RuntimeError: Split pattern data/* 不匹配任何文件

Question

I'm currently trying to convert the TFX-TFRS tutorial into an Airflow pipeline. When I run this Pipeline with LocalDagRunner() it executes without error. When I use the AirflowDagRunner() it throws the following error (Airflow pipeline is triggered through the Airflow Web-UI):

RuntimeError: Split pattern data/* does not match any files.

Inside the data folder resides a CSV dataset.
I use the standard CsvExampleGen component in both cases.

This is the path to the dataset, which gets ingested by the pipeline:

PIPELINE_NAME = 'TFRS-ranking'

# Directory where MovieLens 100K rating data lives
DATA_ROOT = os.path.join('data', PIPELINE_NAME)

Answer 1

我通过为 Airflow 管道提供数据的绝对路径而不是相对路径来解决此问题。