https://github.com/gugarosa/opf_siamese
📄 Official implementation regarding the chapter "Learning to Weight Similarity Measures with Siamese Networks: A Case Study on Optimum-Path Forest".
https://github.com/gugarosa/opf_siamese
chapter implementation machine-learning opf siamese-network
Last synced: 7 days ago
JSON representation
📄 Official implementation regarding the chapter "Learning to Weight Similarity Measures with Siamese Networks: A Case Study on Optimum-Path Forest".
- Host: GitHub
- URL: https://github.com/gugarosa/opf_siamese
- Owner: gugarosa
- License: gpl-3.0
- Created: 2020-07-02T16:52:58.000Z (almost 5 years ago)
- Default Branch: master
- Last Pushed: 2022-10-06T12:59:35.000Z (over 2 years ago)
- Last Synced: 2024-10-18T07:40:00.634Z (8 months ago)
- Topics: chapter, implementation, machine-learning, opf, siamese-network
- Language: Python
- Homepage:
- Size: 3.64 MB
- Stars: 0
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Learning to Weight Similarity Measures with Siamese Networks: A Case Study on Optimum-Path Forest
*This repository holds all the necessary code to run the very-same experiments described in the chapter "Learning to Weight Similarity Measures with Siamese Networks: A Case Study on Optimum-Path Forest".*
---
## References
If you use our work to fulfill any of your needs, please cite us:
```BibTex
@incollection{deRosa:22,
title = {Chapter 7 - Learning to weight similarity measures with Siamese networks: a case study on optimum-path forest☆☆The authors appreciate São Paulo Research Foundation (FAPESP) grants #2013/07375-0, #2014/12236-1, #2017/25908-6, #2018/15597-6, #2018/21934-5 and #2019/02205-5, and CNPq grants 307066/2017-7 and 427968/2018-6.},
editor = {Alexandre Xavier Falcão and João Paulo Papa},
booktitle = {Optimum-Path Forest},
publisher = {Academic Press},
pages = {155-173},
year = {2022},
isbn = {978-0-12-822688-9},
doi = {https://doi.org/10.1016/B978-0-12-822688-9.00015-3},
url = {https://www.sciencedirect.com/science/article/pii/B9780128226889000153},
author = {Gustavo H. {de Rosa} and João Paulo Papa}
}
```
---
## Structure
* `data`: Folder containing the OPF file format datasets;
* `outputs`: Folder for saving the output files, such as `.npy`, `.pkl` and `.txt`;
* `utils`
* `loader.py`: Loads OPF file format datasets;
* `similarity.py`: Calculates the similarity between pairs of samples.
---
## Package Guidelines
### Installation
Install all the pre-needed requirements using:
```Python
pip install -r requirements.txt
```
### Data configuration
Please [download](https://www.recogna.tech/files/opf_siamese/data.tar.gz) the datasets in the OPF file format and put then on the `data/` folder.
---
## Usage
### Learn a Similarity Function
The first step is to learn a similarity function based on the training data and create the dataset's squared similarity matrix. To accomplish such a step, one needs to use the following script:
```Python
python learn_similarities.py -h
```
*Note that `-h` invokes the script helper, which assists users in employing the appropriate parameters.*
### Perform the Classification
After learning the similarity, one needs to classify the data using an OPF-based classifier or Scikit-Learn classifiers. Please, use the following scripts to accomplish such a procedure:
```Python
python classify_with_opf.py -h
```
or
```Python
python classify_without_opf.py -h
```
### Process Classification Reports
After conducting the classification task, one needs to process its report into readable outputs. Please, use the following script to accomplish such a procedure:
```Python
python process_report.py -h
```
*Note that this script converts the .pkl reports into readable .txt outputs.*
### Bash Script
Instead of invoking every script to conduct the experiments, it is also possible to use the provided shell script, as follows:
```Bash
./pipeline.sh
```
Such a script will conduct every step needed to accomplish the experimentation used throughout this chapter. Furthermore, one can change any input argument that is defined in the script.
---
## Support
We know that we do our best, but it is inevitable to acknowledge that we make mistakes. If you ever need to report a bug, report a problem, talk to us, please do so! We will be available at our bests at this repository.
---