jjhelmus / berryconda Goto Github PK
View Code? Open in Web Editor NEWConda based Python distribution for the Raspberry Pi
License: BSD 3-Clause "New" or "Revised" License
Conda based Python distribution for the Raspberry Pi
License: BSD 3-Clause "New" or "Revised" License
I used berryconda to install opencv 3.3.0 into a python 3.5 virtual environment
but "import cv2" yielded this error:
ImportError: numpy.core.multiarray failed to import
Hi!
I am trying to create armhf Dockerfile with automated build of the JupyterLab using this hack: https://resin.io/blog/building-arm-containers-on-any-x86-machine-even-dockerhub/
Here's the link to my repo: https://github.com/movalex/rpi-jupyter-conda/blob/base/Dockerfile
Unfortunately when Berryconda3-2.0.0-Linux-armv7l.sh
is installed, I've got the error on this step:
conda config --system --add channels rpi
qemu: uncaught target signal 4 (Illegal instruction) - core dumped
Illegal instruction
Is it possible to fix this error?
Thank you very much for Berryconda.
There is a problem in installing conda install -c rpi conda-build=2.1.10
in RPi B, having raspbian jessie 8 with Berryconda2-1.0.0-Linux-armv7l.sh. The error is as follows
ERROR conda.core.link:_execute_actions(335): An error occurred while installing package 'defaults::conda-build-2.1.10-py27_0'.
IOError(2, 'No such file or directory')
Attempting to roll back.
IOError(2, 'No such file or directory')
More than the error, further try for the conda command ends in -bash: /home/pi/berryconda2/bin/conda: No such file or directory
, making to suspect the installation command completely removes the conda bin from the berryconda installation. To get back the lost bin/conda, a fresh installation of the berryconda is followed. Similar losing of bin/conda was observed for installing packages such as anaconda-client etc too.
What would be reasons to choose the one or the other?
Hi,
I just successfully installed berryconda on the pi, followed by numpy, matplotlib and jupyter notebook. However, I can't find ipython qtconsole, is it because it is not in the default repo ? or another reason ?
thanks
Importing Pandas on the raspberry pi, the following error crops up
`pi@raspberrypi:~/scripts $ python
Python 3.6.1 | packaged by rpi | (default, Apr 20 2017, 19:35:19)
[GCC 4.9.2] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import pandas
Traceback (most recent call last):
File "/home/pi/berryconda3/lib/python3.6/site-packages/pandas/__init__.py", line 26, in <module>
from pandas._libs import (hashtable as _hashtable,
File "/home/pi/berryconda3/lib/python3.6/site-packages/pandas/_libs/__init__.py", line 3, in <module>
from .tslib import iNaT, NaT, Timestamp, Timedelta, OutOfBoundsDatetime
ImportError: /home/pi/berryconda3/lib/python3.6/site-packages/pandas/_libs/tslib.cpython-36m-arm-linux-gnueabihf.so: file too short
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/pi/berryconda3/lib/python3.6/site-packages/pandas/__init__.py", line 35, in <module>
"the C extensions first.".format(module))
ImportError: C extension: /home/pi/berryconda3/lib/python3.6/site-
packages/pandas/_libs/tslib.cpython-36m-arm-linux-gnueabihf.so: file too short not built. If you want
to import pandas from the source directory, you may need to run 'python setup.py build_ext --inplace
--force' to build the C extensions first.
>>>`
I'm trying to get paramiko working with the Berryian 2.0.0 distribution. Using cryptography 2.3.1, OpenSSL 1.0.2p, and paramiko 2.4.2 to open an SSH connection givens the error:
cryptography.exceptions.UnsupportedAlgorithm: Backend object does not support secp256r1.
I checked with the cryptography
contributors, and they said that the OpenSSL backend should provide support for secp256r1. See pyca/cryptography#4691. I'm wondering if there's any known reason why this might be happening.
Versions of Python, cryptography
, cffi
, pip
, and setuptools
you're using
Python: 3.6.1
cryptography: 2.3.1
cffi: 1.11.5
pip: 18.1
setup tools: 36.0.1
openssl: 1.0.2p
Running on Raspberry Pi 3+ with Raspbian Stretch and the Berryconda 2.0.0 distribution.
