Source code for azure.ai.formrecognizer._models
# coding=utf-8
# ------------------------------------
# Copyright (c) Microsoft Corporation.
# Licensed under the MIT License.
# ------------------------------------
# pylint: disable=protected-access, too-many-lines
from enum import Enum
from collections import namedtuple
from ._helpers import (
adjust_value_type,
adjust_confidence,
get_element
)
def get_bounding_box(field):
return [
Point(x=field.bounding_box[0], y=field.bounding_box[1]),
Point(x=field.bounding_box[2], y=field.bounding_box[3]),
Point(x=field.bounding_box[4], y=field.bounding_box[5]),
Point(x=field.bounding_box[6], y=field.bounding_box[7])
] if field.bounding_box else None
def resolve_element(element, read_result):
element_type, element, page = get_element(element, read_result)
if element_type == "word":
return FormWord._from_generated(element, page=page)
if element_type == "line":
return FormLine._from_generated(element, page=page)
if element_type == "selectionMark":
return FormSelectionMark._from_generated(element, page=page)
raise ValueError("Failed to parse element reference.")
def get_field_value(field, value, read_result): # pylint: disable=too-many-return-statements
if value is None:
return value
if value.type == "string":
return value.value_string
if value.type == "number":
return value.value_number
if value.type == "integer":
return value.value_integer
if value.type == "date":
return value.value_date
if value.type == "phoneNumber":
return value.value_phone_number
if value.type == "time":
return value.value_time
if value.type == "array":
return [
FormField._from_generated(field, value, read_result)
for value in value.value_array
]
if value.type == "object":
return {
key: FormField._from_generated(key, value, read_result)
for key, value in value.value_object.items()
}
if value.type == "selectionMark":
return value.text # FIXME https://github.com/Azure/azure-sdk-for-python/issues/15276
return None
[docs]class FieldValueType(str, Enum):
"""Semantic data type of the field value.
"""
STRING = "string"
DATE = "date"
TIME = "time"
PHONE_NUMBER = "phoneNumber"
FLOAT = "float"
INTEGER = "integer"
LIST = "list"
DICTIONARY = "dictionary"
SELECTION_MARK = "selectionMark"
[docs]class LengthUnit(str, Enum):
"""The unit used by the width, height and bounding box properties.
For images, the unit is "pixel". For PDF, the unit is "inch".
"""
PIXEL = "pixel"
INCH = "inch"
[docs]class TrainingStatus(str, Enum):
"""Status of the training operation.
"""
SUCCEEDED = "succeeded"
PARTIALLY_SUCCEEDED = "partiallySucceeded"
FAILED = "failed"
[docs]class CustomFormModelStatus(str, Enum):
"""Status indicating the model's readiness for use.
"""
CREATING = "creating"
READY = "ready"
INVALID = "invalid"
[docs]class FormContentType(str, Enum):
"""Content type for upload.
.. versionadded:: v2.1-preview
Support for image/bmp
"""
APPLICATION_PDF = "application/pdf"
IMAGE_JPEG = "image/jpeg"
IMAGE_PNG = "image/png"
IMAGE_TIFF = "image/tiff"
IMAGE_BMP = "image/bmp"
[docs]class Point(namedtuple("Point", "x y")):
"""The x, y coordinate of a point on a bounding box.
:ivar float x: x-coordinate
:ivar float y: y-coordinate
"""
__slots__ = ()
def __new__(cls, x, y):
return super(Point, cls).__new__(cls, x, y)
[docs]class FormPageRange(namedtuple("FormPageRange", "first_page_number last_page_number")):
"""The 1-based page range of the form.
:ivar int first_page_number: The first page number of the form.
:ivar int last_page_number: The last page number of the form.
"""
__slots__ = ()
def __new__(cls, first_page_number, last_page_number):
return super(FormPageRange, cls).__new__(cls, first_page_number, last_page_number)
[docs]class FormElement(object):
"""Base type which includes properties for a form element.
:ivar str text: The text content of the element.
:ivar list[~azure.ai.formrecognizer.Point] bounding_box:
A list of 4 points representing the quadrilateral bounding box
that outlines the text. The points are listed in clockwise
order: top-left, top-right, bottom-right, bottom-left.
