The Python OCR SDK supports custom-built APIs.
If your document isn't covered by one of Mindee's Off-the-Shelf APIs, you can create your own API using theAPI Builder.

Quick-Start

from mindee import Client, product

# Init a new client
mindee_client = Client(api_key="my-api-key")

# Load a file from disk
input_doc = mindee_client.source_from_path("/path/to/the/file.ext")
my_endpoint = mindee_client.create_endpoint(
    endpoint_name="my-endpoint",
    account_name="my-account-name",
    version="my-version",
)

# Parse the document as an invoice by passing the appropriate type
result = mindee_client.parse(
    product.CustomV1,
    input_doc,
    endpoint=my_endpoint
)

# Print a brief summary of the parsed data
print(result.document)

Custom Endpoints

You may have noticed in the previous step that in order to access a custom build, you will need to provide an account and an endpoint name at the very least.

Although it is optional, the version number should match the latest version of your build in most use-cases.
If it is not set, it will default to "1".

Field Types

Custom Fields

List Field

A ListField is a special type of custom list that implements the following:

• confidence (float): the confidence score of the field prediction.
• page_id (int): the ID of the page.
• reconstructed (bool): indicates whether or not an object was reconstructed (not extracted as the API gave it).

Since the inner contents can vary, the value isn't accessed through a property, but rather through the following functions:

• contents_list() (-> List[Union[str, float]]): returns a list of values for each element.
• contents_string(separator=" ") (-> str): returns a list of concatenated values, with an optional separator str between them.
• **str()**: returns a string representation of all values, with an empty space between each of them.

Classification Field

A ClassificationField is a special type of custom classification that implements the following:

• value (str): the value of the classification. Corresponds to one of the values specified during training.
• confidence (float): the confidence score of the field prediction.
• **str()**: returns a string representation of all values, with an empty space between each of them.

Attributes

Custom builds always have access to at least two attributes:

Fields

fields (Dict[str: ListListField]):

print(str(result.document.inference.prediction.fields["my-field"]))

Classifications

classifications ([str: ListClassificationField]): The purchase category among predefined classes.

print(str(result.document.inference.prediction.classifications["my-classification"]))

🧪 Custom Line Items

⚠️ Warning: Custom Line Items are an experimental feature, results may vary.

Though not supported directly in the API, sometimes you might need to reconstitute line items by hand.
The library provides a tool for this very purpose:

columns_to_line_items()

The columns_to_line_items() function can be called from the document and page level prediction objects.

It takes the following arguments:

• anchor_names (List[str]): a list of the names of possible anchor (field) candidate for the horizontal placement a line. If all provided anchors are invalid, the LineItemV1 won't be built.
• field_names (List[str]): a list of fields to retrieve the values from
• height_tolerance (float): Optional, the height tolerance used to build the line. It helps when the height of a line can vary unexpectedly.

Example use:

# document-level
response.document.inference.prediction.columns_to_line_items(
  anchor_names,
  field_names,
  0.011 # optional, defaults to 0.01
)

# page-level
response.document.pages[0].prediction.columns_to_line_items(
    anchor_names,
    field_names,
    0.011 # optional, defaults to 0.01
)

It returns a list of CustomLineV1 objects.

CustomlineV1

CustomlineV1 represents a line as it has been read from column fields. It has the following attributes:

• row_number (int): Number of a given line. Starts at 1.
• fields (Dict[str, ListFieldValueV1][]): List of the fields associated with the line, indexed by their column name.
• bbox (BBox): Simple bounding box of the current line representing the 4 minimum & maximum coordinates as float values.

Questions?

Join our Slack