Clear steps for reproducing your bug
Installed Berryconda; pip installed paramiko; ran a simple example:
import paramiko
t = paramiko.Transport(("127.0.0.1", 22))
t.connect(username="pi", password=<password>)
This gives error:
Unknown exception: Backend object does not support secp256r1.
Traceback (most recent call last):
File "/home/producer/berryconda3/lib/python3.6/site-packages/paramiko/transport.py", line 2010, in run
self._handler_table[ptype](self, m)
File "/home/producer/berryconda3/lib/python3.6/site-packages/paramiko/transport.py", line 2126, in _negotiate_keys
self.kex_engine.start_kex()
File "/home/producer/berryconda3/lib/python3.6/site-packages/paramiko/kex_ecdh_nist.py", line 32, in start_kex
self._generate_key_pair()
File "/home/producer/berryconda3/lib/python3.6/site-packages/paramiko/kex_ecdh_nist.py", line 53, in _generate_key_pair
self.P = ec.generate_private_key(self.curve, default_backend())
File "/home/producer/berryconda3/lib/python3.6/site-packages/cryptography/hazmat/primitives/asymmetric/ec.py", line 287, in generate_private_key
return backend.generate_elliptic_curve_private_key(curve)
File "/home/producer/berryconda3/lib/python3.6/site-packages/cryptography/hazmat/backends/openssl/backend.py", line 1346, in generate_elliptic_curve_private_key
_Reasons.UNSUPPORTED_ELLIPTIC_CURVE
cryptography.exceptions.UnsupportedAlgorithm: Backend object does not support secp256r1.
Hi, you mention in your manual that Pi2 will need the AMRv7 version; however, I just realized that this might not be the case for all Pi2. I got an "illegal instruction" error when trying to use conda and by googling discovered that AMRv7 was the culprit. With AMRv6 everything runs smoothly (thanks btw!). I'd suggest adding "cat /proc/cpuinfo" to the manual to determine which version is needed.
Do you tink it is possible to have this package with these bindings available?
I tried to compile opencv on my own to no avail.
Thanks for all your hard work!
At the moment instally pytorch (cpu support, no cuda) can only be done installing from source which takes at least 8 hours! It would be great to have a binary!
Hi there and thanks again for the great work.
I would like to kindly request an update for Pandas from 0.22.0 to the latest version (currently 0.23.1), if possible.
Edit: Bokeh 0.13.0 has just been released, too. This would be much appreciated as well.
Cheers
seaborn 0.8.1 -missing. Can this be added?
//Rajesh
Please report any problems with Berryconda or the packages in the rpi channel by submitting an issue. Also, use this link to request new packages.
I would like to request Bokeh 0.12.15, since only 0.12.13 from December 2017 seems to be available on the rpi channel. I hope this is the right way to go about this. Or should I try to build Bokeh myself?
Many thanks!
I'm using Berryconda3-1.0.0-Linux-armv7l.sh in my Raspberry Pi 3 Model B.
Berryconda3 was installed successfully.
When I run :
jupyter
I get :
usage: jupyter [-h] [--version] [--config-dir] [--data-dir] [--runtime-dir] [--paths] [--json] [subcommand] jupyter: error: one of the arguments --version subcommand --config-dir --data-dir --runtime-dir --paths is required
And when I try to run :
jupyter notebook
I get this error :
Traceback (most recent call last):
File "/home/pi/berryconda3/bin/jupyter-notebook", line 7, in <module>
from notebook.notebookapp import main
File "/home/pi/berryconda3/lib/python3.6/site-packages/notebook/notebookapp.py", line 46, in <module>
from zmq.eventloop import ioloop
File "/home/pi/berryconda3/lib/python3.6/site-packages/zmq/__init__.py", line 47, in <module>
from zmq import backend
File "/home/pi/berryconda3/lib/python3.6/site-packages/zmq/backend/__init__.py", line 40, in <module>
reraise(*exc_info)
File "/home/pi/berryconda3/lib/python3.6/site-packages/zmq/utils/sixcerpt.py", line 34, in reraise
raise value
File "/home/pi/berryconda3/lib/python3.6/site-packages/zmq/backend/__init__.py", line 27, in <module>
_ns = select_backend(first)
File "/home/pi/berryconda3/lib/python3.6/site-packages/zmq/backend/select.py", line 27, in select_backend
mod = __import__(name, fromlist=public_api)
File "/home/pi/berryconda3/lib/python3.6/site-packages/zmq/backend/cython/__init__.py", line 6, in <module>
from . import (constants, error, message, context,
ImportError: /usr/lib/arm-linux-gnueabihf/libstdc++.so.6: version `CXXABI_1.3.9' not found (required by /home/pi/berryconda3/lib/python3.6/site-packages/zmq/backend/cython/../../../../../libzmq.so.5)
I installed jupyter notebook using conda, and I checked that it is installed and it is in (conda list)
How can I run jupyter notebook ?