Units are in pixels for images and inches for PDF.
:ivar int page_number:
The 1-based number of the page in which this content is present.
:ivar str kind:
The kind of form element. Possible kinds are "word", "line", or "selectionMark" which
correspond to a :class:`~azure.ai.formrecognizer.FormWord` :class:`~azure.ai.formrecognizer.FormLine`,
or :class:`~azure.ai.formrecognizer.FormSelectionMark`, respectively.
"""
def __init__(self, **kwargs):
self.bounding_box = kwargs.get("bounding_box", None)
self.page_number = kwargs.get("page_number", None)
self.text = kwargs.get("text", None)
self.kind = kwargs.get("kind", None)
[docs]class RecognizedForm(object):
"""Represents a form that has been recognized by a trained or prebuilt model.
:ivar str form_type:
The type of form the model identified the submitted form to be.
:ivar str form_type_confidence:
Confidence of the type of form the model identified the submitted form to be.
:ivar str model_id:
Model identifier of model used to analyze form if not using a prebuilt
model.
:ivar fields:
A dictionary of the fields found on the form. The fields dictionary
keys are the `name` of the field. For models trained with labels,
this is the training-time label of the field. For models trained
without labels, a unique name is generated for each field.
:vartype fields: dict[str, ~azure.ai.formrecognizer.FormField]
:ivar ~azure.ai.formrecognizer.FormPageRange page_range:
The first and last page number of the input form.
:ivar list[~azure.ai.formrecognizer.FormPage] pages:
A list of pages recognized from the input document. Contains lines,
words, selection marks, tables and page metadata.
.. versionadded:: v2.1-preview
The *form_type_confidence* and *model_id* properties
"""
def __init__(self, **kwargs):
self.fields = kwargs.get("fields", None)
self.form_type = kwargs.get("form_type", None)
self.page_range = kwargs.get("page_range", None)
self.pages = kwargs.get("pages", None)
self.model_id = kwargs.get('model_id', None)
self.form_type_confidence = kwargs.get('form_type_confidence', None)
def __repr__(self):
return "RecognizedForm(form_type={}, fields={}, page_range={}, pages={}, form_type_confidence={}, " \
"model_id={})" \
.format(
self.form_type,
repr(self.fields),
repr(self.page_range),
repr(self.pages),
self.form_type_confidence,
self.model_id
)[:1024]
[docs]class FormField(object):
"""Represents a field recognized in an input form.
:ivar str value_type: The type of `value` found on FormField. Described in
:class:`~azure.ai.formrecognizer.FieldValueType`, possible types include: 'string',
'date', 'time', 'phoneNumber', 'float', 'integer', 'dictionary', 'list', or 'selectionMark'.
:ivar ~azure.ai.formrecognizer.FieldData label_data:
Contains the text, bounding box, and field elements for the field label.
Note that this is not returned for forms analyzed by models trained with labels.
:ivar ~azure.ai.formrecognizer.FieldData value_data:
Contains the text, bounding box, and field elements for the field value.
:ivar str name: The unique name of the field or the training-time label if
analyzed from a custom model that was trained with labels.
:ivar value:
The value for the recognized field. Its semantic data type is described by `value_type`.
:vartype value: str, int, float, :class:`~datetime.date`, :class:`~datetime.time`,
dict[str, :class:`~azure.ai.formrecognizer.FormField`], or list[:class:`~azure.ai.formrecognizer.FormField`]
:ivar float confidence:
Measures the degree of certainty of the recognition result. Value is between [0.0, 1.0].
