Tensor freaking Flow - Anaconda - Python 3 in Windows

Content
  Install
     Anaconda
     PyCharm
     TensorFlow
  Issues
___________________________________________

Install

(1) Download and install Anaconda, as in the video or the link below:

(This video is a really good tutorial, please watch it till the end)

https://medium.com/@GalarnykMichael/install-python-on-windows-anaconda-c63c7c3d1444
Leave the boxes checked and unchecked as default.

💢 If the SETX PATH command return:

  WARNING: The data being saved is truncated to 1024 characters.

Go to Control Panel > System > Environment Variables, and edit PATH manually here:

(2) Download and install PyCharm Community version

Check this box:

See more here: https://medium.com/@GalarnykMichael/setting-up-pycharm-with-anaconda-plus-installing-packages-windows-mac-db2b158bd8c

(3) Install TensorFlow as follow:

https://www.tensorflow.org/install/install_windows#installing_with_anaconda  (💛)

💢 If activate tensorflow fails, install Miniconda in this link:
https://github.com/conda/conda/issues/6053#issuecomment-354473217

💢 NOTE: As when I installed:
Anaconda version for Python was 3.6, however the Python version for TensorFlow was 3.5, as in this command in 💛 :

 C:> conda create -n tensorflow pip python=3.5

Feel free to enter this command. Don't create conda environment by Settings -> Project Interpreter -> ..., this will create a environment with python 3.6, which will create incompatibility later.

More references:
https://pathtogeek.com/installing-tensorflow-on-window-with-pycharm-ide (except the conda environment settings)
___________________________________________

Issues

(1) Warning:

  ... anaconda3\lib\site-packages\h5py\__init__.py:36:                 
  FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. 
  In future, it will be treated as `np.float64 == np.dtype(float).type`. from ._conv import register_converters as _register_converters        

Answer:

https://github.com/h5py/h5py/issues/974#issuecomment-356184944
"This isn't an issue yet (as numpy hasn't made the breaking change), so you can ignore this."

(2) Warning:

 C:\tf_jenkins\workspace\rel-win\M\windows\PY\36\tensorflow\core\platform\cpu_feature_guard.cc:137.
 Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX AVX2

Answer:
(2.1)  https://stackoverflow.com/a/48544614/4505826
Put on top of the code:

import os
os.environ["TF_CPP_MIN_LOG_LEVEL"]="3"

(2.2)  https://stackoverflow.com/a/48486720/4505826
"I just had used warnings() module . I had used these code after your all imports"

import warnings
warnings.filterwarnings('ignore', '.*do not.*',)