Running python 3.7 on miniconda on raspberry pi - python-3.x

I want to use python 3.7 in raspberry pi through miniconda but surfing, I got to know that miniconda provides python version upto 3.4. Some sources said to use berryconda, which I installed but still the conda info states my python version is 3.6. I tried to install python 3.7 through conda install python=3.7. But as u might have guessed that didnt work as well. I do have python 3.7 manually installed but I don't think there is a way to make conda use that version (please correct me if I'm wrong and tell me how).
I want to create an environment with these dependencies, but it seems I'm stuck on the first step (python 3.7):
name: yolov4-cpu
dependencies:
python==3.7
pip
matplotlib
opencv
pip:
opencv-python==4.1.1.26
lxml
tqdm
tensorflow==2.3.0rc0
absl-py
easydict
pillow
Is there a solution for my problem?

Related

How to install Spyder-terminal in Python 3.8 or 3.9 with Spyder v4.3.x or Spyder 5.0.x

Spyder-terminal does not install with Anaconda Navigator running Python v3.8 or v3.9 or v3.10.
My question is "Is there a way to install the Spyder-Terminal Plugin using Anaconda Navigator running Python 3.8 or 3.9 or 3.10, and use Spyder v5.x " ?
Is there a hack for this to run inside Spyder 5.0.x on Python 3.8, 3.9 or 3.10, WITHOUT creating a new virtual environment?
My Software Versions
Spyder-Terminal Version: spyder-terminal-0.5.0 (spyder-ide/win-64)
Python Version: 3.8 or 3.9
Anaconda v2.0.1
Operating system: Windows 10
What I tried:
I tried installing 'conda install -c spyder-ide spyder-terminal' from Anaconda Command Prompt. I did several trials with MANY errors to learn that Spyder-terminal only works with Spyder v4.2 and NOT Spyder v5.0.3.
Spyder-terminal is (nominally) supposed to work in Python 3.9. So I tried upgrading Anaconda base environment to Python 3.9. However, that BROKE the Anaconda base environment, which required uninstalling and reinstalling the entire Anaconda distribution. (What a hassle. Good thing this was on someone else's computer! 8^))
I finally got it to work in a separate conda virtual environment running Python 3.6. Just to be conservative, I used Python 3.6 instead of Python 3.7. (I later installed Spyder-Terminal in a Python 3.7-based virtual environment and it worked fine.)
The steps that finally worked were to Create a new conda environment with Python 3.6 and install spyder-terminal and spyder-notebook there.
I used the following commands...
conda create -n python36 -c conda-forge python=3.6
conda install spyder-terminal -c conda-forge spyder-ide
conda install spyder-notebook -c conda-forge spyder-ide
The correct version of Spyder (for Python 3.6) was installed along with all of its correct dependencies.
What steps will reproduce the problem?
try installing into an Anaconda distribution with Python 3.8 or Python 3.9 kernels
Wait a long time and get many errors.
Rip out a few locks of hair (preferably someone else's)
**What is the expected output?
Spyder-Terminal plugin installs and is available for use.
What do you see instead?
A bunch of errors. The exact errors were lost when I closed the installation console.
Suggestions
The Spyder-terminal plugin failed repeatedly in Anaconda "base" running v3.8.x or a new environment running v3.9.x. I used a new conda virtual environment, running Python 3.6 runtime kernel and install spyder-terminal there.
The Spyder-terminal installer should do version checks for Python and Spyder FIRST, warn the user, and stop if they are not using Python 3.6 or Python 3.7.
Clearer documentation for Spyder-Terminal plugin installation on the Github.com home page (or built into Spyder Menu "Help" will help the "HELP" system. These docs should state WHICH version(s) of Spyder and Python that Spyder-Terminal v0.5.x requires. A dependency table would also be useful.

