Skip to content
/ Boltz.jl Public

Accelerate your ML research using pre-built Deep Learning Models with Lux

License

Notifications You must be signed in to change notification settings

LuxDL/Boltz.jl

Repository files navigation

Boltz ⚡

GitHub Discussions Latest Docs Stable Docs

CI Build status codecov

Downloads Downloads

JET Testing Aqua QA ColPrac: Contributor's Guide on Collaborative Practices for Community Packages Code Style: Blue

Accelerate ⚡ your ML research using pre-built Deep Learning Models with Lux.

Installation

using Pkg
Pkg.add("Boltz")

Getting Started

using Boltz, Lux, Random
using JLD2 # Needed to load pretrained weights

model = Vision.AlexNet(; pretrained="ImageNet1K") # or "DEFAULT"
ps, st = Lux.setup(Random.default_rng(), model)

x = rand(Float32, 224, 224, 3, 1)
model(x, ps, Lux.testmode(st))