- Projects 1
- Followers 1
Akhil M Anil
Kerala
FWD : Fresh Water Quality Detector
We aim to create an accurate and efficient model that can determine fresh water quality based on various factors such as source, location, season, etc. The repository contains the code and data used in the development of the model, as well as the results and findings of the project. ...learn more
Project status: Published/In Market
oneAPI, Artificial Intelligence
Overview / Usage
FWD: Fresh Water Quality Detector
- FWD is a cutting-edge machine learning application designed with Streamlit to predict freshwater quality for drinking and ecosystem use.
- It uses data analytics and machine learning algorithms to analyze datasets and make predictions about the safety and suitability of freshwater.
- The primary function is to provide a Web application that can showcase the powers of the model which has been generated with the help of Intel® oneAPI **Base Toolkit.**
- Main Features**:**
- Analysis of Quality of Water.
- Batch testing.
- Standard Water Quality Metrics.
Methodology / Approach
** 1. Collect and preprocess data: **
Intel has provided the dataset with fresh water quality analysis, using the Intel DevCloud, the preprocessing is done on the dataset, so that the pH, water elements, and chemical properties of the water are used for feature engineering.
2. **Train a machine learning model: **
Used the Intel® oneAPI Base Toolkit to train a machine learning model on the preprocessed data. XGBoost algorithm was used for the generation of the model, and using techniques such as cross-validation to optimize its performance. The model showcased an accuracy of 86%. And the model was saved as a pickle file.
**3. Develop a Streamlit Web application: **
Used Streamlit to develop a Web application that can showcase the powers of the trained machine learning model. The application allows users to input freshwater quality metrics and receive predictions about the safety and suitability of the water for drinking and ecosystem use.
4. Implement batch testing feature :
Developed batch-testing functionality that allows users to upload large datasets of freshwater quality metrics and receive predictions for all the samples in the dataset.
**5. Implement standard water quality metrics: **
Implement standard metrics for freshwater quality, such as pH, temperature, dissolved oxygen, and nutrient levels, to provide users with a comprehensive analysis of the quality of the water based on the Indian Standard of fresh water.
Technologies Used
Model Creation:
- Intel DevCloud
- Intel OneAPI Base Toolkit
- Intel Pytorch kernel
Web Application:
- Streamlit
- Python
- Streamlit Cloud
Documents and Presentations
Repository
https://github.com/ArjunRAj77/FreshWaterQualityDetector
Other links
Collaborators