Our great sponsors
-
WorkOS
The modern identity platform for B2B SaaS. The APIs are flexible and easy-to-use, supporting authentication, user identity, and complex enterprise features like SSO and SCIM provisioning.
-
InfluxDB
Power Real-Time Data Analytics at Scale. Get real-time insights from all types of time series data with InfluxDB. Ingest, query, and analyze billions of data points in real-time with unbounded cardinality.
Neuronika was developed by a colleague of mine and me for a university project. ( we are master's students in CS). The framework offers auto-differentiation and dynamic neural networks, pretty much like Pytorch does.
Just make a PR: https://github.com/anowell/are-we-learning-yet
How does it compare to juice?
Maybe not for learning but as inspiration I have to plug this amazing effort for ML with (vulkan) shaders: https://github.com/charles-r-earp/autograph
The Vulkan docs online are quite good; if i remember they have "hello world" type tutorials as well. An take a look at something like Vulkan fft that has both Vulkan and opencl backends. Looking at "real" projects will of course tens to obscure the main points, but can be helpful for comparison.
there is the rust-gpu effort, an alternative to OpenCL https://github.com/EmbarkStudios/rust-gpu
It already works pretty well. You can see an example here. I have used it myself quite a bit to do ray tracing and simple matrix calculations. In the linked example the shader uses wgsl, however you can also just use glsl compiled to spirv.