1. User Manual¶
- 1.1. A ~5 minute guide to Numba
- 1.2. Overview
- 1.3. Installation
- 1.4. Compiling Python code with
@jit
- 1.5. Flexible specializations with
@generated_jit
- 1.6. Creating Numpy universal functions
- 1.7. Compiling python classes with @jitclass
- 1.8. Creating C callbacks with
@cfunc
- 1.9. Compiling code ahead of time
- 1.10. Automatic parallelization with
@jit
- 1.11. Using the
@stencil
decorator - 1.12. Callback into the Python Interpreter from within JIT’ed code
- 1.13. Performance Tips
- 1.14. The Threading Layers
- 1.15. Troubleshooting and tips
- 1.16. Frequently Asked Questions
- 1.16.1. Programming
- 1.16.1.1. Can I pass a function as an argument to a jitted function?
- 1.16.1.2. Numba doesn’t seem to care when I modify a global variable
- 1.16.1.3. Can I debug a jitted function?
- 1.16.1.4. How can I create a Fortran-ordered array?
- 1.16.1.5. How can I increase integer width?
- 1.16.1.6. How can I tell if
parallel=True
worked?
- 1.16.2. Performance
- 1.16.2.1. Does Numba inline functions?
- 1.16.2.2. Does Numba vectorize array computations (SIMD)?
- 1.16.2.3. Why my loop is not vectorized?
- 1.16.2.4. Does Numba automatically parallelize code?
- 1.16.2.5. Can Numba speed up short-running functions?
- 1.16.2.6. There is a delay when JIT-compiling a complicated function, how can I improve it?
- 1.16.3. Integration with other utilities
- 1.16.4. Miscellaneous
- 1.16.1. Programming
- 1.17. Examples
- 1.18. Talks and Tutorials