I m getting an unforseen error ` ValueError Source path User Kedro #questions

I'm getting an unforseen error - `*ValueError:* So...

Shubham Jha

07/02/2024, 3:53 PM

I'm getting an unforseen error -

*ValueError:* Source path '/Users/user_name/Desktop/science_projects/project_name/src' has to be relative to your project0root '.'

I've tried adding the following but not sure why it is giving this error.

Copy code

project_path = bootstrap_project(
        "."
    )

My project directory structure is standard -

Copy code

/Users/user_name/Desktop/science_projects/project_name/
├── conf/
│   ├── base/
│   │   ├── catalog.yml
│   │   ├── parameters.yml
│   └── local/
├── data/
├── logs/
├── notebooks/
├── src/
│   ├── project_name/
│   │   ├── __init__.py
│   │   ├── pipelines/
│   │   │   ├── recommendation_pipeline/
│   │   │   │   ├── __init__.py
│   │   │   │   ├── nodes.py
│   │   │   │   ├── pipeline.py
│   └── other_modules/
├── pyproject.toml
├── setup.py
└── README.md

Laura Couto

07/02/2024, 4:10 PM

Hi Shubham, could you give a little more context on where you’re receiving this error? What exactly are you trying to do when you get it?

Shubham Jha

07/02/2024, 4:12 PM

I am trying to run my kedro pipeline node. While the pipeline gets created successfully and loads the datacatalog, I get this error when executing the node.

Copy code

INFO    Loading data from params:sample_size (MemoryDataset)...      data_catalog.py:483
          INFO    Running node: evaluate_sample_data_node:                 node.py:340
                evaluate_models([params:model_name;params:train_start_date;params:train_e       
                nd_date;params:sample_size]) -> [sample_evaluation_results]              
          ERROR   Node evaluate_sample_data_node: evaluate_models() -> failed with error: node.py:365
                Source path '/Users/shubham/Desktop/science_projects/project_name/src' has        
                to be relative to your project root '.'.                       
          WARNING  No nodes ran. Repeat the previous command to attempt a new run.

Shubham Jha

07/02/2024, 5:31 PM

I have been able to resolve this error by updating the

kedro catalog

Copy code

def get_kedro_catalog():
    project_path = Path(__file__).resolve().parents[4]
    logger.info(f"Project path: {project_path}")
    bootstrap_project(project_path)
    with KedroSession.create(project_path=project_path) as session:
        context = session.load_context()
        catalog = context.catalog
    return catalog

Open in Slack

Previous Next