Jupyter notebook and python3 not working (installed the miniconda 3.7 package)

As the title suggests, I have installed Miniconda using for Python 3.7. I see that I have ipykernel but my jupyter-kernelspec is still only python2.
Is there a simple way to just override this? Also, why might this have happened even though I installed the python3 version of miniconda?
Edit: conda info also shows that python is version 3.7

Installing Spacy is failing with Python 3.7 on Windows 10

Installing Spacy on Windows 10 with pip is failing. Installing any other of a number of modules is working fine for me.
Here is a pastebin containing the error message:
pip install -U spacy
or
python -m pip install spacy
output:
https://pastebin.com/Y9np4veN
I have tried this both with and without virtualenv and it is failing in the same manner either way. I have also already tried installing and updating setuptools. I've ensured that I am using Python 3.7 and pip3. There is no other version of python or pip installed on my PC.
I downgraded to Python 3.6 and was able to successfully install spacy
Using x86-64 ("64-bit") version of Python 3.7 (3.7.4) instead of x86 ("32-bit") one on Windows fixed the issue for me.

Install working rosbag package for Jupyter Notebook (Python 3)

Trying to plot the data from ros topics in a rosbag file. My system by default had Python 2.7 and I installed Python 3.5 just to work with Jupyter notebooks.
Now I have installed several packages twice, once for Python 2 and again for Python 3. Although most packages are compatible with both versions of Python, I am beginning to wonder if all the ros packages are compatible here.
For the image in this question, initially I was getting an error 'No module named 'rosbag_pandas'', then I installed it with pip3. Now I am facing the error shown in the image with regards to rosbag. I have searched high and low for a quick fix but to no avail. I read online that some packages only work with either Python 2.7 or with Python 3.5. Is that what's happening here?
FYI I have installed pyrosbag with pip and pip3 both. Still facing this issue. Could really use some help with this.
I faced the same problem and found a workaround for me. Basically, on top of the usual python 3 installation, install a python 2 kernel for jupyter notebook, as described here, as follows:
python2 -m pip install pip
python2 -m pip install --upgrade ipykernel
python2 -m ipykernel install
Not sure if there is indeed a python 3 package for rosbag but using the python 2 kernel you can import rosbag files to save them in another format for further processing in python 3.

Changing Default Python env.(from 2.7 to 3.3) in Anaconda on Mac OS 10.8

I have the Anaconda distribution installed for Mac. I have Mac OSX 10.8 (Mountain lion). The problem I don't often use Anaconda is because the default Python which it uses is 2.7 while I work on 3.3 or atleast prefer to work on that.
I really like the Spyder IDE of Anaconda. Is there a way I can get the default environment on Anaconda changed to 3.3 instead of 2.7 so that Spyder and iPython all use 3.3 default?
I see the following help from Anaconda site:
$ conda create -n py3k python=3 anaconda
Here python=3 and anaconda are package specifications, and it is the job of the SAT solver inside conda to find a consistent set of packages which satisfies these requirements. As the root environment uses Python 2, we had to specify the major version explicitly.
After adding the binary directory of the newly created environment to the PATH environment variable, which may be done using
$ source activate py3k
My question is I have separately installed Python 3.3 and associated Scientific Python packages like Pandas, numpy, scipy, scikit-learn etc using Homebrew so that it doesn't conflict with my Mac OS default Python 2.7. So now if I run the above Conda commands in Mac Terminal will it interfere with other Python packages I have installed using Homebrew? or will it automatically install/upgrade the python and other packages in the Anaconda library without interfering with either the Homebrew installed Python or Mac OS default Python?
Please advise.
No, the Homebrew and Anaconda Pythons will stay completely independent of one another. Just make sure you don't have PYTHONPATH set, which causes this to not be true.
Also, you should know that Spyder is not available for Python 3 in Anaconda yet, because PySide has not yet been built for Python 3.
I had installed Anaconda with python 2.7, but even after adding python3:
conda create --name Py3 python=3
spyder continued to call python2.7
creating an environment with BOTH spyder and python=3 worked for me:
conda create --name SpyPy3 python=3 spyder

Resources