projplot: Utilities for Creating Projection Plots
projplot provides a set of tools to assess whether or not an optimization algorithm has converged to a local optimum. Its main function does this by visualizing the "projection plots" of the objective function f(x) -- that is, by plotting f against each coordinate of x with the other coordinates fixed at the corresponding elements of the candidate optimal solution x_opt.
This package has a similar goal to the R package optimCheck.
Installation
This package is available on PyPi and can be installed as follows:
pip install projplot
Documentation
Available on Read the Docs.
Example
An overview of the package functionality is illustrated with the following example. Let f(x) = x^TAx - 2b^Tx denote a quadratic objective function in x, which is in the d-dimensional real space. If A is a positive-definite d x d matrix, then the unique minimum of f(x) is x_opt =A^{-1}b.
For example, suppose we have
import numpy as np
A = np.array([[3., 2.],
[2., 7.]])
b = np.array([1., 10.])
Then we have that the optimal solution is x_opt = (-0.765, 1.647). Now, projplot allows us to complete a visual check. The following information will need to be provided:
- The objective function (
obj_fun): This can be either a vectorized or non-vectorized function. - Optimal values (
x_opt): This will be the optimal solution for your function. - Upper and lower bounds for each parameter (
x_lims): This will provide an initial range of values to observe. - Parameter names (
x_names): These are the names of your parameters, i.e. theta, mu, sigma - The number of points to plot for each parameter (
n_pts): This is the number of points that each parameter will be evaluated at for their respective plot.
# Optimal values
x_opt = np.array([-0.765, 1.647])
# Upper and lower bounds for each component of x
x_lims = np.array([[-3., 1], [0, 4]])
# Parameter names
x_names = ["x1", "x2"]
# Number of evaluation points per coordinate
n_pts = 10
import projplot as pjp
def obj_fun(x):
'''Compute x'Ax - 2b'x.'''
y = np.dot(np.dot(x.T, A), x) - 2 * np.dot(b, x)
return y
# Obtain plots with vertical x lines
pjp.proj_plot(obj_fun, x_opt=x_opt, x_lims=x_lims,
x_names=x_names, n_pts=n_pts,
opt_vlines=True)
Further Reading
See documentation for advanced use cases and an FAQ.
Metadata
Release files for projplot 1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| projplot-1.0-py3-none-any.whl | Python 3 | none | any | Details for projplot-1.0-py3-none-any.whl |
Release files / projplot-1.0-py3-none-any.whl
| Download URL | projplot-1.0-py3-none-any.whl |
|---|---|
| Size | 5.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
323bc79c88803c6d75da4948bd1590a18739bcfc1bdbb4d8e01e5414fc5f6fb9
|
|
BLAKE2b-256 checksum How to use checksums |
1e1f79610239b679b496853d050ebf52f42a935376a740cef23d75acace55f17
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.1 CPython/3.8.13
|
Security for projplot 1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.