Example usage of GraphGrid Python SDK
The first step to clone this repository. Using this bash command clone the repository.
git clone https://github.com/graphgrid/graphgrid-sdk-python-examples.gitThe next step is to start running a local CDP deployment.
Download CDP ai-edition (version 2.0) from https://www.graphgrid.com/cdp-downloads/ and visit https://docs.graphgrid.com/ for more information.
We need to build a docker image based on our dockerfile.
This image is what Airflow uses when the DAG is triggered.
Be sure to COPY the necessary files for your DAG.
docker build -t graphgrid-sdk-python-examples .With our DAG image built we now need to upload it to Airflow.
From within your CDP directory run the graphgrid command:
./bin/graphgrid airflow upload <PATH/TO/example_dag.py>Be sure to replace <PATH/TO/example_dag.py> with the actual path to your DAG python file.
Airflow may take up to 1 minute to add new DAGs to the Webserver UI.
We can start training NLP models!
You can use the Airflow Webserver browser to trigger your DAG (CDP defaults this to localhost:8081
and signing in with username/password airflow.
From the home screen you should see your custom DAG (train_model_with_sdk) and the nlp_model_training DAG.
You can manually trigger your custom DAG by hitting the arrow under the Actions column.
You can monitor the jobs manually through the Airflow Webserver, or you can learn about how to use the GraphGrid SDK to programmatically monitor and interact with your model trainings.
Following these steps you will have trained the model(s) specified within your DAG using GraphGrid's NLP model training service.
These trained models can be loaded in and used with the NLP Module of CDP.
Visit the GraphGrid docs site for more detailed information on running your custom NLP model training jobs and learning about the CDP platform.