Hi,
I downloaded the file for Python 2 on a RaspberryPi 3B+ on my Mac and then used FTP to put it on the raspberry pi. I then made the .sh file executable using instructions provided but when I then try to execute it I get the following message:
ERROR: size of Berryconda2-2.0.0-Linux-armv7l.sh should be 33565270 bytes
Any idea what I a doing wrong?
Now that bokeh-charts has been deprecated in favor of holoviews (which is pretty awesome), it would be nice to have holoviews packages. I think to be useful also requires packages for param, paramnb, and parambokeh.
Hi! Just did a fresh install of this on a Raspberry Pi Zero W and am seeing strange behavior.
I'm on Raspbarian Stretch Lite, 2018-06-27, and I just installed Berryconda. I've attached the terminal session where I discovered the issue.
Any invocation of conda
dies with Illegal instruction
after about a second.
Same with conda info
conda install
conda develop
etc.
python works, and is 3.6, though I can't tell if it's the conda version
Welcome to Berryconda3 2.0.0
In order to continue the installation process, please review the license
agreement.
Please, press ENTER to continue
>>>
Copyright (c) 2016-2017 Jonathan J. Helmus
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are
met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following
disclaimer in the documentation and/or other materials provided
with the distribution.
* Neither the name of the developers nor the names of any
contributors may be used to endorse or promote products derived
from this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
# (ISO/IEC-6429). (Lack of such support is extremely rare, and such
Do you approve the license terms? [yes|no]
[no] >>> y
Please answer 'yes' or 'no':
>>> yes
Berryconda3 will now be installed into this location:
/home/pi/berryconda3
- Press ENTER to confirm the location
- Press CTRL-C to abort the installation
- Or specify a different location below
[/home/pi/berryconda3] >>>
PREFIX=/home/pi/berryconda3
installing: python-3.6.1-0 ...
^[[B^[[B^[[B^[[B^[[B^[[B^[[BPython 3.6.1
installing: asn1crypto-0.22.0-py36_0 ...
installing: ca-certificates-2017.4.17-0 ...
installing: certifi-2017.4.17-py36_0 ...
installing: cffi-1.10.0-py36_0 ...
installing: chardet-3.0.4-py36_0 ...
installing: conda-4.3.22-py36_0 ...
installing: conda-env-2.6.0-0 ...
installing: cryptography-1.9-py36_0 ...
installing: idna-2.5-py36_0 ...
installing: libffi-3.2.1-0 ...
installing: ncurses-5.9-0 ...
installing: openssl-1.0.2l-0 ...
installing: packaging-16.8-py36_0 ...
installing: pip-9.0.1-py36_0 ...
installing: pycosat-0.6.1-py36_0 ...
installing: pycparser-2.17-py36_0 ...
installing: pyopenssl-16.2.0-py36_0 ...
installing: pyparsing-2.2.0-py36_0 ...
installing: readline-6.3-0 ...
installing: requests-2.18.1-py36_0 ...
installing: ruamel_yaml-0.11.14-py36_0 ...
installing: setuptools-36.0.1-py36_0 ...
installing: six-1.10.0-py36_0 ...
installing: sqlite-3.15.0-0 ...
installing: tk-8.6.6-0 ...
installing: urllib3-1.21.1-py36_0 ...
installing: wheel-0.29.0-py36_0 ...
installing: xz-5.2.2-0 ...
installing: yaml-0.1.6-0 ...
installing: zlib-1.2.11-0 ...
installation finished.
