An extension to PyLops for distributed linear operators.
:vertical_traffic_light: :vertical_traffic_light: This library is under early development. Expect things to constantly change until version v1.0.0. :vertical_traffic_light: :vertical_traffic_light:
This library is an extension of PyLops for distributed operators.
Doing so, linear operators can be distributed across several processes on a single node or even across multiple nodes. Their forward and adjoint are first lazily built as directed acyclic graphs and evaluated only when requested by the user (or automatically within one of our solvers).
Here is a simple example showing how a diagonal operator can be created, applied and inverted using PyLops:
import numpy as np from pylops import Diagonal n = 10 x = np.ones(n) d = np.arange(n) + 1 Dop = Diagonal(d) # y = Dx y = Dop*x # x = D'y xadj = Dop.H*y # xinv = D^-1 y xinv = Dop / y
and similarly using PyLops-distributed:
import numpy as np import dask.array as da import pylops_distributed from pylops_distributed import Diagonal # set-up client client = pylops_distributed.utils.backend.dask() n = 10 x = da.ones(n, chunks=(n//2,)) d = da.from_array(np.arange(n) + 1, chunks=(n//2, n//2)) Dop = Diagonal(d) # y = Dx y = Dop*x # x = D'y xadj = Dop.H*y # xinv = D^-1 y xinv = Dop / y da.compute((y, xadj, xinv)) client.close()
It is worth noticing two things at this point:
- in this specific case we did not even need to reimplement the
Derivativeoperator. Calling numpy operations as methods (e.g.,
x.sum()) instead of functions (e.g.,
np.sum(x)) makes it automatic for our operator to act as a distributed operator when a dask array is provided instead. Unfortunately not all numpy functions are also implemented as methods: in those cases we reimplement the operator directly within PyLops-distributed.
.H*is still possible also within PyLops-distributed, however when initializing an operator we will need to decide whether we want to simply create dask graph or also evaluation. This gives flexibility as we can decide if and when apply evaluation using the
computemethod on a dask array of choice.
You need Python 3.5 or greater.
You can also directly install from the master node
pip install https://firstname.lastname@example.org/equinor/pylops-distributed.git@master
Feel like contributing to the project? Adding new operators or tutorial?
Follow the instructions from PyLops official documentation.
The official documentation of PyLops-distributed is available here.
Moreover, if you have installed PyLops using the developer environment you can also build the documentation locally by typing the following command:
Once the documentation is created, you can make any change to the source code and rebuild the documentation by simply typing
Note that if a new example or tutorial is created (and if any change is made to a previously available example or tutorial) you are required to rebuild the entire documentation before your changes will be visible.
PyLops-Distributed was initially written and it is currently maintained by Equinor. It is an extension of PyLops for large-scale optimization with distributed linear operators that can be tailored to our needs, and as contribution to the free software community.
- Matteo Ravasi, mrava87
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, size||File type||Python version||Upload date||Hashes|
|Filename, size pylops_distributed-0.2.0-py3-none-any.whl (52.2 kB)||File type Wheel||Python version py3||Upload date||Hashes View|
|Filename, size pylops_distributed-0.2.0.tar.gz (35.3 kB)||File type Source||Python version None||Upload date||Hashes View|
Hashes for pylops_distributed-0.2.0-py3-none-any.whl