This project utilizes the Acoustic Extinguisher Fire Dataset to predict whether acoustic wave with given features will be able to extinguish a 🔥flame🔥 of a given size. For more information on dataset, features, expirements dataset is based on - refer to the Jupyter notebook.
The model is served using FastAPI/Uvicorn and can be deployed using Docker.
For an interesting demonstration of the underlying concept (though not directly related to dataset), check out these YouTube videos:
For a detailed exploration of the dataset, feature importance analysis, modeling and tuning - refer to the Jupyter notebook included in the repository.
The dataset can be downloaded from Kaggle or the author's website. We use the dataset file provided in the project's GitHub repository, and the code for obtaining the dataset for colab use is included in the notebook.
Clone the repository and navigate to the project directory:
git clone https://github.com/larin92/Acoustic_fire_extinguisher.git
cd Acoustic_fire_extinguisherTo set up the Python environment and install dependencies using pipenv:
pip install pipenv
pipenv installTo run the training script using pipenv:
pipenv run python .\training_script.pyTo serve the model using Uvicorn (without Docker):
pipenv run python .\serve.pyTo build and run the Docker container:
docker build -f Dockerfile -t acoustic_fire_extinguisher:01 .
docker run -d --name acoustic_fire_extinguisher -p 8000:8000 acoustic_fire_extinguisher:01To stop container and clean up:
docker rm $(docker stop acoustic_fire_extinguisher)You can test the served model using curl with the following commands:
- On Unix:
curl -i -X POST -H "Content-Type: application/json" -d '{"SIZE": 1, "FUEL": "Gasoline", "DISTANCE": 10, "DECIBEL": 72, "AIRFLOW": 0, "FREQUENCY": 1}' http://localhost:8000/predict
curl -i -X POST -H "Content-Type: application/json" -d '{"SIZE": 4, "FUEL": "Kerosene", "DISTANCE": 100, "DECIBEL": 92.5, "AIRFLOW": 8.5, "FREQUENCY": 38}' http://localhost:8000/predict
curl -i -X POST -H "Content-Type: application/json" -d '{"SIZE": 1, "FUEL": "Gasoline", "DISTANCE": 10, "DECIBEL": 109, "AIRFLOW": 4.5, "FREQUENCY": 67}' http://localhost:8000/predict- On Windows (use
cmd, notPowerShell):
curl -i -X POST -H "Content-Type: application/json" -d "{\"SIZE\": 1, \"FUEL\": \"Gasoline\", \"DISTANCE\": 10, \"DECIBEL\": 72, \"AIRFLOW\": 0, \"FREQUENCY\": 1}" http://localhost:8000/predict
curl -i -X POST -H "Content-Type: application/json" -d "{\"SIZE\": 4, \"FUEL\": \"Kerosene\", \"DISTANCE\": 100, \"DECIBEL\": 92.5, \"AIRFLOW\": 8.5, \"FREQUENCY\": 38}" http://localhost:8000/predict
curl -i -X POST -H "Content-Type: application/json" -d "{\"SIZE\": 1, \"FUEL\": \"Gasoline\", \"DISTANCE\": 10, \"DECIBEL\": 109, \"AIRFLOW\": 4.5, \"FREQUENCY\": 67}" http://localhost:8000/predict For more information on the dataset and related studies, please refer to the following citations:
1: KOKLU M., TASPINAR Y.S., (2021). Determining the Extinguishing Status of Fuel Flames With Sound Wave by Machine Learning Methods. IEEE Access, 9, pp.86207-86216, Doi: 10.1109/ACCESS.2021.3088612
Link: https://ieeexplore.ieee.org/document/9452168 (Open Access)
https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9452168
2: TASPINAR Y.S., KOKLU M., ALTIN M., (2021). Classification of Flame Extinction Based on Acoustic Oscillations using Artificial Intelligence Methods. Case Studies in Thermal Engineering, 28, 101561, Doi: 10.1016/j.csite.2021.101561
Link: https://www.sciencedirect.com/science/article/pii/S2214157X21007243 (Open Access)
https://www.sciencedirect.com/sdfe/reader/pii/S2214157X21007243/pdf
3: TASPINAR Y.S., KOKLU M., ALTIN M., (2022). Acoustic-Driven Airflow Flame Extinguishing System Design and Analysis of Capabilities of Low Frequency in Different Fuels. Fire Technology, Doi: 10.1007/s10694-021-01208-9
Link: https://link.springer.com/content/pdf/10.1007/s10694-021-01208-9.pdf"