Inspect data for a parallel workflow configuration¶

inspect_dataset¶

Once you have a WorkflowConfig class object, a saved configuration file, or at the very least an input directory of image data, you may want to check which and how many images would be used running that workflow. This can be helpful in troubleshooting metadata filters among other configuration options.

plantcv.parallel.inspect_dataset(config)

returns summary_df, df - summary_df shows whether images were kept and if not where they why they were removed (status column) then groups images by status and any metadata from config.metadata_filters and returns how many unique values each term has and what they are (if there are 3 or fewer unique values). - df shows all the images collected by the workflow, including the same status information as in summary_df.

• Parameters:
  • config - Plantcv.parallel.WorkflowConfig object or str. String input should be a filepath to a configuration json file or a path to be used in place of config.input_dir. Note that there are several useful defaults expected from a WorkflowConfig object, so results from using just a file path may be less detailed.
• Context:
  • Used to check what would be collected by a given workflow configuration.
• Example use:
  • See below

from plantcv import parallel as pcvpar

config = pcvpar.WorkflowConfig()
config.input_dir = "./my_images"

summary, meta = pcvpar.inspect_dataset(config)