The seventh month of work focused on work in the following areas:

  • Performance analysis
    • Initial performance comparisons of Phylanx to Numpy, Dask, SciKit Learn, and TensorFlow
    • Performance analysis of ALS
  • APEX
    • Adding support for thread parent tracking
  • Visualization
  • HPX version 1.1.0 Release

Organizational activities and results

Development activities and results

  • Algorithms Highlighted: Logistic Regression, ALS, K-Means

  • Initial integration of interactive visualization tools inside Jupyter Notebook:jupyter_notebook_vis_1.png

  • Mapped select Numpy functions to Phylanx primitives when placed inside of Phylanx decorators
  • Added an OpenScop backend (see related Phylanx seminar)
  • HPX
    • Released HPX version 1.1.0
    • Added trait to enable runtime to schedule an action on a new or current thread

Repository activity March 2nd – April 2nd:

Capture.png

Code statistics

Language No. of Files Blank Lines Comment Lines Code Lines
C++           231       9850       3184      44867
C/C++ Header 155       3014       2202      12167
CMake           64       625       528        2521
Python          32       602       407        2032
YAML              2         13         15           79
Dockerfile        1          2          4           32
Markdown          1         11          0           29
INI               1          0          0            8
SUM       487       4117       6340      61735

Impact on other projects

  • Vcpkg Pull Requests:
    • Pull Requests: Updating HPX to V1.1 (#3120)
  • HPX Pull Requests:
    • Apex refactoring with guids (#3250, #3245)
    • Adding trait for actions allowing to make runtime decision on whether to execute it directly (#3254)