I am trying to run the Breusch-Pagan test in python 3 using the below code. It works perfectly in python 2.7, but when I run it in Anaconda with python 3.6 instead, I get the following the error: "module 'statsmodels.stats.api' has no attribute 'het_breuschpagan'".
I have looked at the statsmodel documentation at this link, https://www.statsmodels.org/devel/generated/statsmodels.stats.diagnostic.het_breuschpagan.html, and know that I am running the right code.
import statsmodels.stats.api as sms
breuschpagan_test = sms.het_breuschpagan(model_run.resid, model.model.exog)
Does anyone know a solution to this or a different way to call this statsmodel function in python 3?
Also, due to limitations at work, I cannot uninstall/re-install or update my statsmodel library at the moment either.
Thanks in advance!
Related
I am trying to use those codes to process satellite images (https://github.com/uw-cryo/skysat_stereo)
The issue i have is that they are using a library which called Imview. You can find the github just here : https://github.com/dshean/imview. This library was made for Python 2.7 and i am using Python 3.8. Does someone knows an equivalent of Imview for python 3.8 ? I will try to replace it but i could not find anything yet based on my knowledge.
If I find something i will let it know in the comments.
Thanks for the help.
I want to run a python script in my flutter app. Thanks to starflut package which helped to do this. Now I want to run a python script that uses NumPy and OpenCV libraries. I want to import these libraries into my python script. I researched a lot about how can I achieve this I couldn't find a way. So I'm posting here so that whoever got the solution for this problem can suggest to me how to do this.
I'm trying to run some scala code in python 3 with jupyter notebook. I have installed pixiedust to make this easier. I have imported it and managed to do that successfully. However, according to the tutorial (https://pixiedust.github.io/pixiedust/scalabridge.html) I should be able to use %scala and then run scala code. This is not working for me and I'm getting an error like this: UsageError: Cell magic %%scala not found.
I have tried with both %scala and %%scala, but neither work.
Does anyone know of a different syntax or how this could work?
Thanks!
I developed a a tool in python 3.5 some time ago which currently only uses differential evolution from scipy to do its task. For some reasons I had to change settings on my machine and switch to use python in a virtual environment.
My specs now:
win 10 64 bit
used pip 9.0.1
numpy 1.12.1+mkl
scipy 0.19.0
python 3.6.1
I have a different env using python 2.7 flying around somewhere else.
Now my problem....everytime the differential evolution function gets its first set of data it crashes after returning the differential_evolution step value.
The differential_evolution call can be found under ./libraries/methods/differential_evoluation.py line 76
The pop-up Error is "Entry point 'mkl_blas_dgemm_alloc wasn't found in 'mkl_intel_thread.dll'." And the printed error is "Intel MKL FATAL ERROR: Cannot load mkl_intel_thread.dll." Please notice that my system language is german therfore the pop-up message was translated by me.
I don't know whether this is relevant but my directory structure is:
>some_place/location1/goal.py
>some_place/location2/env/
I didn't work with virtualenv before and in addition to this i used python 3.5. I'd appreciate any help or instructions on how to add more information to this case to help clarify my problem.
Yours sincerely
OK I'm back with more information. Maybe someone else will stumble uppon it. To be clear: I do not know the fix or reason. I just tried a bunch of stuff.
Using
python 3.5.3,numpy 1.11.1+mkl and scipy-0.18.0 or scipy-0.19.0
made the error disappear. I couldn't try numpy 1.11.1 or 1.11.2 for python 3.6 because this kind soul sadly doesn't offer those versions anymore. I found the 1.11.1 version for python 3.5 somewhere on my disk.
From my testing i can tell that it breaks once i use numpy 1.11.3 or higher. 1.11.1 works fine. Therefore I assume that some changes happened either in numpy between 1.11.1 and 1.11.3 which break it OR in how this kind soul creates/builds his wheels. So I'm going to use python 3.5 for now as i do not have numpy 1.11.1+mkl for python 3.6.
Yours sincerely
I am currently working with converting Pycrypto over to Python 3.X
Whilst I seem to have the cryptography side working the same cannot be said for the tests
provided with the module :(
I have used the tests under Python 2.64 and all works fine.
I then ran '2to3' over the tests to generate new files in 3.X format.
There are several references to the following:
from .common import make_block_tests
Whenever I run the tests I get:
ValueError: Attempted relative import in non-package
If someone would point me towards a way to fix this it would be much appreciated :)
Cheers
Grail
You are trying to run the test files directly, then you can't have relative imports. Change them to be absolute imports, and it will solve the problem.