"""
def __init__(self, **kwargs):
self.value_type = kwargs.get("value_type", None)
self.label_data = kwargs.get("label_data", None)
self.value_data = kwargs.get("value_data", None)
self.name = kwargs.get("name", None)
self.value = kwargs.get("value", None)
self.confidence = kwargs.get("confidence", None)
@classmethod
def _from_generated(cls, field, value, read_result):
return cls(
value_type=adjust_value_type(value.type) if value else None,
label_data=None, # not returned with receipt/supervised
value_data=FieldData._from_generated(value, read_result),
value=get_field_value(field, value, read_result),
name=field,
confidence=adjust_confidence(value.confidence) if value else None,
)
@classmethod
def _from_generated_unlabeled(cls, field, idx, page, read_result):
return cls(
value_type="string", # unlabeled only returns string
label_data=FieldData._from_generated_unlabeled(field.key, page, read_result),
value_data=FieldData._from_generated_unlabeled(field.value, page, read_result),
value=field.value.text,
name="field-" + str(idx),
confidence=adjust_confidence(field.confidence),
)
def __repr__(self):
return "FormField(value_type={}, label_data={}, value_data={}, name={}, value={}, confidence={})" \
.format(
self.value_type,
repr(self.label_data),
repr(self.value_data),
self.name,
repr(self.value),
self.confidence
)[:1024]
[docs]class FieldData(object):
"""Contains the data for the form field. This includes the text,
location of the text on the form, and a collection of the
elements that make up the text.
:ivar int page_number:
The 1-based number of the page in which this content is present.
:ivar str text: The string representation of the field or value.
:ivar list[~azure.ai.formrecognizer.Point] bounding_box:
A list of 4 points representing the quadrilateral bounding box
that outlines the text. The points are listed in clockwise
order: top-left, top-right, bottom-right, bottom-left.
Units are in pixels for images and inches for PDF.
:ivar field_elements:
When `include_field_elements` is set to true, a list of
elements constituting this field or value is returned. The list
constitutes of elements such as lines, words, and selection marks.
:vartype field_elements: list[Union[~azure.ai.formrecognizer.FormElement, ~azure.ai.formrecognizer.FormWord,
~azure.ai.formrecognizer.FormLine, ~azure.ai.formrecognizer.FormSelectionMark]]
.. versionadded:: v2.1-preview
*FormSelectionMark* is added to the types returned in the list of field_elements
"""
def __init__(self, **kwargs):
self.page_number = kwargs.get("page_number", None)
self.text = kwargs.get("text", None)
self.bounding_box = kwargs.get("bounding_box", None)
self.field_elements = kwargs.get("field_elements", None)
@classmethod
def _from_generated(cls, field, read_result):
if field is None or all(field_data is None for field_data in [field.page, field.text, field.bounding_box]):
return None
return cls(
page_number=field.page,
text=field.text,
bounding_box=get_bounding_box(field),
field_elements=[resolve_element(element, read_result) for element in field.elements]
if field.elements else None
)
@classmethod
def _from_generated_unlabeled(cls, field, page, read_result):
return cls(
page_number=page,
text=field.text,
bounding_box=get_bounding_box(field),
field_elements=[resolve_element(element, read_result) for element in field.elements]
if field.elements else None
)
def __repr__(self):
return "FieldData(page_number={}, text={}, bounding_box={}, field_elements={})" \
.format(
self.page_number,
self.text,
self.bounding_box,
repr(self.field_elements)
)[:1024]
[docs]class FormPage(object):
"""Represents a page recognized from the input document. Contains lines,
words, selection marks, tables and page metadata.
:ivar int page_number:
The 1-based number of the page in which this content is present.
:ivar float text_angle:
The general orientation of the text in clockwise direction, measured in
degrees between (-180, 180].
:ivar float width:
The width of the image/PDF in pixels/inches, respectively.
:ivar float height:
The height of the image/PDF in pixels/inches, respectively.
:ivar str unit:
The :class:`~azure.ai.formrecognizer.LengthUnit` used by the width,
height, and bounding box properties. For images, the unit is "pixel".
For PDF, the unit is "inch".
:ivar list[~azure.ai.formrecognizer.FormTable] tables:
A list of extracted tables contained in a page.
:ivar list[~azure.ai.formrecognizer.FormLine] lines:
When `include_field_elements` is set to true, a list of recognized text lines is returned.
For calls to recognize content, this list is always populated. The maximum number of lines
returned is 300 per page. The lines are sorted top to bottom, left to right, although in
certain cases proximity is treated with higher priority. As the sorting order depends on
the detected text, it may change across images and OCR version updates. Thus, business
logic should be built upon the actual line location instead of order.
:ivar selection_marks: List of selection marks extracted from the page.
