[SOLVED] Vectorbt import hangs on Windows unless Numba JIT is disabled

Vectorbt import hangs on Windows unless Numba JIT is disabled

I’m trying to use vectorbt in Python on Windows, but the import hangs indefinitely:

import vectorbt as vbt  # hangs here

Environment / Versions I’ve tested:

Python: 3.9, 3.10, 3.11, 3.12, 3.13 (all 64-bit)
NumPy: 1.2 , 2.3
Numba: 0.58 → 0.63
vectorbt: latest

Observations:

The import hangs regardless of NumPy, Numba, or Python version.
It only works if I disable Numba JIT:

import os
os.environ["NUMBA_DISABLE_JIT"] = "1"
import vectorbt  # works, but calculations are slow

Goal:
I want to import Vectorbt without hanging, while keeping Numba JIT fully enabled for fast computations.

Question:
Why does Vectorbt hang on import?
Is there a way to safely initialize Vectorbt on Windows with full JIT enabled?

Solution

After about 8 hours, I finally realized the problem was super simple 😅
Vectorbt hangs in Jupyter because Numba can’t access its cache folder and I could fix it by setting a writable cache path before import or Run Jupyter as Administrator:

import os
os.environ["NUMBA_CACHE_DIR"] = r"C:\temp\numba_cache"
import vectorbt as vbt