aryn-sdk 0.1.0

The client library for Aryn services

aryn-sdk is a simple client library for interacting with Aryn cloud services.

Aryn Partitioning Service

Partition pdf files with the Aryn Partitioning Service (APS) through aryn-sdk:

from aryn_sdk.partition import partition_file

with open("partition-me.pdf", "rb") as f:
    data = partition_file(
        f,
        use_ocr=True,
        extract_table_structure=True,
        extract_images=True
    )
elements = data['elements']

Convert a partitioned table element to a pandas dataframe for easier use:

from aryn_sdk.partition import partition_file, table_elem_to_dataframe

with open("partition-me.pdf", "rb") as f:
    data = partition_file(
        f,
        use_ocr=True,
        extract_table_structure=True,
        extract_images=True
    )

# Find the first table and convert it to a dataframe
df = None
for element in data['elements']:
    if element['type'] == 'table':
        df = table_elem_to_dataframe(element)
        break

Or convert all partitioned tables to pandas dataframes in one shot:

from aryn_sdk.partition import partition_file, tables_to_pandas

with open("partition-me.pdf", "rb") as f:
    data = partition_file(
        f,
        use_ocr=True,
        extract_table_structure=True,
        extract_images=True
    )
elements_and_tables = tables_to_pandas(data)
dataframes = [table for (element, table) in elements_and_tables if table is not None]

Visualize partitioned documents by drawing on the bounding boxes:

from aryn_sdk.partition import partition_file, draw_with_boxes

with open("partition-me.pdf", "rb") as f:
    data = partition_file(
        f,
        use_ocr=True,
        extract_table_structure=True,
        extract_images=True
    )
page_pics = draw_with_boxes("partition-me.pdf", data, draw_table_cells=True)

from IPython.display import display
display(page_pics[0])

Note: visualizing documents requires poppler, a pdf processing library, to be installed. Instructions for installing poppler can be found here