Deep learning applications for quality control in particle detector construction:
https://arxiv.org/abs/2203.08969
It's important to specify this version of python as Tenserflow has not been working on the latest release
git clone https://github.com/NotJoeMartinez/APDL_Quality_Control
cd APDL_Quality_Control
virtualenv env --python=python3.9.2
source env/bin/activate
pip install -r requirements.txt
If you still get an error installing tenserflow deactivate and remove
pip install --upgrade tensorflow
Flask based API for interacting with the Tenserflow models
curl -X POST -F 'image=@imgs/[PATH/TO/IMAGE].jpg' http://127.0.0.1:5000/get-labels
{
"All Predictions": {
"Broken Wire": "42%",
"Glue": " 1%",
"Good": "12%",
"No Wires": " 0%",
"One Third Wire": "37%",
"Two Third Wires": " 5%",
"Unknown Debris": " 3%"
},
"Image Size": [
640,
480
],
"Predicted Label": "Broken Wire",
"Prediction Confidence": "42%"
}
pip install -r requirements.txt
export FLASK_APP=APDL_QC_API/__init__.py
export FLASK_ENV=development
flask run