Currently we only support local mode on Windows. Windows 10.1809 is well tested and recommended.
Anaconda python(64-bit) is highly recommended.
When you use PowerShell to run script for the first time, you need run PowerShell as administrator with this command:
Set-ExecutionPolicy -ExecutionPolicy Unrestricted
-
Install NNI through pip
Prerequisite:
python(64-bit) >= 3.5
python -m pip install --upgrade nni
-
Install NNI through source code
Prerequisite:
python >=3.5
,git
,powershell
git clone -b v0.7 https://github.com/Microsoft/nni.git cd nni powershell ./install.ps1
When these things are done, use the config_windows.yml configuration to start an experiment for validation.
nnictl create --config nni/examples/trials/mnist/config_windows.yml
For other examples you need to change trial command python3
into python
in each example YAML.
Make sure C++ 14.0 compiler installed.
building 'simplejson._speedups' extension error: [WinError 3] The system cannot find the path specified
If you run PowerShell script for the first time and did not set the execution policies for executing the script, you will meet this error below. Try to run PowerShell as administrator with this command first:
Set-ExecutionPolicy -ExecutionPolicy Unrestricted
...cannot be loaded because running scripts is disabled on this system.
This error caused by missing LIBIFCOREMD.DLL and LIBMMD.DLL and fail to install SciPy. Anaconda python is highly recommended. If you use official python, make sure you have one of Visual Studio
, MATLAB
, MKL
and Intel Distribution for Python
installed on Windows before running NNI. If not, try to install one of products above or Anaconda python(64-bit).
ImportError: DLL load failed
Please check the trial log file stderr for more details. If there is no such file and NNI is installed through pip, then you need to run PowerShell as administrator with this command first:
Set-ExecutionPolicy -ExecutionPolicy Unrestricted
If there is a stderr file, please check out. Two possible cases are as follows:
- forget to change the trial command
python3
intopython
in each experiment YAML. - forget to install experiment dependencies such as TensorFlow, Keras and so on.
- SMAC is not supported
- BOHB is supported, make sure C++ 14.0 compiler and dependencies installed successfully.
Note:
- If there is any error like
Segmentation fault
, please refer to FAQ