I tried to install pip install google-colab today, which failed for some reason. Since then, nothing works anymore. When I run code on the Python 3 kernel, it just never gets executed (restarting kernel doesnt help). When I run it on my venv kernel, I can execute code, which doesn't require any modules, such as print("hi"). But when I import modules, I get the following error:
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
/opt/anaconda3/envs/torch/lib/python3.8/codeop.py in __call__(self, source, filename, symbol)
134
135 def __call__(self, source, filename, symbol):
--> 136 codeob = compile(source, filename, symbol, self.flags, 1)
137 for feature in _features:
138 if codeob.co_flags & feature.compiler_flag:
TypeError: required field "type_ignores" missing from Module
Any help is appreciated!
I had the same problem in a virtual environment after installing google.colab -> removing the virtual environment and then created the virtual environment again. The following solved that problem:
pip3 uninstall notebook
pip3 install --ignore-installed --no-cache-dir --upgrade notebook
Related
Here's what I'm doing:
I have a blog that uses blogdown to render .Rmd files.
Some of the code snippets in the blog are in Python. I'm using reticulate for that.
I'm using a GitHub workflow to build and publish the blog as part of a larger website. This workflow sets up the environment and package dependencies in miniconda.
The last time this ran was six months ago. At that time, it worked. Now, it does not. I can't seem to replicate the behavior locally for more detailed debugging.
It seems to be trying to put a mamba command into normalizePath instead of a filesystem path (www-main is the name of the repository):
conda activate www-main
Rscript -e 'blogdown::build_site(local=FALSE, run_hugo=FALSE, build_rmd="content/blog/2020-08-28-api.Rmd")'
shell: /usr/bin/bash -l {0}
env:
CONDA_PKGS_DIR: /home/runner/conda_pkgs_dir
Rendering content/blog/2020-08-28-api.Rmd...
[...]
Quitting from lines 401-410 (2020-08-28-api.Rmd)
Error in normalizePath(conda, winslash = "/", mustWork = TRUE) :
path[1]="# cmd: /usr/share/miniconda/condabin/mamba update --name www-main --file /home/runner/work/www-main/www-main/conda": No such file or directory
Calls: local ... python_munge_path -> get_python_conda_info -> normalizePath
Execution halted
Error: Failed to render content/blog/2020-08-28-api.Rmd
Execution halted
Lines 401-410 of 2020-08-28-api.Rmd are a Python code block:
400 ```{python python-data, dev='svg'}
401 import covidcast
402 from datetime import date
403 import matplotlib.pyplot as plt
404
405 data = covidcast.signal("fb-survey", "smoothed_hh_cmnty_cli",
406 date(2020, 9, 8), date(2020, 9, 8),
407 geo_type="state")
408 covidcast.plot_choropleth(data, figsize=(7, 5))
409 plt.title("% who know someone who is sick, Sept 8, 2020")
410 ```
The useful bits of the output of conda info, in case it helps:
active environment : www-main
active env location : /usr/share/miniconda/envs/www-main
shell level : 1
user config file : /home/runner/.condarc
populated config files : /home/runner/.condarc
conda version : 4.12.0
conda-build version : not installed
python version : 3.9.12.final.0
virtual packages : __linux=5.15.0=0
__glibc=2.31=0
__unix=0=0
__archspec=1=x86_64
base environment : /usr/share/miniconda (writable)
conda av data dir : /usr/share/miniconda/etc/conda
conda av metadata url : None
channel URLs : https://conda.anaconda.org/conda-forge/linux-64
https://conda.anaconda.org/conda-forge/noarch
https://repo.anaconda.com/pkgs/main/linux-64
https://repo.anaconda.com/pkgs/main/noarch
https://repo.anaconda.com/pkgs/r/linux-64
https://repo.anaconda.com/pkgs/r/noarch
package cache : /home/runner/conda_pkgs_dir
envs directories : /usr/share/miniconda/envs
/home/runner/.conda/envs
platform : linux-64
user-agent : conda/4.12.0 requests/2.27.1 CPython/3.9.12 Linux/5.15.0-1020-azure ubuntu/20.04.5 glibc/2.31
UID:GID : 1001:121
netrc file : None
offline mode : False
I found this, but their workaround doesn't make sense for me since I'm not using papermill: https://github.com/rstudio/reticulate/issues/1184
I found this, but my paths don't have spaces: https://github.com/rstudio/reticulate/issues/1149
I found this, but their problem includes an entirely reasonable value for path[1], unlike mine: How can I tell R where the conda environment is via a docker image?
The build environment for this is a bit of a bear but I can probably put together a minimum working (/nonworking) example if needed, lmk
I tracked this down to at least two bits of weird/buggy behavior in reticulate and found a workaround: switch from vanilla miniconda to Mambaforge.
The TL;DR seems to be that whatever wacky ubuntu-latest setup-miniconda#v2 environment started putting into meta/history doesn't include a create line, which is what reticulate needs in order to figure out which conda goes with which python, because (1) it ignores the reticulate.conda_binary setting for some reason, and (2) it uses a more restrictive regex to parse the lines of the history file than the regex it uses to select them. Mambaforge does include the create line, so reticulate is happy.
