pymc-devs / pymc-devs/pytensor

BUG: <```uint64 dtype``` is broken for ```Max```>

Open
#770 3 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug C-backend
Dominant language
Python
Stars
644
Forks
208
Avg merge
2d 14h
Merged PRs (30d)
16

Description

Describe the issue:

The Max op which is a subclass of CAReduce fails for 64 bit unsigned integers. This is also evident in the PR 731 and its test. The exact number where uint64 starts failing for is 9223372036854775.
Error can also be reproduced with the following example:

Reproducable code example:
import pytensor.tensor as pt
import numpy as np

dtype="uint64"
n = pt.vector("n", shape=(None,), dtype=dtype)
test_n = np.array([0, 9223372036854775], dtype)

assert pt.max(n).dtype == dtype
print(pt.max(n).eval({n: test_n})) # 9223372036854776
print(test_n.max()) # 9223372036854775
assert pt.max(n).eval({n: test_n}) == test_n.max()  # Fails, returns 1 larger
Error message:

No response

PyTensor version information:

Python 3.11.8
Pytensor 2.20.0

Context for the issue:

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the Max operation and its CAReduce base, then inspect the uint64 case and the test referenced from PR 731. Reproduce the supplied Python/NumPy example; done when pt.max(n) returns the same uint64 maximum as test_n.max() for the failing value.

Written by the indexing model from the issue text.

Assessment

Tech stack
numpy, python
Domain
data
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.