Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

RTD failure due to plots involving convolution Kernels

See original GitHub issue

Description

There are RTD failures due to a missing ndim in latest PRs, see:

I believe this may have to do with the latest version of matplotlib (https://github.com/matplotlib/matplotlib/releases/tag/v3.5.2) interacting poorly with how plots for Box1DKernel, Gaussian1DKernel, RickerWavelet1DKernel, and Trapezoid1DKernel are coded.

Note that a simple attempt at fixing this was tried in #13205; however, I think the fix will be more involved than this.

This is related to matplotlib/matplotlib#22973

Issue Analytics

State:
Created a year ago
Comments:15 (15 by maintainers)

Top GitHub Comments

1reaction

WilliamJamiesoncommented, May 18, 2022

This has now been fixed by matplotlib/matplotlib#22975 so the next release of matplotlib should contain the needed fix.

1reaction

WilliamJamiesoncommented, May 4, 2022

@WilliamJamieson - the difference from your minimal example and the real problem is, I think, that Kernel has a .shape defined and matplotlib sees this and treats the object as “sufficiently array-like”; without the shape, the object gets coerced to ndarray, and that works because of the __array__ method.

@mhvk you are exactly correct. See https://github.com/matplotlib/matplotlib/issues/22973#issuecomment-1117845296 where I just posted a better minimal reproducer. My above comment was me doing a first quick attempt.

Top Results From Across the Web

A Self-Adaptive 1D Convolutional Neural Network for Flight ...

In this paper, a novel one-dimension convolutional neural network (CNN) is ... resistive temperature detectors (RTD), and piezoelectric lead–zirconate ...

On the Anatomy of Predictive Models for ... - ACM Digital Library

Our analysis shows that models learned from GTB—and more in general techniques based on Decision Trees—are more effective than other models because of...

International Journal of Heat and Mass Transfer - NSF PAR

This paper addresses the significance of and need for understanding the thermal impulse response func- tion for (1) validating in-situ thermocouple models; ...

On the Anatomy of Predictive Models for Accelerating GPU ...

In this work, we analyze different machine learning techniques and predictive models to accelerate the convolution operator and GEMM. Moreover, ...

A Basic Guide to RTD Measurements - Texas Instruments

RTDs, or resistance temperature detectors, are sensors used to measure ... This means that there is little variation from sensor to sensor because....