FS#69494 - [python-pytorch] mismatch numpy
Attached to Project:
Community Packages
Opened by DarcyHu (DarcyHu) - Monday, 01 February 2021, 08:08 GMT
Last edited by Konstantin Gizdov (kgizdov) - Monday, 01 February 2021, 16:35 GMT
Opened by DarcyHu (DarcyHu) - Monday, 01 February 2021, 08:08 GMT
Last edited by Konstantin Gizdov (kgizdov) - Monday, 01 February 2021, 16:35 GMT
|
Details
Description:
could not import torch due to numpy version issue Additional info: * python-numpy 1.19.5-1 * python-pytorch-opt-cuda 1.7.1-4 Traceback (most recent call last): File "<string>", line 1, in <module> File "/usr/lib/python3.9/site-packages/torch/__init__.py", line 190, in <module> from torch._C import * RuntimeError: module compiled against API version 0xe but this version of numpy is 0xd Steps to reproduce: run this command: python -c "import torch" Maybe you build pytorch with the python-numpy from testing repo by accident? |
This task depends upon
Closed by Konstantin Gizdov (kgizdov)
Monday, 01 February 2021, 16:35 GMT
Reason for closing: Fixed
Additional comments about closing: correct python-numpy is now in [extra]
Monday, 01 February 2021, 16:35 GMT
Reason for closing: Fixed
Additional comments about closing: correct python-numpy is now in [extra]
2021-02-01 17:34:06.096069: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.11.0
RuntimeError: module compiled against API version 0xe but this version of numpy is 0xd
RuntimeError: module compiled against API version 0xe but this version of numpy is 0xd
ImportError: numpy.core._multiarray_umath failed to import
ImportError: numpy.core.umath failed to import
2021-02-01 17:34:06.394154: F tensorflow/python/lib/core/bfloat16.cc:714] Check failed: PyBfloat16_Type.tp_base != nullptr