Do you wish the installer to prepend the Berryconda3 install location
to PATH in your /home/pi/.bashrc ? [yes|no]
[no] >>>
You may wish to edit your .bashrc or prepend the Berryconda3 install location:
$ export PATH=/home/pi/berryconda3/bin:$PATH
Thank you for installing Berryconda3!
pi@demozero:~ $ less /.bash_rc
/.bash_rc: No such file or directory
pi@demozero:~ $ conda
-bash: conda: command not found
pi@demozero:~ $ vim .bashrc
pi@demozero:~ $ bash
pi@demozero:~ $ conda
Illegal instruction
pi@demozero:~ $ python --version
Python 3.6.1
pi@demozero:~ $ python
Python 3.6.1 | packaged by rpi | (default, Apr 20 2017, 19:35:19)
[GCC 4.9.2] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>>
Hi jjhelmus,
Would it be possible to add Scikit-optimize (SKOPT) (https://github.com/scikit-optimize/scikit-optimize) to the rpi channel?
SKOPT is implementing a Bayesian Optimization routine for optimizing hyperparameters in e.g. machine learnings. Furthermore, it has proven succesfull in optimization of scientific experiments.
Many thanks for the berryconda work!
Best
/sqbl
Do you know when version v3.4.2 for OpenCV is going to be made available? I'm saying this because I'm interested in using a new feature that came with this version called EAST text detector.
Here's an article of what I'm talking about:
https://www.pyimagesearch.com/2018/08/20/opencv-text-detection-east-text-detector/
Thank you!
Any chance you could implement wxPython?
I'm interested in creating some recipes for other Python packages to help add to the rpi channel, but I figured I'd throw in this request while I am learning to do so. Thanks!
Having ipyleaflet would be awesome giving its recent improvements.
More generally speaking, maybe it would be helpful to provide a section about how to contribute such additional packages as external contributions to this project? Thanks a lot, BTW!
@:~ $ openssl
WARNING: can't open config file: /home/jhelmus/bc3_v7/conda-bld/openssl_1547610726478/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_place/ssl/openssl.cnf
I have installed tensorflow using conda and it did not show any error, but when I tried to check if the installation works properly it shows the following error:
no module named tensorflow
Hi,
I'm trying to install jupyterHub on raspberry, is there any chance to get it from this package?
Thank you very much!
Hi Jonathan,
I'm keen on biomedical text/data mining and I'd like to suggest Biopython, Gensim and NLTK as candidates for the rpi-channel.
Best regards,
Daniel
The installation instructions are missing a step, I said "yes" to add anaconda reference links in the bashrc file however this is not as easy and simple as ubuntu where I have anaconda working no problem. In debian the bashrc file is in a different location and the typical source command doesn't work in debian so I need some assistance on which command or where the bashrc file is located on the raspberry pi
I'm used to the behaviour of pip in the system-provided python and virtualenv/virtualenvwrapper, where pip install -e
a package from git would clone the repo into a subdirectory of the virtualenv. For instance, using virtualenvwrapper, I can do this:
~ $ mkvirtualenv foobarenv
~ $ pip install -e git+https://github.com/foobar/foobar.git
~ $ ls -ld /home/ksze/.virtualenvs/foobarenv/src/
total 8
drwxr-xr-x 8 ksze ksze 4096 Jul 31 10:38 foobar
-rw-r--r-- 1 ksze ksze 185 Jul 31 10:39 pip-delete-this-directory.txt
And foobar's git repo would be cloned into /home/ksze/.virtualenvs/foobar/src/foobar
, regardless of where I run pip from.
However, using berryconda, pip install -e
just clones to a subdirectory under the current working directory. This behaviour is quite annoying, not to mention being incompatible with what people expect when using a requirements file.
~ $ conda create -n foobar pip
~ $ source activate foobar
~ $ pip install -e git+https://github.com/foobar/foobar.git
~ $ ls -ld ./src/
total 8
drwxr-xr-x 8 ksze ksze 4096 Jul 31 10:41 foobar
-rw-r--r-- 1 ksze ksze 185 Jul 31 10:40 pip-delete-this-directory.txt
I'm not sure if this is an upstream conda issue or specific to berryconda, as I have never used the "vanilla" conda.
I ran Microsoft-ELL tutorial on Pi 0 W, and it seems the opencv package is not compiled properly.
My version of opencv is
opencv: 3.3.0-py36hcfffc66_0
It is installed by this command
conda install -c microsoft-ell/label/pi_zero opencv
And the packages installed are
Package plan for installation in environment /home/pi/berryconda3/envs/py36:
The following NEW packages will be INSTALLED:
bzip2: 1.0.6-1 lisaong
ffmpeg: 3.2.4-hf767ee3_0
giflib: 5.1.4-h3b1de47_0
hdf5: 1.8.17-0
jasper: 1.900.1-4 microsoft-ell
jpeg: 9b-0 microsoft-ell
libgfortran: 3.0.0-1
libpng: 1.6.29-0 microsoft-ell
libtiff: 4.0.6-0
libwebp: 0.5.2-hcb3eeb6_7
numpy: 1.14.0-py36hebd9de5_0
openblas: 0.2.19-0
opencv: 3.3.0-py36hcfffc66_0
x264: 20131217-hc5d671e_0
Here are the errors.
