Tensorflow not running on windows and pycharm

Question

I´m trying to make my first try with Tensorflow using windows 8.1 and Pycharm but I get an Tensorflow error.

I also installed everything in a virtual env with pip and runned the code in the command line with the same result.

Some things I tried

1. I read other posts relating the issue to the msvcp140.dll and I do have the C++ distributable installed.

2. Also found info related to downgrading to python 3.5. I actually use Python 3.7 and wouldn´t like to downgrade. I´m worried other apps won´t work. Can anyone confirm it won´t work with Python greater than 3.5?

3. Also read info about using Conda, but a the same time other info saying to avoid it, naming pip as the officialy supported method.

4. Also found info about my Intel® Pentium® Processor B980 not supporting AVX instructions. Is this a must when using CPU or only when using the GPU?

Any clues? Thanks in advance!

The following is the error message I get:

Using TensorFlow backend. Traceback (most recent call last): File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 58, in from tensorflow.python.pywrap_tensorflow_internal import * File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow\python\pywrap_tensorflow_internal.py", line 28, in _pywrap_tensorflow_internal = swig_import_helper() File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow\python\pywrap_tensorflow_internal.py", line 24, in swig_import_helper _mod = imp.load_module('_pywrap_tensorflow_internal', fp, pathname, description) File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\imp.py", line 243, in load_module return load_dynamic(name, filename, file) File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\imp.py", line 343, in load_dynamic return _load(spec) ImportError: DLL load failed: No se puede encontrar el módulo especificado.

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "C:/Users/Lia love/TestAi/Test1.py", line 4, in from keras.models import Sequential File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\keras__init__.py", line 3, in from . import utils File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\keras\utils__init__.py", line 6, in from . import conv_utils File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\keras\utils\conv_utils.py", line 9, in from .. import backend as K File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\keras\backend__init__.py", line 89, in from .tensorflow_backend import * File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\keras\backend\tensorflow_backend.py", line 5, in import tensorflow as tf File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow__init__.py", line 24, in from tensorflow.python import pywrap_tensorflow # pylint: disable=unused-import File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow\python__init__.py", line 49, in from tensorflow.python import pywrap_tensorflow File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 74, in raise ImportError(msg) ImportError: Traceback (most recent call last): File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 58, in from tensorflow.python.pywrap_tensorflow_internal import * File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow\python\pywrap_tensorflow_internal.py", line 28, in _pywrap_tensorflow_internal = swig_import_helper() File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\site-packages\tensorflow\python\pywrap_tensorflow_internal.py", line 24, in swig_import_helper _mod = imp.load_module('_pywrap_tensorflow_internal', fp, pathname, description) File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\imp.py", line 243, in load_module return load_dynamic(name, filename, file) File "C:\Users\Lia love\AppData\Local\Programs\Python\Python37\lib\imp.py", line 343, in load_dynamic return _load(spec) ImportError: DLL load failed: No se puede encontrar el módulo especificado.

Failed to load the native TensorFlow runtime.

Test code

I estimate this is not a problem about my code, but I include it just in case.

import pandas as pd
from sklearn import preprocessing
from sklearn.model_selection import train_test_split
from keras.models import Sequential
from keras.layers import Dense

df = pd.read_csv("housepricedata.csv")

dataset = df.values

X = dataset[:, 0:10]
Y = dataset[:, 10]

min_max_scaler = preprocessing.MinMaxScaler()
X_scale = min_max_scaler.fit_transform(X)

X_train, X_val_and_test, Y_train, Y_val_and_test = train_test_split(X_scale, Y, test_size=0.3)

X_val, X_test, Y_val, Y_test = train_test_split(X_val_and_test, Y_val_and_test, test_size=0.5)

print("Keras model setup")
model = Sequential([
    Dense(32, activation='relu', input_shape=(10,)),
    Dense(32, activation='relu'),
    Dense(1, activation='sigmoid'),
])

Answer 1

Intel Pentium processors do not support Advanced Vector Instructions ( AVX ) which are needed by TensorFlow if installed from PyPI through :

pip install tensorflow

Since, your CPU doesn't support AVX, you have two options to choose from:

1. Use Anaconda

Anaconda uses conda distribution index which is similar to PyPI. The TensorFlow conda build uses MKL ( Intel Math Kernel Library ). It works without AVX.

Download Anaconda as mentioned here and create a new conda enviroment as mentioned here. Run this command:

conda install tensorflow

2. Use builds from tensorflow-windows-wheel repo.

This repo contains a number of TensorFlow pip wheel files which are build using SSE instead of AVX. SSE build run without any compilation errors. Use this file from the repo.

Hope this helps.

Answer 2

Python virtual environments are used to isolate package installation from the system (recommended)

Make sure these installation:

python3 --version
pip3 --version
virtualenv --version

Create a new virtual environment by choosing a Python interpreter and making a .\venv directory to hold it:

virtualenv --system-site-packages -p python3 ./venv

Activate the virtual environment:

.\venv\Scripts\activate

Install packages within a virtual environment without affecting the host system setup. Start by upgrading pip:

pip install --upgrade pip
pip list  # show packages installed within the virtual environment

pip install --upgrade tensorflow

#Verify the install:
python -c "import tensorflow as tf; tf.enable_eager_execution(); print(tf.reduce_sum(tf.random_normal([1000, 1000])))"

Open your project in pycharm and go to project setting and change the python run time to target this virtual env. Hope this will help.