Duration Oracle Application
User Guide
1. Introduction
This guide provides instructions for setting up and using the Duration Oracle Application. This application is a self-contained desktop app that uses a pre-trained machine-learning model to estimate the duration of projects based on certain parameters.
2. Prerequisites
The user will need Python installed on their computer to run this application. If Python is not installed, it can be downloaded from the official Python website: https://www.python.org/downloads/
The application has been tested on Python 3.7+.
3. Installation
Step 1: Clone the repository or download the application code
If you have Git installed, you can clone the repository by running the following command in the terminal/command prompt:
git clone <https://github.com/OriAshkenazi/DurationOracle>
Alternatively, you can download a zip file of the project from the repository page and extract it.
Step 2: Install the necessary Python packages
Open the terminal/command prompt, navigate to the directory containing the application code, and install the necessary Python packages using pip (Python's package installer). The required packages are Flask and Pywebview. Run the following command:
pip install Flask pywebview keras tensorflow pandas
This installs Flask and Pywebview, which are needed for the application's UI and server functionality, as well as keras and tensorflow for loading and using the pre-trained model, and pandas for handling data.
Step 3: Launch the application
In the terminal/command prompt, navigate to the directory containing the application code. Then, run the Flask application (app.py) by using the following command:
python app.py
This command starts the backend server. Keep this running and open another terminal/command prompt window.
In the new terminal/command prompt window, navigate to the same directory and run the window script (window.py) using the following command:
python window.py
This command opens the application in a new window.
4. Using the Application
The application window contains input fields for entering the parameters based on which the duration of the project will be predicted.
-
Price Estimation [million NIS]: Enter the estimated price of the project in millions of NIS.
-
Complexity: Select the complexity of the project from the options - 'Simple', 'Normal', and 'Complex'.
-
Type: Select the type of the project from the options - 'Building' and 'Infrastructure'.
After filling in the required details, click on the "Predict Duration" button. The application will display the estimated duration of the project in years, months, and days.
If any changes are made to the input fields after a duration prediction has been displayed, the prediction and the "Predict Duration" button will turn red. They will turn green again once a new duration prediction is displayed.
5. Troubleshooting
-
Python or pip not recognized: If the terminal/command prompt does not recognize 'python' or 'pip', it is likely that Python was not added to the PATH during its installation. Refer to Python's official installation instructions and ensure that Python is correctly installed and added to the PATH.
-
Package installation failed: If pip fails to install a package, it might be due to an older version of pip. Try updating pip using python -m pip install --upgrade pip and then install the packages again.
-
The application doesn't start or an error is displayed: Make sure that both the Flask server (app.py) and the application window (window.py) are running. Both need to be started from separate terminal/command prompt windows.
In case of other problems, refer to the error message displayed in the terminal/command prompt, or contact the application support.