`ndarray.max()` does not call CUB in some conditions
See original GitHub issueAs mentioned in #4232, a daily test fails where ndarray.max()
does not call cupy.core._routines_statistics.cub.device_segmented_reduce
for some reasons.
https://jenkins.preferred.jp/job/chainer/job/daily_master_cupy-shuffled/378/ID=20,label=mn1-p100/
The following shows the condition of the daily shuffle test.
11:22:17 + ./run_cupy_combination_test.py --id 20
11:22:17 --- Shuffle Parameters ---
11:22:17 cuda_libs: ('cuda90', 'cudnn73-cuda9', 'nccl2.4-cuda9', 'none')
11:22:17 scipy: None
11:22:17 base: ubuntu16_py35
11:22:17 numpy: 1.18
11:22:17 --- Configuration ---
11:22:17 nccl: nccl2.4-cuda9
11:22:17 cutensor: none
11:22:17 cudnn: cudnn73-cuda9
11:22:17 base: ubuntu16_py35
11:22:17 cuda: cuda90
11:22:17 requires: ['numpy<1.19']
11:23:55 sha256:28646ab8950d2c4793c87a7631006aa0a7c063fd6ba98cfeef674f1c0dff3798
Issue Analytics
- State:
- Created 3 years ago
- Reactions:2
- Comments:7 (7 by maintainers)
Top Results From Across the Web
how to use np.max for empty numpy array without ValueError
Yes, calling numpy.max() on an empty array throws an error. Just check if the array is empty, if it is don't call max() ......
Read more >How to use the NumPy max function - Sharp Sight
max() function computes the maximum value of the numeric values contained in a NumPy array. It can also compute the maximum value of...
Read more >NumPy's max() and maximum(): Find Extreme Values in Arrays
max() method has scanned the whole array and returned the largest element. Using this method is exactly equivalent to calling np.max(n_scores) .
Read more >TensorRT 8.4.1 Release Notes - NVIDIA Documentation Center
This release includes several fixes from the previous TensorRT releases as well as the following additional changes. These Release Notes are ...
Read more >VBA Call / Run Sub From another Subroutine - Automate Excel
However, it has no effect whatsoever on how the code is run or stored. Sub TestRoutine() Call RunRoutine1 Call RunRoutine2 End Sub. vba...
Read more >Top Related Medium Post
No results found
Top Related StackOverflow Question
No results found
Troubleshoot Live Code
Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start FreeTop Related Reddit Thread
No results found
Top Related Hackernoon Post
No results found
Top Related Tweet
No results found
Top Related Dev.to Post
No results found
Top Related Hashnode Post
No results found
Top GitHub Comments
Looks good to skip the mock tests for CUDA 9.0 (than to vary the range of
dtype
). Yes, your PR is welcome.Now
test_cub_max
andtest_cub_min
report the same error.