Debug Mode for TensorFlow Graphics
Stay organized with collections
Save and categorize content based on your preferences.
Tensorflow Graphics heavily relies on L2 normalized tensors, as well as
trigonometric functions that expect their inputs to be in a certain range.
During optimization, an update can make these variables take values that cause
these functions to return Inf or NaN values. To make debugging such issues
simpler, TensorFlow Graphics provides a debug flag that injects assertions to
the graph to check for the right ranges and the validity of the returned values.
As this can slow down the computations, debug flag is set to False by default.
Users can set the -tfg_debug flag to run their code in debug mode. The flag
can also be set programmatically by first importing these two modules:
from absl import flags
from tensorflow_graphics.util import tfg_flags
and then by adding the following line to the code.
flags.FLAGS[tfg_flags.TFG_ADD_ASSERTS_TO_GRAPH].value = True
Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
Last updated 2021-01-28 UTC.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2021-01-28 UTC."],[],[]]