OpenCV Error: Unspecified error (The function is not implemented. Rebuild the library with Windows, GTK+ 2.x or Carbon support. If you are on Ubuntu or Debian, install libgtk2.0-dev and pkg-config, then re-run cmake or configure script) in cvWaitKey, file /home/jhelmus/conda/conda-bld/opencv_1512232341522/work/modules/highgui/src/window.cpp, line 676
Traceback (most recent call last):
File "tutorial.py", line 89, in
main()
File "tutorial.py", line 49, in main
while (cv2.waitKey(1) & 0xFF) == 0xFF:
cv2.error: /home/jhelmus/conda/conda-bld/opencv_1512232341522/work/modules/highgui/src/window.cpp:676: error: (-2) The function is not implemented. Rebuild the library with Windows, GTK+ 2.x or Carbon support. If you are on Ubuntu or Debian, install libgtk2.0-dev and pkg-config, then re-run cmake or configure script in function cvWaitKey
There appears to be an issue running Jupyter notebook in Berryconda.
I'm using a Raspberry Pi 3 Model B, with Raspbian Jessie 8. Using Berryconda3-1.0.0-Linux-armv7l.sh.
Typing jupyter run or jupyter-run yields the following error:
pi@raspberrypi:~ $ jupyter-run
[RunApp] WARNING | Native kernel (python3) is not available
Traceback (most recent call last):
File "/home/pi/berryconda3/lib/python3.5/site-packages/jupyter_client/kernelspec.py", line 201, in get_kernel_spec
resource_dir = d[kernel_name.lower()]
KeyError: 'python3'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/home/pi/berryconda3/bin/jupyter-run", line 6, in <module>
sys.exit(jupyter_client.runapp.main())
File "/home/pi/berryconda3/lib/python3.5/site-packages/jupyter_core/application.py", line 267, in launch_instance
return super(JupyterApp, cls).launch_instance(argv=argv, **kwargs)
File "/home/pi/berryconda3/lib/python3.5/site-packages/traitlets/config/application.py", line 657, in launch_instance
app.initialize(argv)
File "<decorator-gen-7>", line 2, in initialize
File "/home/pi/berryconda3/lib/python3.5/site-packages/traitlets/config/application.py", line 87, in catch_config_error
return method(app, *args, **kwargs)
File "/home/pi/berryconda3/lib/python3.5/site-packages/jupyter_client/runapp.py", line 76, in initialize
JupyterConsoleApp.initialize(self)
File "/home/pi/berryconda3/lib/python3.5/site-packages/jupyter_client/consoleapp.py", line 334, in initialize
self.init_kernel_manager()
File "/home/pi/berryconda3/lib/python3.5/site-packages/jupyter_client/consoleapp.py", line 288, in init_kernel_manager
self.kernel_manager.start_kernel(**kwargs)
File "/home/pi/berryconda3/lib/python3.5/site-packages/jupyter_client/manager.py", line 244, in start_kernel
kernel_cmd = self.format_kernel_cmd(extra_arguments=extra_arguments)
File "/home/pi/berryconda3/lib/python3.5/site-packages/jupyter_client/manager.py", line 175, in format_kernel_cmd
cmd = self.kernel_spec.argv + extra_arguments
File "/home/pi/berryconda3/lib/python3.5/site-packages/jupyter_client/manager.py", line 87, in kernel_spec
self._kernel_spec = self.kernel_spec_manager.get_kernel_spec(self.kernel_name)
File "/home/pi/berryconda3/lib/python3.5/site-packages/jupyter_client/kernelspec.py", line 203, in get_kernel_spec
raise NoSuchKernel(kernel_name)
jupyter_client.kernelspec.NoSuchKernel: No such kernel named python3
As of right now, Jupyter Lab on the rpi channel is stuck in a pre-alpha version of a computational environment for Jupyter. Any chance an updated build could be done?