- uses: conda-incubator/setup-miniconda#v2
with:
python-version: 3.9
activate-environment: www-main
miniforge-variant: Mambaforge
miniforge-version: latest
use-mamba: true
use-only-tar-bz2: true # (for caching support)
- name: Update environment
run: mamba env update -n www-main -f environment.yml
I am trying to install and configure Globus Connect Personal for Linux (i have a CentOS 8), following this tutorial. However, when I try to set up Globus connect personal by running ./globusconnectpersonal -start i get this error
Could not find platform independent libraries <prefix>
Could not find platform dependent libraries <exec_prefix>
Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>]
Python path configuration:
PYTHONHOME = (not set)
PYTHONPATH = (not set)
program name = 'gc.py'
isolated = 0
environment = 1
user site = 1
import site = 1
sys._base_executable = ''
sys.base_prefix = '/tmp/build/80754af9/python_1599203911753/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placeho'
sys.base_exec_prefix = '/tmp/build/80754af9/python_1599203911753/_h_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placeho'
sys.executable = ''
sys.prefix = '/tmp/build/80754af9/python_1599203911753/_h_env_placehold_
Subprocess pid 1722896 exited, rc=1
Traceback (most recent call last):
File "./gc-ctrl.py", line 369, in <module>
start(debug=False)
File "./gc-ctrl.py", line 191, in start
send2clients(fds[2:], mesg.encode('utf-8'))
AttributeError: 'bytes' object has no attribute 'encode'
does anybody know what this could mean?
I think there needs to be PYTHONHOME and PYTHONPATH. I created a conda environment with just the correct version of python in it. Then I ran ./globusconnectpersonal inside the conda environment.
Using a conda environment also works for the non-GUI form of globus.
I have not tried setting the paths manually.
I ran into the same problem when I was using Python3.8 in a Miniconda environment. When I disabled conda with:
conda deactivate
Then I could run "globusconnectpersonal -start" with my native Python2.7. I don't know if it was because the client needed Python2 or if conda was interfering, but his resolved the problem for me.
I am unable to setup & run a simple darkflow program. Infact can't even configure darkflow library:
from darkflow.net.build import TFNet
==> ModuleNotFoundError: No module named 'darkflow'
My Target is to run the following program:
from darkflow.net.build import TFNet
import cv2
options = {"model": "cfg/yolo.cfg", "load": "bin/yolo.weights", "threshold": 0.1}
tfnet = TFNet(options)
imgcv = cv2.imread("./test/dog.jpg")
result = tfnet.return_predict(imgcv)
print(result
Please suggest steps so that I could configure darkflow on Jupyter Notebook (with no GPU) and run the above code
Fixed by creating the file in ipynb file in darkflow directory (downloaded from github) and executing the following from the notebook:
!python3 setup.py build_ext --inplace
!pip install -e .
!pip install .
ERROR (theano.sandbox.gpuarray): Could not initialize pygpu, support disabled
Traceback (most recent call last):
File "/home/mesayantan/.local/lib/python2.7/site-packages/theano/sandbox/gpuarray/init.py", line 95, in
init_dev(config.device)
File "/home/mesayantan/.local/lib/python2.7/site-packages/theano/sandbox/gpuarray/init.py", line 46, in init_dev
"Make sure Theano and libgpuarray/pygpu "
RuntimeError: ('Wrong major API version for gpuarray:', -9997, 'Make sure Theano and libgpuarray/pygpu are in sync.')
[Elemwise{exp,no_inplace}()]
Looping 1000 times took 26.248654 seconds
Result is [ 1.23178029 1.61879337 1.52278066 ..., 2.20771813 2.29967761
1.62323284]
Used the cpu
The problem is solved by installing the developer version of theano.
I agree with Sayantan Ghosh.
I used developer version Theano in Anaconda using
git clone https://github.com/Theano/Theano.git
cd Theano
pip install -e .
One more step was specific Pygpu library by
conda install -c mila-udem pygpu
After then I have to put command line in Terminal export THEANO_FLAGS=mode=FAST_RUN,device=cuda,floatX=float32 to solve Old Backend GPU error.
For your reference, my environment, Python 3.6, Cuda 9, Cudnn 7
I am trying to install OpenCL on a machine with the following specs:
CentOS 6.8
Kernel version 2.6.32-642.el6.x86_64
4G of RAM
FPGA board "Nallatech 395 – with Stratix V D8"
aocl version
aocl 16.0.0.211 (Altera SDK for OpenCL, Version 16.0.0 Build 211, Copyright (C) 2016 Altera Corporation)
However, I receive error [Q 2-3] described in the following page:
http://www.alteraforum.com/forum/showthread.php?t=46355
/#aocl install aocl install: Running install from
/root/altera/16.0/hld/board/nalla_pcie/linux64/libexec Using kernel
source files from /usr/src/kernels/2.6.32-642.el6.x86_64 make: ***
/usr/src/kernels/2.6.32-642.el6.x86_64: No such file or directory.
Stop. cp: cannot stat `./aclpci_drv.ko': No such file or directory
FATAL: Module aclpci_drv not found.
The page above suggests to execute the following command:
yum install kernel-devel
However I still receive the same error.
Please advise!
I found a thread with a similar error, the following solution could work for you. However, this solution is based on version 14.
First navigate to the linux/driver of the BSP folder which for you should be located at root/altera/16.0/hld/board/nalla_pcie/linux64/driver
Find and edit the file aclpci.c: At the top of the file add the following lines:
#define __devinit
#define __devexit
Secondly edit aclpci_cmd.c: On line 83 remove the assignment
Change:
result = pci_restore_state(aclpci->pci_dev);
to:
pci_restore_state(aclpci->pci_dev);
On lines 438-440 comment out the following statements:
if (dev->aer_firmware_first) {
return -EIO;
}
You should now be able to run "aocl install" successfully.