A seed project for starting machine learning projects, written in python.

• python3

Setup the env and dependencies:

• python3 -m venv env
• source ./env/bin/activate
• pip install -r requirements.txt

To build the model, run ./create-new-model.sh. By default, this will save the model in the /models/ directory.

The network variables can be edited here: create_new_model.py.

Once the model is built, the model can be served using the docker-compose file. This uses tensorflow/serving to serve the model. It can be run using docker-compose up.

Once the model is running inputs can be sent to the model to get a prediction. For the test data set included with this project you could get a prediction with the following request:

curl --location --request POST 'http://localhost:8501/v1/models/model:predict' \
--header 'Content-Type: application/json' \
--data-raw '{
    "instances": [{
        "longitude":[-114.31],
        "latitude":[34.19],
        "housing_median_age":[15],
        "total_rooms":[5612],
        "total_bedrooms":[1283],
        "population":[1015],
        "households":[472],
        "median_income":[1.4936]
    }]
}'