FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated

Question

After updating my Numpy and Tensorflow I am getting these kind of warnings. I had already tried these, but nothing works, every suggestion will be appreciated.

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
2018-01-19 17:11:38.695932: I C:\tf_jenkins\home\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 1

This might or might not be your case, but the same warning is also spit out from h5py package:

/home/user/bin/conda3/lib/python3.6/site-packages/h5py/__init__.py:34: 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

For anyone coming here with this problem, it is a known h5py issue, introduced with numpy 1.14. As stated by the devs:

You can ignore the warning, it's not going to cause any issues at the moment, but you should upgrade to the next release of h5py when it becomes available.

... so it's harmless. The fix has just been merged to master. But until the update is released, the workaround is to downgrade numpy to a previous version:

pip install numpy==1.13.0

Update: h5py has released the RC build with the fix. The following command should do it:

pip install h5py==2.8.0rc1

Update (FINAL): there's a full-fledged release now. So you can simply run:

pip install --upgrade h5py

Answer 2

You could upgrade h5py

pip install --upgrade h5py

Answer 3

I had tried with these and it had solved same problem for me , just put these at top of your code

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

Answer 4

This is due to a version conflict between h5py and numpy. All you need to do is degrade your numpy version through command as below:

pip install numpy==1.13.0