This project is part of the seminar presented by myself, [Rodrigo Cunha](https://orcid.org/ 0000-0002-9374-9568), for the discipline MAE5871 - Análise Espectral de Séries Temporais, ministered by Professor Pedro Morettin and Professor Chang Chiann at Instituto de Matemática e Estatística da Universidade de São Paulo (IME-USP).
This project applies the ideas from the article Daily streamflow forecasting in Sobradinho Reservoir using machine learning models coupled with wavelet transform and bootstrapping, by Saraiva et. al (2021). The article main idea is to use Discrete Wavelet Transform (DWT) to decompose the time series before feeding it to machine learning models while training. The results presented by the authors show an increase on both MLP and SVM models' performance. The purpose of the models is to forecast the daily streamflow in Sobradinho Reservoir.
In this seminar, the objective is to apply DWT for forecasting time series based on an article of choice of each student. Henceforth, an idea similar to the applied in the presented article is applied to forecast the Affluent Natural Energy (or Energia Afluente Natural, in portuguese) at the Paraná River basin using MLP models. The data is provided by the Operador Nacional do Sistema (ONS), a government agency responsible for the coordination and monitoring of electric power generation and transmission facilities connected to Brazil's national grid.
The data used in this project is available here.
Follow the instructions here to install Docker Engine for you OS.
Follow the instructions here to install Git for you OS.
Choose the code editor of you preference (PyCharm, Atom, VSCode, etc...) to help you following the tutorial.
Open a terminal (for Windows' users, the Windows Power Shell) to run the following commands.
Remember to use a valid path in your machine, replacing /path/to/folder.
mkdir /path/to/folder && cd /path/to/folder
git clone https://github.com/rodra-go/enaforecast.git
After installing Docker in your machine (see section 1. Prerequisites), open a terminal (for Windows' users, the Windows Power Shell) to run the following commands.
Remember to replace to a valid path in your machine.
cd /path/to/folder
docker build -t enaforecast .
docker run --rm --name enaforecast -dit -p 8888:8888 -p 5000:5000 -v $(pwd):/usr/src/code/ enaforecast
Replace the $(pwd) on the command above for the path to your folder /path/to/folder in the command below, then run it on Windows Power Shell.
docker run --rm --name enaforecast -dit -p 8888:8888 -p 5000:5000 -v C:/path/to/folder:/usr/src/code/ enaforecast
docker exec -it enaforecast bash
After opening a bash in the enaforecast container, run the following commands.
kedro info
kedro jupyter notebook --ip 0.0.0.0
After a while, the command above will print the address for your Jupyter Notebook and the authentication token. You can access the Jupyter Notebook Server with your internet browser, at the local address http://0.0.0.0:8888.
It will be necessary to open a new terminal. After doing that, run the following
cd /path/to/folder
docker exec -it enaforecast bash
tensorboard --logdir . --host 0.0.0.0 --port 5000
You can access tensorboard with your internet browser, at the local address http://0.0.0.0:5000.
This is your new Kedro project, which was generated using Kedro 0.16.6.
Take a look at the Kedro documentation to get started.
In order to get the best out of the template:
- Don't remove any lines from the
.gitignorefile we provide - Make sure your results can be reproduced by following a data engineering convention
- Don't commit data to your repository
- Don't commit any credentials or your local configuration to your repository. Keep all your credentials and local configuration in
conf/local/
Declare any dependencies in src/requirements.txt for pip installation and src/environment.yml for conda installation.
To install them, run:
kedro install
You can run your Kedro project with:
kedro run
Have a look at the file src/tests/test_run.py for instructions on how to write your tests. You can run your tests as follows:
kedro test
To configure the coverage threshold, go to the .coveragerc file.
To generate or update the dependency requirements for your project:
kedro build-reqs
This will copy the contents of src/requirements.txt into a new file src/requirements.in which will be used as the source for pip-compile. You can see the output of the resolution by opening src/requirements.txt.
After this, if you'd like to update your project requirements, please update src/requirements.in and re-run kedro build-reqs.
Further information about project dependencies
Note: Using
kedro jupyterorkedro ipythonto run your notebook provides these variables in scope:context,catalog, andstartup_error.Jupyter, JupyterLab, and IPython are already included in the project requirements by default, so once you have run
kedro installyou will not need to take any extra steps before you use them.
To use Jupyter notebooks in your Kedro project, you need to install Jupyter:
pip install jupyter
After installing Jupyter, you can start a local notebook server:
kedro jupyter notebook
To use JupyterLab, you need to install it:
pip install jupyterlab
You can also start JupyterLab:
kedro jupyter lab
And if you want to run an IPython session:
kedro ipython
You can move notebook code over into a Kedro project structure using a mixture of cell tagging and Kedro CLI commands.
By adding the node tag to a cell and running the command below, the cell's source code will be copied over to a Python file within src/<package_name>/nodes/:
kedro jupyter convert <filepath_to_my_notebook>
Note: The name of the Python file matches the name of the original notebook.
Alternatively, you may want to transform all your notebooks in one go. Run the following command to convert all notebook files found in the project root directory and under any of its sub-folders:
kedro jupyter convert --all
To automatically strip out all output cell contents before committing to git, you can run kedro activate-nbstripout. This will add a hook in .git/config which will run nbstripout before anything is committed to git.
Note: Your output cells will be retained locally.
Further information about building project documentation and packaging your project