Running setup.py install for fbprophet … error

Question

I cannot install fbprophet or gcc7.

I have manually installed a precompiled ephem.

Running setup.py install for fbprophet ... error

I have tried with python 3.6 and 3.7. I have tried running as administrator and without.

My anaconda prompt cannot install anything without throwing errors. I would rather use pip.

The problem may be related to pystan.

File "d:\python37\lib\site-packages\pystan\api.py", line 13, in <module> import pystan._api  # stanc wrapper
ImportError: DLL load failed: The specified module could not be found.

I am using windows 10.

Answer 1

Reason: The python distribution on Anaconda3 uses an old version of gcc (4.2.x)

Please use anaconda prompt as administrator

set a new environment for a stan

conda create -n stan python=<your_version> numpy cython

install pystan and gcc inside the virtual environment.

conda activate stan   

or

source activate stan
(stan)  pip install pystan
(stan)  pip install gcc

verify your gcc version:

gcc --version
gcc (GCC) 4.8.5

Answer 2

If all of the answers did not work lets clone pystan and do not use the above solutions:

git clone --recursive https://github.com/stan-dev/pystan.git
cd pystan
python setup.py install

Answer 3

To solve this problem, I uninstalled my existing python 3.7 and anaconda. I re-installed anaconda with one key difference.

I registered Anaconda as my default Python 3.7 during the Anaconda installation. This lets visual studio, PyDev and other programs automatically detect Anaconda as the primary version to use.

Answer 4

Use: The first step is to remove pystan and cache:

pip uninstall fbprophet pystan
pip --no-cache-dir install pystan==2.17  #any version
pip --no-cache-dir install fbprophet==0.2 #any version
conda install Cython --force

pip install pystan
conda install pystan -c conda-forge
conda install -c conda-forge fbprophet

It creates a wheel and update the environment necessary for the package. pip install fbprophet creates the similar issue.

Be sure that pystan is working.

import pystan
model_code = 'parameters {real y;} model {y ~ normal(0,1);}'
model = pystan.StanModel(model_code=model_code)
y = model.sampling().extract()['y']
y.mean()  # with luck the result will be near 0

Use this link: Installing PyStan on windows