:vartype selection_marks: list[~azure.ai.formrecognizer.FormSelectionMark]
.. versionadded:: v2.1-preview
*selection_marks* property
"""
def __init__(self, **kwargs):
self.page_number = kwargs.get("page_number", None)
self.text_angle = kwargs.get("text_angle", None)
self.width = kwargs.get("width", None)
self.height = kwargs.get("height", None)
self.unit = kwargs.get("unit", None)
self.tables = kwargs.get("tables", None)
self.lines = kwargs.get("lines", None)
self.selection_marks = kwargs.get("selection_marks", None)
def __repr__(self):
return "FormPage(page_number={}, text_angle={}, width={}, height={}, unit={}, tables={}, lines={}," \
"selection_marks={})" \
.format(
self.page_number,
self.text_angle,
self.width,
self.height,
self.unit,
repr(self.tables),
repr(self.lines),
repr(self.selection_marks)
)[:1024]
[docs]class FormLine(FormElement):
"""An object representing an extracted line of text.
:ivar str text: The text content of the line.
:ivar list[~azure.ai.formrecognizer.Point] bounding_box:
A list of 4 points representing the quadrilateral bounding box
that outlines the text. The points are listed in clockwise
order: top-left, top-right, bottom-right, bottom-left.
Units are in pixels for images and inches for PDF.
:ivar list[~azure.ai.formrecognizer.FormWord] words:
A list of the words that make up the line.
:ivar int page_number:
The 1-based number of the page in which this content is present.
:ivar str kind: For FormLine, this is "line".
:ivar appearance: Text appearance properties.
:vartype appearance: ~azure.ai.formrecognizer.Appearance
.. versionadded:: v2.1-preview
*appearance* property
"""
def __init__(self, **kwargs):
super(FormLine, self).__init__(kind="line", **kwargs)
self.words = kwargs.get("words", None)
self.appearance = kwargs.get("appearance", None)
@classmethod
def _from_generated(cls, line, page):
line_appearance = line.appearance if hasattr(line, "appearance") else None
return cls(
text=line.text,
bounding_box=get_bounding_box(line),
page_number=page,
words=[FormWord._from_generated(word, page) for word in line.words]
if line.words else None,
appearance=line_appearance
)
def __repr__(self):
return "FormLine(text={}, bounding_box={}, words={}, page_number={}, kind={}, appearance={})" \
.format(
self.text,
self.bounding_box,
repr(self.words),
self.page_number,
self.kind,
self.appearance
)[:1024]
[docs]class FormWord(FormElement):
"""Represents a word recognized from the input document.
:ivar str text: The text content of the word.
:ivar list[~azure.ai.formrecognizer.Point] bounding_box:
A list of 4 points representing the quadrilateral bounding box
that outlines the text. The points are listed in clockwise
order: top-left, top-right, bottom-right, bottom-left.
Units are in pixels for images and inches for PDF.
:ivar float confidence:
Measures the degree of certainty of the recognition result. Value is between [0.0, 1.0].
:ivar int page_number:
The 1-based number of the page in which this content is present.
:ivar str kind: For FormWord, this is "word".
"""
def __init__(self, **kwargs):
super(FormWord, self).__init__(kind="word", **kwargs)
self.confidence = kwargs.get("confidence", None)
@classmethod
def _from_generated(cls, word, page):
return cls(
text=word.text,
bounding_box=get_bounding_box(word),
confidence=adjust_confidence(word.confidence),
page_number=page
)
def __repr__(self):
return "FormWord(text={}, bounding_box={}, confidence={}, page_number={}, kind={})" \
.format(
self.text,
self.bounding_box,
self.confidence,
self.page_number,
self.kind
)[:1024]
[docs]class FormSelectionMark(FormElement):
"""Information about the extracted selection mark.
:ivar str text: The text content - not returned for FormSelectionMark.
:ivar list[~azure.ai.formrecognizer.Point] bounding_box:
A list of 4 points representing the quadrilateral bounding box
that outlines the text. The points are listed in clockwise
order: top-left, top-right, bottom-right, bottom-left.
Units are in pixels for images and inches for PDF.
:ivar float confidence:
Measures the degree of certainty of the recognition result. Value is between [0.0, 1.0].
