The fastest way to assess the predictive potential of a data problem in the industry
Pandora Box enables analysts and data scientists to discover the predictive potential of an industrial process in minutes.
Pandora does quick feasibility analysis of the problems you want to predictively model in the industry (quality prediction, virtual sensors, etc).
Sientia Predictive Power Evaluator (SPPE) can be used to analyze correlations between process variables and quality (linear and non-linear correlation), do hypothesis testing (check causality), calculate lag time, synchronize variables in time, and provide information on the predictive potential of the problem.
You know those times when you need to do a quick feasibility analysis (predictive potential) of a problem and you don't have time to do all that python coding or you don't have anyone available to help you?
If you've ever been in this situation, then SPPE is the perfect app to save you in those moments!
Analyzing whether a quality variable of an industrial process can be predictively modeled is a complex task that requires evaluating a lot of information. SPPE makes this process easier, making all the necessary steps for it more automatic and faster.
Of course, we don't reinvent the wheel. However, we have made this process more democratic and accessible to analysts, engineers and decision makers in the industry, allowing a minimal analysis to be done on the data you want to explore/model.
Such a preliminary analysis, using SPPE, allows the user to have a minimum of necessary information regarding the predictive potential of the problem, facilitating the decision-making on whether or not to proceed with the data project.
SPPE also provides the user with a synchronized dataset ready to be explored more consistently and systematically.
- Clone the project and navigate to the project folder
git clone https://github.com/Aignosi/sientia-predictivepower-evaluation.git
cd <project folder name>
- Run your virtual environment (in this example we will use virtualenv)
python -m venv myenv
Linux
source myenv/bin/activate
Windows
. myenv/Scripts/activate
- Install dependencies
pip install -r requirements.txt
- Run the application in the command terminal
streamlit run sppe_app.py
- A practical example
- User Name: default
- Password: 12345
Loading data into SPPE
Visualizing the data and its basic characteristics
Visualizing and Interpreting Correlations
- Join us at Discord
Open an issue
Send an e-mail to [email protected]
Want to share your findings, export datasets, and assess the predictive potential of a large-scale problem?
enter our lista de espera do Pandorabox Cloud
Our app is open source under the terms of use of the GPL 3.0 license