So I got a new Raspberry Pi Zero W and tried getting Berryconda to run on it. I encountered an error when trying to import Numpy although the install seemed to have run without issues.
Please find the error and outputs of conda info and conda list below:
pi@raspberrypi:~/berryconda3/pkgs $ which python
/home/pi/berryconda3/bin/python
pi@raspberrypi:~/berryconda3/pkgs $ python
Python 3.6.1 | packaged by rpi | (default, Apr 21 2017, 01:13:08)
[GCC 4.9.2] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import numpy
Traceback (most recent call last):
File "/home/pi/berryconda3/lib/python3.6/site-packages/numpy/core/__init__.py", line 16, in
<module>
from . import multiarray
ImportError: libgfortran.so.3: cannot open shared object file: No such file or directory
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "/home/pi/berryconda3/lib/python3.6/site-packages/numpy/__init__.py", line 142, in <module>
from . import add_newdocs
File "/home/pi/berryconda3/lib/python3.6/site-packages/numpy/add_newdocs.py", line 13, in
<module>
from numpy.lib import add_newdoc
File "/home/pi/berryconda3/lib/python3.6/site-packages/numpy/lib/__init__.py", line 8, in <module>
from .type_check import *
File "/home/pi/berryconda3/lib/python3.6/site-packages/numpy/lib/type_check.py", line 11, in
<module>
import numpy.core.numeric as _nx
File "/home/pi/berryconda3/lib/python3.6/site-packages/numpy/core/__init__.py", line 26, in
<module>
raise ImportError(msg)
ImportError:
Importing the multiarray numpy extension module failed. Most
likely you are trying to import a failed build of numpy.
If you're working with a numpy git repo, try `git clean -xdf` (removes all
files not under version control). Otherwise reinstall numpy.
Original error was: libgfortran.so.3: cannot open shared object file: No such file or directory
Output of conda info:
pi@raspberrypi:~/berryconda3/pkgs $ conda info
Current conda install:
platform : linux-armv6l
conda version : 4.3.22
conda is private : False
conda-env version : 4.3.22
conda-build version : not installed
python version : 3.6.1.final.0
requests version : 2.18.1
root environment : /home/pi/berryconda3 (writable)
default environment : /home/pi/berryconda3
envs directories : /home/pi/berryconda3/envs
/home/pi/.conda/envs
package cache : /home/pi/berryconda3/pkgs
/home/pi/.conda/pkgs
channel URLs : https://conda.anaconda.org/rpi/linux-armv6l
https://conda.anaconda.org/rpi/noarch
config file : /home/pi/berryconda3/.condarc
netrc file : None
offline mode : False
user-agent : conda/4.3.22 requests/2.18.1 CPython/3.6.1 Linux/4.9.35+ debian/8.0 glibc/2.19
UID:GID : 1000:1000
Output of conda list:
pi@raspberrypi:~/berryconda3/pkgs $ conda list
# packages in environment at /home/pi/berryconda3:
#
asn1crypto 0.22.0 py36_0
beautifulsoup4 4.6.0 py36_1
ca-certificates 2017.4.17 0
certifi 2017.4.17 py36_0
cffi 1.10.0 py36_0
chardet 3.0.4 py36_0
conda 4.3.22 py36_0
conda-env 2.6.0 0
cryptography 1.9 py36_0
cython 0.26 py36_0
idna 2.5 py36_0
libffi 3.2.1 0
libgfortran 3.0.0 0
ncurses 5.9 0
numpy 1.13.1 py36_0
openblas 0.2.19 0
openssl 1.0.2l 0
packaging 16.8 py36_0
pip 9.0.1 py36_0
pycosat 0.6.1 py36_0
pycparser 2.17 py36_0
pyopenssl 16.2.0 py36_0
pyparsing 2.2.0 py36_0
python 3.6.1 0
python-dateutil 2.6.1 py_0
pytz 2017.2 py_0
readline 6.3 0
requests 2.18.1 py36_0
ruamel_yaml 0.11.14 py36_0
setuptools 36.0.1 py36_0
six 1.10.0 py36_0
sqlite 3.15.0 0
tk 8.6.6 0
urllib3 1.21.1 py36_0
wheel 0.29.0 py36_0
xz 5.2.2 0
yaml 0.1.6 0
zlib 1.2.11 0
I am using Conda on an ARM board quite similar to the Raspberry Pi, the BeagleBone Black, and it would be quite convenient to be able to use the Berryconda channel for this – which almost works. The problem is the following:
The board has an AM335x 1GHz Cortex-A8 CPU and the stock Debian (jessie) distribution is armv7l with hardware floating point, which should be compatible with Raspbian. However, while the CPU supports NEON, it doesn't support VFPv4, which is enabled on the Berryconda armv7l binaries. (Symptom: Python crashes with SIGILL due to an v4-only FMA instruction.)
