​
Parsing Task Properties

​
Required Properties

​
parsed_format
string
default:"base64"

The format of the parsed output.

Allowed values:

  • "base64": Base64 encoded output
  • "markdown": Markdown formatted output
  • "image": Image output

When the parsed_format is set to markdown, make sure to set model and api_key properties.

Checkout the Generation Task for more parameters.

​
file_type
string
default:"pdf"

The format of the input file.

Allowed values:

  • "pdf": PDF files
  • "png": PNG images
  • "jpeg": JPEG images
  • "jpg": JPG images

​
Optional Properties

​
max_depth
integer
default:"5"

Maximum depth for parsing nested structures.

​
directory
string
default:"*"

Directory to search for files to parse.

​
Usage Examples

tasks:
    parse_graphs:
        task_type: parsing
        task_properties:
            parsed_format: markdown
            file_type: pdf
            max_depth: 3
            directory: ./documents
It’s currently work in progress. If there something else you’d like to see, we’d love to know.