The goal of this project is to build a classification model to predict the severity of phone screen damage. To achieve that goal, The model is training using 3 different classes which are:
- Not Damage : - It means the phone screen is not damage.
- Less Damage : - It means the phone screen is less damage
- Severely Damage : - It means the phone screen is severely damage.
- low confidence : It means that the user might upload completely different images, or it's not a picture of a phone.
By making 3 different classes, it would be easier to identify & classify, and achieve the goal, and objective set for this project.
- To check the demo & see the result, please click on this link Here
- The demo might be slow in response, it's due to been deployed in a free hosting resource online.
- To speed up the response of the API is to allocate more resource and move to the paid resource, so that it would be much faster if it's needed.
- Hence, it's served the idea Testing, and Seen the demo
First check if you've python installed in your computer or not. then if you don't have please installed python first, and follow this guideline
pip install virtualenv
To use venv in your project, in your terminal, create a new project folder, cd to the project folder in your terminal, and run the following command:
python<version> -m venv <virtual-environment-name>
mkdir dir_name
cd dirname
python -m venv env
source env/bin/activate
pip install -r requirements.txt
Then run the app locally once you installed all the packaged successfully.
python app.py
Then you go, you can now use the project locally to test it.
Just run accordingly, and add the path of your datasets, and then it would work smoothly. Hence, once you trained a new model, you can just replace the model in the flask app, and it would work accordingly.