:ivar str state: State of the selection mark. Possible values include: "selected",
"unselected".
:ivar int page_number:
The 1-based number of the page in which this content is present.
:ivar str kind: For FormSelectionMark, this is "selectionMark".
"""
def __init__(
self,
**kwargs
):
super(FormSelectionMark, self).__init__(kind="selectionMark", **kwargs)
self.confidence = kwargs['confidence']
self.state = kwargs['state']
@classmethod
def _from_generated(cls, mark, page):
return cls(
confidence=adjust_confidence(mark.confidence),
state=mark.state,
bounding_box=get_bounding_box(mark),
page_number=page
)
def __repr__(self):
return "FormSelectionMark(text={}, bounding_box={}, confidence={}, page_number={}, state={})" \
.format(
self.text,
self.bounding_box,
self.confidence,
self.page_number,
self.state
)[:1024]
[docs]class FormTable(object):
"""Information about the extracted table contained on a page.
:ivar int page_number:
The 1-based number of the page in which this table is present.
:ivar list[~azure.ai.formrecognizer.FormTableCell] cells:
List of cells contained in the table.
:ivar int row_count:
Number of rows in table.
:ivar int column_count:
Number of columns in table.
:ivar list[~azure.ai.formrecognizer.Point] bounding_box:
A list of 4 points representing the quadrilateral bounding box
that outlines the table. The points are listed in clockwise
order: top-left, top-right, bottom-right, bottom-left.
Units are in pixels for images and inches for PDF.
.. versionadded:: v2.1-preview
The *bounding_box* property.
"""
def __init__(self, **kwargs):
self.page_number = kwargs.get("page_number", None)
self.cells = kwargs.get("cells", None)
self.row_count = kwargs.get("row_count", None)
self.column_count = kwargs.get("column_count", None)
self.bounding_box = kwargs.get("bounding_box", None)
def __repr__(self):
return "FormTable(page_number={}, cells={}, row_count={}, column_count={}, bounding_box={})" \
.format(
self.page_number,
repr(self.cells),
self.row_count,
self.column_count,
self.bounding_box
)[:1024]
[docs]class FormTableCell(object): # pylint:disable=too-many-instance-attributes
"""Represents a cell contained in a table recognized from the input document.
:ivar str text: Text content of the cell.
:ivar int row_index: Row index of the cell.
:ivar int column_index: Column index of the cell.
:ivar int row_span: Number of rows spanned by this cell.
:ivar int column_span: Number of columns spanned by this cell.
:ivar list[~azure.ai.formrecognizer.Point] bounding_box:
A list of 4 points representing the quadrilateral bounding box
that outlines the text. The points are listed in clockwise
order: top-left, top-right, bottom-right, bottom-left.
Units are in pixels for images and inches for PDF.
:ivar float confidence:
Measures the degree of certainty of the recognition result. Value is between [0.0, 1.0].
:ivar bool is_header: Whether the current cell is a header cell.
:ivar bool is_footer: Whether the current cell is a footer cell.
:ivar int page_number:
The 1-based number of the page in which this content is present.
:ivar field_elements:
When `include_field_elements` is set to true, a list of
elements constituting this cell is returned. The list
constitutes of elements such as lines, words, and selection marks.
For calls to begin_recognize_content(), this list is always populated.
