MSE-CNN Implementation 1
Code database with the implementation of MSE-CNN, from the paper 'DeepQTMT: A Deep Learning Approach for Fast QTMT-based CU Partition of Intra-mode VVC'
|
Functions | |
echo (string, padding=80) | |
Prints and clears. | |
update_docstring (path) | |
Creates and updates the beginning docstring of python file. | |
functions_used_from_each_module (str_file) | |
Detect the functions associated with each import. | |
multi_thresholding (rs, preds) | |
Variables | |
str | DOCSTRING_BEGINNING_CREATE |
str | DOCSTRING_BEGINNING_UPDATE |
str | STRING_PATH_1 = "updated_docs" |
str | STRING_PATH_1_2 = "created_docs" |
str | STRING_PATH_2 = ".py" |
str | NONE_STRING = "- None" |
str | AUTHOR = "Raul Kevin Viana" |
@package docstring @file utils.py @brief Usefull functions that are used to not directly for training the model or anything related to this work actual implementation. @section libraries_utils Libraries - os - torch - datetime - re @section classes_utils Classes - None @section functions_utils Functions - echo(string, padding=80) - update_docstring(path) - functions_used_from_each_module(str_file) - multi_thresholding(rs, preds) @section global_vars_utils Global Variables - STRING_PATH_1 - STRING_PATH_1_2 - STRING_PATH_2" - NONE_STRING - AUTHOR @section todo_utils TODO - None @section license License MIT License Copyright (c) 2022 Raul Kevin do Espirito Santo Viana Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. @section author_utils Author(s) - Created by Raul Kevin Viana - Last time modified is 2023-01-29 22:22:04.157947
msecnn_raulkviana.utils.echo | ( | string, | |
padding = 80 |
|||
) |
Prints and clears.
[in] | string | String that will be printed |
[in] | padding | Padding to avoid concatenations |
msecnn_raulkviana.utils.functions_used_from_each_module | ( | str_file | ) |
Detect the functions associated with each import.
[in] | str_file | String representation of the file contents |
msecnn_raulkviana.utils.multi_thresholding | ( | rs, | |
preds | |||
) |
Accordingly, the multi-threshold values can be chosen in the following strategies, ensuring the overall prediction accuracy of MSE-CNN. • Case 1 (more time saving): if the average threshold (1/5) * sum(rs) >= 0.4, then τ2 ≥ τ6 ≥ τ3 ≈ τ4 ≈ τ5. • Case 2 (better RD performance): if the average threshold (1/5) * sum(rs) < 0.4, then τ2 ≥ τ4 ≈ τ3 ≈ τ5 ≥ τ6. @brief Implementation of the multi-threshold for MSE-CNN @param [in] rs: Constant to control the minimum amount of threshold probabilities @param [in] preds: Predictions made by the MSE-CNN @param [out] search_RD: Vector with the information of which modes to compute the RD cost and make a decision regarding in which split to make
msecnn_raulkviana.utils.update_docstring | ( | path | ) |
Creates and updates the beginning docstring of python file.
[in] | path | Path to a script |
str msecnn_raulkviana.utils.AUTHOR = "Raul Kevin Viana" |
str msecnn_raulkviana.utils.DOCSTRING_BEGINNING_CREATE |
str msecnn_raulkviana.utils.DOCSTRING_BEGINNING_UPDATE |
str msecnn_raulkviana.utils.NONE_STRING = "- None" |
str msecnn_raulkviana.utils.STRING_PATH_1 = "updated_docs" |
str msecnn_raulkviana.utils.STRING_PATH_1_2 = "created_docs" |
str msecnn_raulkviana.utils.STRING_PATH_2 = ".py" |