Would it be possible to build the armv7l binaries without vfpv4? The performance difference should be unnoticable for typical berryconda use cases.
Alternatively, is there a way to force conda to use the armv6l binaries? These seem to work.
First. thanks very much for creating this package!
Next: How do I make pip play nice with berryconda? I have a project on a Pi3 that needs setproctitle. It's not available thru conda. On OS X using Anaconda I'm used to being able to install packages with pip and have all the paths come out correctly in the virtual environment.
The terminal session below shows some system info and my path in a Python 3.5 root environment. Then I switch to a Python 2.7 environment and show the path. Trying pip install setproctitle
says it's already available under /usr/local/lib.
Then I start Python, check sys.path and then try to import setproctitle but get an ImportError -- which makes sense because sys.path only contains the pwd and the berryconda paths.
What's the recommended fix?
Thanks,
Mike
py27) pi@pi2:~ $ uname -a
Linux pi2 4.4.34-v7+ #930 SMP Wed Nov 23 15:20:41 GMT 2016 armv7l GNU/Linux
pi@pi2:~ $ echo $PATH
/home/pi/berryconda3/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/games:/usr/games
pi@pi2:~ $ source activate py27
(py27) pi@pi2:~ $ echo $PATH
/home/pi/berryconda3/envs/py27/bin:/home/pi/berryconda3/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/local/games:/usr/games
(py27) pi@pi2:~ $ which pip
/usr/local/bin/pip
(py27) pi@pi2:~ $ pip install setproctitle
Requirement already satisfied: setproctitle in /usr/local/lib/python2.7/dist-packages
(py27) pi@pi2:~ $ python
Python 2.7.13 | packaged by rpi | (default, Jan 15 2017, 19:52:30)
[GCC 4.9.2] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import sys
>>> sys.path
['', '/home/pi/berryconda3/envs/py27/lib/python27.zip', '/home/pi/berryconda3/envs/py27/lib/python2.7', '/home/pi/berryconda3/envs/py27/lib/python2.7/plat-linux2', '/home/pi/berryconda3/envs/py27/lib/python2.7/lib-tk', '/home/pi/berryconda3/envs/py27/lib/python2.7/lib-old', '/home/pi/berryconda3/envs/py27/lib/python2.7/lib-dynload', '/home/pi/berryconda3/envs/py27/lib/python2.7/site-packages']
>>> import setproctitle
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ImportError: No module named setproctitle
>>>
I'm wondering if it is possible to get Qt5, PyQt and PyOpenGL recipes ready.
There are some suggestions out there to compile Qt5 natively on raspberry PI 2&3. I would give it a try, but would need some guidance in which dependent packages need to come with berryconda and which are considered system-packages (coming with raspbian). The iteration cycle will be extremely slow, considering the approx. 2 days compilation time for Qt5 on an overclocked PI3 (https://wiki.qt.io/Native_Build_of_Qt5_on_a_Raspberry_Pi)
@jjhelmus In the end I would only need PyQt and PyOpenGL. So sharing your ideas would be very much welcome.
Update: Unfortunately Qt4 is needed also.
Hi Jonathan,
After reading https://github.com/samjabrahams/tensorflow-on-raspberry-pi and seeing that you published Tensorflow in one of your other channels I am wondering how feasible it is to have Tensorflow in Berryconda. It seems a bit less straightforward than other packages.
Thanks.
I have a package build request for pandas-datareader so it will be available on the rpi channel (per the Anaconda docs here):
https://anaconda.org/anaconda/pandas-datareader
Also regarding compatibility you may wish to know that I successfully installed Berry Conda on OSMC, after installing bzip2, as described in this thread:
https://discourse.osmc.tv/t/can-i-install-berry-conda-python-package-manager-on-osmc/36630
Thank you for any attention you may be able to devote to this package request.
Hi, it would be interesting to add support to RPI3b+ (with non-raspbian OS as ubuntu or suse) using the arm64 architecture.
