I have 4GB memory in the beaglebone black board received and 92% space is used. First I tried to connect the board using serial from iMac and able to talk to the system.
After that I connected the board from windows 10 and connected through ssh and did pandas installation using command “pip install pandas”. After some time it came back with following error.
import platform
issues with named module “_build_utils.apple_accelerate”. Subsequently I am unable to install pandas.
How to resolve this “build_utils.apple_accelerate” issue?
On Tue, 5 Feb 2019 23:14:55 -0800 (PST), rkanhiroten@gmail.com declaimed the
following:
After that I connected the board from windows 10 and connected through ssh
and did pandas installation using command "pip install pandas". After some
time it came back with following error.
import platform
issues with named module "_build_utils.apple_accelerate". Subsequently I am
unable to install pandas.
That does not appear to be any sort of Python related error traceback.
Don't paraphrase error messages -- cut&paste the exact text (not an image
please).
The platform module should not be calling anything with _build_utils in
the name.
How to resolve this "build_utils.apple_accelerate" issue?
2. How can add additional space with mmc card?
By not using the eMMC... Download the LXQT OS http://beagleboard.org/latest-images write it to an 8-32GB uSD card, insert
the uSD card, and reboot the BBB.
OR -- if you do not need x-window (graphical environment), download the
IoT image, write it to a 4-32GB uSD card, mount it and follow the
instructions to turn it into a flasher image. Then reboot to let it flash
the IoT image to the eMMC (after which you need to remove it as it will
attempt to reflash each time you boot with it -- you can, after booting off
eMMC, mount the card and convert it back to non-flasher mode). The IoT
image is about half the size of the LXQT image.
For installing pandas, try apt-get instead. If that doesn’t work, then as Dennis said, just copy/paste the whole error message.
$ apt-cache search pandas
python-pandas - data structures for “relational” or “labeled” data
python-pandas-doc - documentation and examples for pandas
python3-pandas - data structures for “relational” or “labeled” data - Python 3
There are several options. Choose python-pandas or python3-pandas and install from apt instead.
Thanks for your quick response. Listed the error message I get regarding the build_utils.apple_accelerate error with pandas installation.
File "/usr/lib/python2.7/dist-packages/numpy/distutils/misc_util.py", line 966, in add_subpackage
caller_level = 2)
File "/usr/lib/python2.7/dist-packages/numpy/distutils/misc_util.py", line 935, in get_subpackage
caller_level = caller_level + 1)
File "/usr/lib/python2.7/dist-packages/numpy/distutils/misc_util.py", line 872, in _get_configuration_from_setup_py config = setup_module.configuration(*args)
File "numpy/setup.py", line 10, in configuration
from os.path import join as pjoin
File "/usr/lib/python2.7/dist-packages/numpy/distutils/misc_util.py", line 966, in add_subpackage
caller_level = 2)
File "/usr/lib/python2.7/dist-packages/numpy/distutils/misc_util.py", line 935, in get_subpackage
caller_level = caller_level + 1)
File "/usr/lib/python2.7/dist-packages/numpy/distutils/misc_util.py", line 847, in _get_configuration_from_setup_py ('.py', 'U', 1))
File "numpy/core/setup.py", line 13, in <module>
import platform
ImportError: No module named _build_utils.apple_accelerate
On Thu, 7 Feb 2019 07:43:56 -0800, Ram Kanhirotentavida
<rkanhiroten@gmail.com> declaimed the
following:
File "/usr/lib/python2.7/dist-packages/numpy/distutils/misc_util.py",
line 847, in _get_configuration_from_setup_py ('.py', 'U', 1))
File "numpy/core/setup.py", line 13, in <module>
import platform
ImportError: No module named _build_utils.apple_accelerate
That still looks strange... Line 13 of numpy/core/setup does import
_build_utils, but the line is:
from numpy._build_utils.apple_accelerate import (
uses_accelerate_framework, get_sgemv_fix
)
"""
from __future__ import division, print_function
import os
import sys
import pickle
import copy
import warnings
import platform
from os.path import join
from numpy.distutils import log
from distutils.dep_util import newer
from distutils.sysconfig import get_config_var
from numpy._build_utils.apple_accelerate import (
uses_accelerate_framework, get_sgemv_fix
)
from numpy.compat import npy_load_module
"""
import platform is on line 8
I'd echo the other response: if you are installing for a system
provided Python -- use apt-get, as that should be a tested/good package,
whereas pip has to download a generic source tree and then run setup
functions to build anything needed locally (and apparently either the numpy
that pip is fetching doesn't have the _build_utils files, OR there is a
numpy already installed which doesn't have said file, but keeps pip from
downloading a different numpy.