I have a question about the cache loading mechanism in Numba. Here’s a piece of test code I wrote: import numba as nb @nb.njit(cache=True) def test_jit(): a = 0 for i in range(1000000): a += i return a If I execute this code from scratch, the execution time is 12.8 ms: %time…

Understanding Initial Execution Delay in Numba Cached Functions

gmarkall November 5, 2024, 2:44pm 2

Topic		Replies	Views
Numba Warm-up Speed With Cached Functions Support: How do I do ...?	3	1843	January 17, 2023
Some questions about Numba behavior Community Support	1	448	October 18, 2021
Re-execution of the same jitted function causes re-compilation for same type of data Community Support	9	517	October 23, 2020
Reduce Compilation Time Support: How do I do ...?	8	1847	November 3, 2021
Caching vs AOT: What are the use cases? Support: How do I do ...?	0	214	August 15, 2023