Regards
I've installed berryconda, and it was very successful in downloading some requisite software for spectrometer use. However, I'm looking for a script editor that I can easily modify and run code in that needs to be run through berryconda (for the spectrometer). I've tried installing spyder and python-editor, but had no luck (Packages missing in current channels). Any suggestions?
PS: I'm fairly new to all this, so please bear me! Maybe the solution is super easy.
How we can install opencv3 ?
conda install -c menpo opencv3
gives
Solving environment: failed
PackagesNotFoundError: The following packages are not available from current channels:
- opencv3
Current channels:
- https://conda.anaconda.org/menpo/linux-armv7l
- https://conda.anaconda.org/menpo/noarch
- https://conda.anaconda.org/rpi/linux-armv7l
- https://conda.anaconda.org/rpi/noarch
To search for alternate channels that may provide the conda package you're
looking for, navigate tohttps://anaconda.org
and use the search bar at the top of the page.
Using dask.distributed in the notebook requires the ipywidgets package to show progress bars.
Any chance of seeing Pystan on Berryconda?
would it be possible to get the pyserial package?
Python 3.7 was released last week. It would be awesome if berryconda3 could be updated to install it.
Hello Jonathan,
I tried to import Excel data in Pandas (with : pd.read_excel) and got :
ImportError: No module named 'xlrd'
xlrd is not in your repo and also tried from conda but no success :
PackageNotFoundError: Package not found: '' Package missing in current linux-armv7l channels:
Thanks !!
(Thanks for you very useful repo on a Pi !! great time saver)
PREFIX=/home/simon/berryconda3
installing: python-3.6.1-0 ...
./Berryconda3-2.0.0-Linux-armv7l.sh: line 212: /home/simon/berryconda3/bin/python: No such file or directory
ERROR:
cannot execute native linux-armv7l binary, output from 'uname -a' is:
Linux localhost 4.9.65-perf+ #2 SMP PREEMPT Wed Jan 9 22:16:26 CST 2019 aarch64 aarch64 aarch64 GNU/Linux
In attempting to view images in openCV I run into this issue:
OpenCV Error: Unspecified error (The function is not implemented. Rebuild the library with Windows, GTK+ 2.x or Carbon support. If you are on Ubuntu or Debian, install libgtk2.0-dev and pkg-config, then re-run cmake or configure script) in cvShowImage, file /home/jhelmus/conda/conda-bld/opencv_1512148862464/work/modules/highgui/src/window.cpp, line 605
Traceback (most recent call last):
File "stereoCam.py", line 18, in <module>
cv2.imshow('Camera', image)
cv2.error: /home/jhelmus/conda/conda-bld/opencv_1512148862464/work/modules/highgui/src/window.cpp:605: error: (-2) The function is not implemented. Rebuild the library with Windows, GTK+ 2.x or Carbon support. If you are on Ubuntu or Debian, install libgtk2.0-dev and pkg-config, then re-run cmake or configure script in function cvShowImage
So it looks like openCV wasn't compiled with UI support. Is this something that you could provide?
For completeness this is the code snippet that generates the error:
from picamera.array import PiRGBArray
from picamera import PiCamera
import time
import cv2
camera = PiCamera()
camera.resolution = (640, 480)
camera.framerate = 30
rawCapture = PiRGBArray(camera, size=camera.resolution)
# pause while the camera inits
time.sleep(0.1)
for frame in camera.capture_continuous(rawCapture, format='bgr', use_video_port=True):
image = frame.array
# display using opencv
cv2.imshow('Camera', image)
key = cv2.waitKey(1) & 0xFF
# clear the stream for the next frame
rawCapture.truncate(0)
if key == ord('q'):
break
In this page ,the link to Berryconda is wrong to the https://https//github.com/jjhelmus/berryconda
Hi,
I want to use Berryconda to do my project and I need an OpenCV package. Would you mind to create an OpenCV package? Thank you.
Hey @jjhelmus : how hard would it be it package Py-ART?
Just got my Pi today and think this could be a cool "Py-ART in a can" tool
Hello:
Thank you
I tried to install numba using berryconda but i obtained this error:
Loading channels: done
PackagesNotFoundError: The following packages are not available from current channels:
Current channels:
To search for alternate channels that may provide the conda package you're
looking for, navigate to
https://anaconda.org
and use the search bar at the top of the page.
Any suggestion?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.