:vartype field_elements: list[Union[~azure.ai.formrecognizer.FormElement, ~azure.ai.formrecognizer.FormWord,
~azure.ai.formrecognizer.FormLine, ~azure.ai.formrecognizer.FormSelectionMark]]
.. versionadded:: v2.1-preview
*FormSelectionMark* is added to the types returned in the list of field_elements
"""
def __init__(self, **kwargs):
self.text = kwargs.get("text", None)
self.row_index = kwargs.get("row_index", None)
self.column_index = kwargs.get("column_index", None)
self.row_span = kwargs.get("row_span", 1)
self.column_span = kwargs.get("column_span", 1)
self.bounding_box = kwargs.get("bounding_box", None)
self.confidence = kwargs.get("confidence", None)
self.is_header = kwargs.get("is_header", False)
self.is_footer = kwargs.get("is_footer", False)
self.page_number = kwargs.get("page_number", None)
self.field_elements = kwargs.get("field_elements", None)
@classmethod
def _from_generated(cls, cell, page, read_result):
return cls(
text=cell.text,
row_index=cell.row_index,
column_index=cell.column_index,
row_span=cell.row_span or 1,
column_span=cell.column_span or 1,
bounding_box=get_bounding_box(cell),
confidence=adjust_confidence(cell.confidence),
is_header=cell.is_header or False,
is_footer=cell.is_footer or False,
page_number=page,
field_elements=[resolve_element(element, read_result) for element in cell.elements]
if cell.elements else None
)
def __repr__(self):
return "FormTableCell(text={}, row_index={}, column_index={}, row_span={}, column_span={}, " \
"bounding_box={}, confidence={}, is_header={}, is_footer={}, page_number={}, field_elements={})" \
.format(
self.text,
self.row_index,
self.column_index,
self.row_span,
self.column_span,
self.bounding_box,
self.confidence,
self.is_header,
self.is_footer,
self.page_number,
repr(self.field_elements)
)[:1024]
[docs]class CustomFormModel(object):
"""Represents a model trained from custom forms.
:ivar str model_id: The unique identifier of this model.
:ivar str status:
Status indicating the model's readiness for use,
:class:`~azure.ai.formrecognizer.CustomFormModelStatus`.
Possible values include: 'creating', 'ready', 'invalid'.
:ivar ~datetime.datetime training_started_on:
The date and time (UTC) when model training was started.
:ivar ~datetime.datetime training_completed_on:
Date and time (UTC) when model training completed.
:ivar list[~azure.ai.formrecognizer.CustomFormSubmodel] submodels:
A list of submodels that are part of this model, each of
which can recognize and extract fields from a different type of form.
:ivar list[~azure.ai.formrecognizer.FormRecognizerError] errors:
List of any training errors.
:ivar list[~azure.ai.formrecognizer.TrainingDocumentInfo] training_documents:
Metadata about each of the documents used to train the model.
:ivar str model_name: Optional user defined model name.
:ivar properties: Optional model properties.
:vartype properties: ~azure.ai.formrecognizer.CustomFormModelProperties
.. versionadded:: v2.1-preview
The *model_name* and *properties* properties.
"""
def __init__(self, **kwargs):
self.model_id = kwargs.get("model_id", None)
self.status = kwargs.get("status", None)
self.training_started_on = kwargs.get("training_started_on", None)
self.training_completed_on = kwargs.get("training_completed_on", None)
self.submodels = kwargs.get("submodels", None)
self.errors = kwargs.get("errors", None)
self.training_documents = kwargs.get("training_documents", None)
self.model_name = kwargs.get("model_name", None)
self.properties = kwargs.get("properties", None)
@classmethod
def _from_generated(cls, model, api_version):
model_name = model.model_info.model_name if hasattr(model.model_info, "model_name") else None
properties = CustomFormModelProperties._from_generated(model.model_info) \
if hasattr(model.model_info, "attributes") else None
return cls(
model_id=model.model_info.model_id,
status=model.model_info.status,
training_started_on=model.model_info.created_date_time,
training_completed_on=model.model_info.last_updated_date_time,
submodels=CustomFormSubmodel._from_generated_unlabeled(model)
if model.keys else CustomFormSubmodel._from_generated_labeled(
model, api_version, model_name=model_name
),
errors=FormRecognizerError._from_generated(model.train_result.errors)
if model.train_result else None,
training_documents=TrainingDocumentInfo._from_generated(model.train_result)
if model.train_result else None,
properties=properties,
model_name=model_name
)
@classmethod
def _from_generated_composed(cls, model):
return cls(
model_id=model.model_info.model_id,
status=model.model_info.status,
training_started_on=model.model_info.created_date_time,
training_completed_on=model.model_info.last_updated_date_time,
submodels=CustomFormSubmodel._from_generated_composed(model),
training_documents=TrainingDocumentInfo._from_generated_composed(model),
properties=CustomFormModelProperties._from_generated(model.model_info),
model_name=model.model_info.model_name
)
def __repr__(self):
return "CustomFormModel(model_id={}, status={}, training_started_on={}, training_completed_on={}, " \
"submodels={}, errors={}, training_documents={}, model_name={}, properties={})" \
.format(
self.model_id,
self.status,
self.training_started_on,
self.training_completed_on,
repr(self.submodels),
repr(self.errors),
repr(self.training_documents),
self.model_name,
repr(self.properties)
)[:1024]
[docs]class CustomFormSubmodel(object):
"""Represents a submodel that extracts fields from a specific type of form.
