Issue installing Keras with Tensorflow backend on Anaconda

Question

I am trying to install keras, theano and tensorflow on Anaconda. Following are the steps I have followed: Step 1: Created an environment 'nnet'

conda create -nnnet python=3.5 anaconda

Step 2:

activate nnet

Step 3: Installing Spyder on the new environment

conda install spyder

Step 4: Installing the packages

pip install theano
pip install tensorflow
pip install keras

Step 5: update all packages

conda update --all

Step 6: run spyder in nnet environment

Now when I run my code which tries to import keras, I get this error:

ImportError: Importing the multiarray numpy extension module failed. Most likely you are trying to import a failed build of numpy. If you're working with a numpy git repo, try git clean -xdf (removes all files not under version control). Otherwise reinstall numpy.

Original error was: DLL load failed: The specified module could not be found.

Then I uninstalled numpy and reinstalled. However the error still persists.

Also upon checking the packages installed in anaconda navigator under nnet environment, I see tensorflow and theano missing. Here is the list I am referring to

I have used these steps before and have successfully used these packages before.(Imported and used in code) However, I am unable to figure out the issue now. Any sort of help or direction would be appreciated.

Answer 1

I was able to resolve it by modifying the commands in step 4:

Step 4:

conda install -c anaconda theano 
conda install -c conda-forge tensorflow 
conda install -c conda-forge keras 

The packages now work fine and I can import them too.