Frequently Asked Questions

What’s Likely the Cause of Installation Failure?

pyEMAPS installation includes version requirements on pip, its setuptools and wheel that make downloading and installing the package correct and fast. These tools and their right versions may not come with your python installation and environment. Without them, pip tries to re-build pyEMAPS from source and will fail due to the fact that pyEMAPS contains extensions modules.

Make sure you have updated versions of pip, setuptools, and wheel:

pip install -U pip setuptools wheel

before trying installation again.

How Do I Install and Run pyEMAPS in Anaconda Environment With Jupyter Notebook?

Installing pyEMAPS in an virtual environment provides isolation and prevents conflicts. Anaconda and python’s own virtual environment are good choices.

Follow the steps below to install and run pyEMAPS in Anaconda with Jupyter Notebooks:

  • Install Anaconda.

    Note

    You may need to make a clean install by doing a “Full Uninstall” of existing Anaconda by deleting old environment and packages folders as old packages can cause dependencies issues.

  • Create a new environment in Anaconsa called ‘pyemaps’ in Anaconda command prompt:

    conda create -n pyemaps python=3.7

    Note

    Python 3.7 is the only supported python version for now, more version support in development.

  • Switch to the new pyemaps environment by activating it:

    conda activate pyemaps

  • Install latest pyemaps in the new environment:

    pip install pyemaps (or pyemaps==X.X.X)

  • Install Jupyter:

    pip install jupyter

  • Run

    conda install ipywidgets
conda install widgetsnbextension
python -m ipykernel install --user --name pyemaps --display-name "pyemaps (python 3.7)"

    Note

    replace –display-name value with your own string if desired.

  • Run Jupyter local server:

    jupyter notebook

    Create a new notebook file to run pyemaps tasks.

What can I do to speed up dynamic diffraction simulations?

pyEMAPS Bloch simulation, e. g. dynamic diffraction simulation costs significant computation resource and operations. As a result, it is much slower than that of kinematic simulation.

While pyEMAPS performance has improved significantly since its inception, there are still rooms for enhancements and we are still looking for opportunities to make constant progress.

Meanwhile, you can also add to this effort in your simulation with pyEMAPS by taking advantages of python features such as parallell processing.

For example:

  • Using python mutiprocessing if your simulation involves multiple controls input.

with concurrent.futures.ProcessPoolExecutor(max_workers=MAX_PROCWORKERS) as e:

      for ec in emclist:
          fs.append(e.submit(cr.generateBloch,
                             disk_size=dsize,
                             sampling = 20,
                             sample_thickness=(1750,1750,100),
                             em_controls = ec))

The above code snippet can be found in samples folder in si_bloch.py.

  • Assisting simulation computation by setting %TMPDIR% environment to a file location where file I/O performance is much higher than that of the normal folder.