Python-based Single Particle Imaging Simulation for Free-Electron Lasers

Start by downloading a copy of pysingfel package and data folder

git clone https://github.com/chuckie82/pysingfel.git  
cd pysingfel/examples/input  
Download a copy of lcls.tar.gz in the /input folder from https://stanford.box.com/s/e7c30tvhfz0485j2xr48rnrvel8a4yby
tar -xvf lcls.tar.gz

Run unit tests

pytest

Simulate open and closed states of chaperones on an LCLS pnccd detector

python pysingfel/examples/scripts/ExampleMultipleChaperone.py

python [>2.6 >3.5]
numpy >1.10
numba >0.46
six >1.2
scitkit-learn >0.19
pytest >4.5
matplotlib >2.1
setuptools >44.0
h5py >2.6
scipy >1.1
mpi4py >=2.0

Optional:
psana-conda >1.3 (Required for LCLS simulations only)

1. Install LCLS2 on psana Refer to README.md on the webpage https://github.com/slac-lcls/lcls2

2. Install CuPy on psana

git clone https://github.com/cupy/cupy.git
pip install --no-cache-dir .

To quickly install a standalone version of pysingfel with psana2 on Summit, run

./setup/easy_install_summit_psana2.sh

from the root of the package.

This will:

• downloads a fresh copy of Conda;
• creates an environment with Python 3 and all the required packages for psana and pysingfel;
• downloads and installs psana2; and
• installs pysingfel.

To recover the environment, run source setup/env.sh.

1. Install

mkdir test_proxy
cd test_proxy
wget https://nodejs.org/dist/v12.16.3/node-v12.16.3-linux-ppc64le.tar.gz
tar xvfz node-v12.16.3-linux-ppc64le.tar.gz
ln -s node-v12.16.3-linux-ppc64le latest
wget https://raw.githubusercontent.com/ExaFEL/installation/master/proxy.js
wget https://raw.githubusercontent.com/ExaFEL/installation/master/run_proxy.sh
chmod +x run_proxy.sh
export PATH=$PWD/latest/bin:$PATH
npm --version
node --version
npm install http-proxy -save

2. Change proxy listen port -- use your own port number to avoid conflicting with other users.

In proxy.js, change 6749 to any other number.

var httpProxy = require('http-proxy');
httpProxy.createProxyServer({
  target: {
    protocol: 'https:',
    host: 'pswww.slac.stanford.edu',
    port: 443
  },
  changeOrigin: true,
}).listen(6749); 

3. Run the proxy script

./run_proxy.sh

4. Set environment variable for psana2

export LCLS_CALIB_HTTP=http://YOUR_HOST_NAME:LISTEN_PORT_NUMBER/calib_ws

where YOUR_HOST_NAME is your current node (type printenv HOSTNAME to see the name) and LISTEN_PORT_NUMBER is the number that you chose in step 2.