Overview DeepCausality v.0.6 now supports multiple contexts. The previous context API remains the same, meaning all existing code should compile as before. However, new API functionality was added to interact with additional contexts to enable more advanced use cases.
Problem To deal with problems in the financial industry when modelling synthetics such as future spreads, multiple contexts apply to the instrument. Specifically, when modelling a classic long-short-term spread, in total three contexts apply, one for the long-term future contract, a second one for the short-term contract, and a final one for the resulting spread. In previous versions of DeepCausality, all three context could only be stored in one context, which becomes cumbersome to maintain over time.
Why Rust for machine learning? The Good The Not-So-Great Python! The Ugly Conclusion About This post briefly summarizes my current thoughts on machine learning in Rust. I wonder how these may change five years from now.
Why Rust for machine learning? Those companies that moved their ML production to Rust see more benefits than just a 25X speedup. Others do the obvious: interfacing Rust with C++ code since most ML libraries are still solid C++ implementations.
ArrayGrid - A Faster Tensor For Low Dim. Data Problem Solution Index Storage API Storage Implementation Grid Type ArrayGrid Usage About ArrayGrid - A Faster Tensor For Low Dim. Data DeepCausality allows fast and efficient adjustment of all values stored in a context hyper-graph. Often, this requires the formulation of an adjustment matrix. The matrix can already be attached to each element of the context graph but may require periodic updates depending on the required changes.
Welcome to our first blog post.
About DeepCausality is a hyper-geometric computational causality library that enables fast and deterministic context-aware causal reasoning in Rust. Please give us a star on GitHub.
