OS X, and Windows.