Skip to content

CNTK 2.0 Setup

Jump to bottom
sayanpa edited this page Oct 25, 2016 · 24 revisions

Step-by-Step installation

This page will walk through the manual installation steps for binary install one component at a time. This is intended for those users who want to have a customized installation.

If you are a developer intending to contribute to CNTK please follow the instructions to build from sources

Binary installation (manual steps)

This section is for those individuals who want deploy CNTK V2 binary package following manual steps of Windows and Linux.

Steps for Windows

[Note: If you previously installed an earlier version of the CNTK 2.0 Python pip package, you can jump to step 3 to update existing CNTK 2.0 package installation from your Python 3.4 environment]

Step 1: Install pre-requisites

CNTK V2 on windows requires the following prerequisites to be installed on your system. Please install them from the links below:

Step 2: Python setup

  • If you do not have Anaconda environment: install Anaconda Python for Windows

  • If you already have existing Anaconda env or after your have installed the environment above: Create a conda environment, activate the cntk-py34 environment, update the pip version by running the following commands from the anaconda / windows shell

conda create --name cntk-py34 python=3.4.3 numpy scipy jupyter matplotlib
activate cntk-py34
python -m pip install --upgrade pip**

Step 3: Install CNTK (TODO: UPDATE THE PATHS)

Python

Brainscript (optional)

  • Download the required binary package from CNTK Releases page and extract it to your machine.

  • From Windows cmd prompt run: set PATH=%PATH%;[binary folder]\cntk\cntk

Step 4: Download the examples and tutorials

  • Install GIT

  • Clone the CNTK repository (we assume c:\repos\CNTK as the repository root) to get the python examples and tutorials and include examples directory in PYTHONPATH by running:

git clone --recursive https://github.com/Microsoft/CNTK/
cd c:\repos\cntk 
git checkout v2.alpha4 # TODO update
setx PYTHONPATH [CNTK repo root]\bindings\python\examples;%PYTHONPATH%

Step 5: Verify setup

Python

Run the examples from inside the [CNTK clone root]/bindings/python directory, to verify your installation. Run python examples/NumpyInterop/feedforwardNet.py. You should following output on the console:

Minibatch: 0, Train Loss: 0.7915553283691407, Train Evaluation Criterion: 0.48

Minibatch: 20, Train Loss: 0.6266774368286133, Train Evaluation Criterion: 0.48

Minibatch: 40, Train Loss: 1.0378565979003906, Train Evaluation Criterion: 0.64

Minibatch: 60, Train Loss: 0.6558118438720704, Train Evaluation Criterion: 0.56

Brainscript (optional)

Please run the following self contained example:

cd C:\repos\CNTK\Examples\Other\Simple2d

CNTK configFile=Config/Simple.cntk currentDirectory=Data

You should be able to see the results getting printed on the console.

Steps for Linux

[Note: If you previously installed an earlier version of the CNTK 2.0 Python pip package, you can skip steps 1 through 3 below and directly jump to step 4 to update your existing CNTK 2.0 package installation from your Python 3.4 environment]

Step 1: Download and install pre-requisites

Docker users please follow the instructions here. Others please continue reading.

CNTK V2 on Linux requires the following prerequisites to be installed from the links below:

  • C++ Compiler
  • Open MPI IMPORTANT! We strongly recommend to follow Open MPI installation procedure described by the link above to ensure the correct work of CNTK.
  • For GPU systems ensure that you have the latest NVIDIA driver

Step 2: Python setup

  • If you do not have Anaconda environment: install Anaconda Python 3.5 for Linux

  • If you already have existing Anaconda env or after your have installed the environment above, create a conda environment, upgrade the pip package by running (if needed under elevated command prompt)

conda create --name cntk-py34 python=3.4.3 numpy scipy jupyter matplotlib
activate cntk-py34
python -m pip install --upgrade pip

Note: Make sure that this Python version above is what you use for the remainder of the instructions.]

Step 3: Install CNTK2.0 (TODO: UPDATE THE PATHS)

Python

Brainscript

  • Set the following environment variables (we assume that the CNTK archive is extracted to /home/username/cntkbin):
# TODO make sure: if openmpi built yourself, add to LD_LIBRARY_PATH (.../openmpi-1.10.3/lib)
export PATH=/home/username/cntkbin/cntk/bin:$PATH
export LD_LIBRARY_PATH=/home/username/cntkbin/cntk/lib:/home/username/cntkbin/cntk/dependencies/lib:$LD_LIBRARY_PATH

Step 4: Download the examples and tutorials

  • Install GIT

  • Clone the CNTK repository (we assume c:\repos\CNTK as the repository root) to get the python examples and tutorials and include examples directory in PYTHONPATH by running:

    git clone --recursive https://github.com/Microsoft/CNTK/
    cd $HOME/[USERNAME]/repos/cntk 
    git checkout v2.alpha4 # TODO update
    export PYTHONPATH=$PWD/bindings/python:$PYTHONPATH

Step 5: Verify setup

Python

  • Run the examples from inside the [CNTK clone root]/bindings/python directory, to verify your installation. Run python examples/NumpyInterop/feedforwardNet.py. You should following output on the console:
Minibatch: 0, Train Loss: 0.7915553283691407, Train Evaluation Criterion: 0.48

Minibatch: 20, Train Loss: 0.6266774368286133, Train Evaluation Criterion: 0.48

Minibatch: 40, Train Loss: 1.0378565979003906, Train Evaluation Criterion: 0.64

Minibatch: 60, Train Loss: 0.6558118438720704, Train Evaluation Criterion: 0.56
  • Additional test (Optional): You may run the Python test included in the CNTK module
pip install pytest

python -c "import cntk, os; print(os.path.dirname(os.path.abspath(cntk.__file__)))"

pytest [the directory output by the previous command]

Brainscript (optional)

  • Please run the following self contained example:
cd /home/username/repos/CNTK/Examples/Other/Simple2d

cntk configFile=Config/Simple.cntk currentDirectory=Data

You should be able to see the results getting printed on the console.

New Documentation Site

Iteration Plans

Clone this wiki locally