NUS AMD Workshop

---

Installation

1. Download and install the Anaconda python distribution, (python3 version, use 64bit unless your machine is old and only supports 32bit). The installation instructions are found here.

2. Clone demo repository

   • Windows

     • Launch Anaconda Prompt
     • Type

       git clone https://github.com/acceleratedmaterials/NUS_AMDworkshop.git

       and press enter (hereafter called "issue" command)
       • If the above command fails with error message 'git' is not recognized ..., you need to install git. One option is to use conda to install git: issue

         conda install -c anaconda git

         Alternatively, check here on installation instructions for git on Windows
     • Change directory by issuing

       cd NUS_AMDworkshop

   • MacOS, Linux

     • Launch a terminal
     • Type

       git clone https://github.com/acceleratedmaterials/NUS_AMDworkshop.git

       and press enter (hereafter called "issue" command)
       • MacOS: If you are using Mac OS 10.13 (High Sierra), you may get the error message: xcrun: error: invalid active developer path (/Library/Developer/CommandLineTools). In this case, reinstall xcode by issuing

         xcode-select --install

         and clone the repository again
     • Change directory by issuing

       cd NUS_AMDworkshop

3. Install additional required packages

   • Note for MacOS/Linux machines: Check you are indeed using the anaconda binaries by issuing

     which python

     You should see outputs with something similar to /User/<your name>/anaconda3/bin/python (For linux machines, this will be /home/... instead of /User/...). If you see this, no further action is required. If instead, you get something like /usr/bin/python (i.e. without the anaconda3), you are using your system's python binaries. In this case, please add /User/<your name>/anaconda3/bin to your $PATH environment variable (Guide) and then proceed to the next step.

   • Windows. In your opened anaconda prompt, issue the following:

     • Create a conda environment:

       conda create -c rdkit -n NUSworkshop rdkit anaconda python=3.6

       You can use any other name in place of NUSworkshop but be consistent hereafter

     • Activate the environment:

       activate NUSworkshop

       Your prompt should change, from C:> to (NUSworkshop)C:>

     • Install the required packages found in the requirements file:

       pip install -r requirements.txt

     • Note: you can deactivate the environment by issuing deactivate. But make sure to activate the environment again before running any demos from the prompt, since the required packages are installed in this environment! For more information on conda environments, click here

   • MacOS, Linux. In your opened terminal, issue the following:

     • Create a conda environment:

       conda create -c rdkit -n NUSworkshop rdkit anaconda python=3.6

       You can use any other name in place of NUSworkshop but be consistent hereafter

     • Activate the environment:

       source activate NUSworkshop

       Your prompt should change, from $ to (NUSworkshop)$

     • Install the required packages found in the requirements file:

       pip install -r requirements.txt

       Note: on MacOS, if the tensorflow installation fails, try issuing

       pip install --ignore-installed --upgrade https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-1.10.1-py3-none-any.whl

     • Note: you can deactivate the environment by issuing source deactivate. But make sure to activate the environment again before running any demos from the terminal, since the required packages are installed in this environment! For more information on conda environments, click here

Getting started

1. Launch Anaconda Navigator and change environment to NUSworkshop by clicking on the drop down menu

2. For demos using Jupyter notebook

   • Launch Jupyter notebook (If "Install" appears in the place of "Launch", click "Install" then "Launch" will appear.)
   • Navigate to the NUS_AMDworkshop_demo folder
   • Navigate to the corresponding subfolders and click on <demo name>.ipynb to begin interacting with the demo (Guide)

3. For demos using Spyder

   • Launch Spyder (If "Install" appears in the place of "Launch", click "Install" then "Launch" will appear.)
   • Navigate to the NUS_AMDworkshop_demo folder
   • Navigate to the corresponding subfolders Run the <demo name>.py files as required from Spyder IDE (Guide)

License

This project is licensed under the MIT license.