Tullio is a very flexible einsum macro. It understands many array operations written in index notation -- not just matrix multiplication and permutations, but also convolutions, stencils, scatter/gather, and broadcasting. Used by itself the macro writes ordinary nested loops much like Einsum.@einsum. One difference is that it can parse more expressions, and infer ranges for their indices. Another is that it will use multi-threading (via Threads.@spawn) and recursive tiling, on large enough arrays.
Features
- It uses LoopVectorization.@avx to speed many things up
- It uses KernelAbstractions.@kernel to make a GPU version
- The macro also tries to provide a gradient for use with Tracker
- The entire right hand side is summed over the full possible range of any indices not appearing on the left
- Co-operates with various other packages
- One difference is that it can parse more expressions
Categories
Data VisualizationLicense
MIT LicenseFollow Tullio.jl
You Might Also Like
Boost application security and continuity with SKUDONET ADC, our Open Source Load Balancer, that maximizes IT infrastructure flexibility. Additionally, save up to $470 K per incident with AI and SKUDONET solutions, further enhancing your organization’s risk management and cost-efficiency strategies.
Rate This Project
Login To Rate This Project
User Reviews
Be the first to post a review of Tullio.jl!