25daysofserverless

1. In the Azure Portal: Create a function app which you will use in the next steps Under the associated storage account for the function create a container named jokes

2. Clone repo

git clone git@github.com:simonaco/25daysofserverless.git
cd 25daysofserverless
npx func azure functionapp publish <your-function-name> --python

1. Train the model that you will upload to the created storage container

python3.7 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt  # inside the 25daysofserverless folder
python create_model.py        # trains the model for that you will use for your API

1. Go to the storage section in the azure portal and upload the resulting textgenrnn_weights.hdf5

2. This should trigger the function to execute. If everything has worked then you should see no errors in the Live Metrics stream.