Worker for taskbridge which can handle tasks of type classifyimage.
{
...
"type" : "classifyimage",
"worker" : "ROG",
"file" : "123456789",
"data": {
"numberofpredictions" : 2,
"targetlanguage" : "de"
},
...
"result" : {
"predictions" : [
{
"class" : "n02066245",
"name" : "Grauwal",
"probability": 75.4
},
{
"class" : "n02423022",
"name" : "Gazelle",
"probability": 32.8
}
],
"duration" : 12,
"repository" : "https://github.com/hilderonny/taskworker-classifyimage",
"version" : "1.0.0",
"library" : "tensorflow-2.17.0",
"model" : "mobilenetv3large_model.keras"
}
}The type must be classifyimage.
worker contains the unique name of the worker.
The worker expects a file property defining the filename which contains the image to classify.
In the data property you need to define the targetlanguage (en or de). The names of the returned classes will be written in this language. Additionally you need to define in numberofpredictions, how many predictions (the best ones) you want to get in return.
When the worker finishes the task, it sends back a result property. This property is an object. It contains an array predictions. The array size depends on the previosly defined number of predictions. Each element contains a class property. This is one of the classes defined in ILSVRC2011. The nameproperty contains the name of the class in the target language defined before. The probability property contains the probability of the detection between 0 and 100.
First install Python 3.12. The run the following commands in the folder of the downloaded repository.
python3.12 -m venv python-venv
python-venv\Scripts\activate # Windows
source ./python-venv/bin/activate # Linux
pip install tensorflow==2.17.0 pillow==10.4.0Adopt the shell script classifyimage.sh to your needs and create SystemD config files (if you want to run the worker as Linux service).
/etc/systemd/system/taskworker-classifyimage.service:
[Unit]
Description=Task Worker - Image classifier
[Service]
ExecStart=/taskworker-classifyimage/classifyimage.sh
Restart=always
User=user
WorkingDirectory=/taskworker-classifyimage/
[Install]
WantedBy=multi-user.target
Finally register and start the services.
chmod +x ./classifyimage.sh
sudo systemctl daemon-reload
sudo systemctl enable taskworker-classifyimage.service
sudo systemctl start taskworker-classifyimage.service
Running the program the first time, ai models with about 4 GB (depending on the used model) get downloaded automatically.
python classifyimage.py --taskbridgeurl http://127.0.0.1:42000/ --worker RH-WORKBOOK