:ivar str model_id: Model identifier of the submodel.
:ivar float accuracy: The mean of the model's field accuracies.
:ivar fields: A dictionary of the fields that this submodel will recognize
from the input document. The fields dictionary keys are the `name` of
the field. For models trained with labels, this is the training-time
label of the field. For models trained without labels, a unique name
is generated for each field.
:vartype fields: dict[str, ~azure.ai.formrecognizer.CustomFormModelField]
:ivar str form_type: Type of form this submodel recognizes.
.. versionadded:: v2.1-preview
The *model_id* property
"""
def __init__(self, **kwargs):
self.model_id = kwargs.get("model_id", None)
self.accuracy = kwargs.get("accuracy", None)
self.fields = kwargs.get("fields", None)
self.form_type = kwargs.get("form_type", None)
@classmethod
def _from_generated_unlabeled(cls, model):
return [
cls(
model_id=model.model_info.model_id,
accuracy=None,
fields=CustomFormModelField._from_generated_unlabeled(fields),
form_type="form-" + cluster_id
) for cluster_id, fields in model.keys.clusters.items()
]
@classmethod
def _from_generated_labeled(cls, model, api_version, model_name):
if api_version == "2.0":
form_type = "form-" + model.model_info.model_id
elif model_name:
form_type = "custom:" + model_name
else:
form_type = "custom:" + model.model_info.model_id
return [
cls(
model_id=model.model_info.model_id,
accuracy=model.train_result.average_model_accuracy,
fields={field.field_name: CustomFormModelField._from_generated_labeled(field)
for field in model.train_result.fields} if model.train_result.fields else None,
form_type=form_type
)
] if model.train_result else None
@classmethod
def _from_generated_composed(cls, model):
return [
cls(
accuracy=train_result.average_model_accuracy,
fields={field.field_name: CustomFormModelField._from_generated_labeled(field)
for field in train_result.fields} if train_result.fields else None,
form_type="custom:" + train_result.model_id,
model_id=train_result.model_id
) for train_result in model.composed_train_results
]
def __repr__(self):
return "CustomFormSubmodel(accuracy={}, model_id={}, fields={}, form_type={})" \
.format(
self.accuracy,
self.model_id,
repr(self.fields),
self.form_type,
)[:1024]
[docs]class CustomFormModelField(object):
"""A field that the model will extract from forms it analyzes.
:ivar str label: The form fields label on the form.
:ivar str name: Canonical name; uniquely identifies a field within the form.
:ivar float accuracy: The estimated recognition accuracy for this field.
"""
def __init__(self, **kwargs):
self.label = kwargs.get("label", None)
self.name = kwargs.get("name", None)
self.accuracy = kwargs.get("accuracy", None)
@classmethod
def _from_generated_labeled(cls, field):
return cls(
name=field.field_name,
accuracy=field.accuracy
)
@classmethod
def _from_generated_unlabeled(cls, fields):
return {
"field-{}".format(idx): cls(
name="field-{}".format(idx),
label=field_name,
) for idx, field_name in enumerate(fields)
}
def __repr__(self):
return "CustomFormModelField(label={}, name={}, accuracy={})" \
.format(
self.label,
self.name,
self.accuracy
)[:1024]
[docs]class TrainingDocumentInfo(object):
"""Report for an individual document used for training
a custom model.
:ivar str name:
The name of the document.
:ivar str status:
The :class:`~azure.ai.formrecognizer.TrainingStatus`
of the training operation. Possible values include:
'succeeded', 'partiallySucceeded', 'failed'.
:ivar int page_count:
Total number of pages trained.
:ivar list[~azure.ai.formrecognizer.FormRecognizerError] errors:
List of any errors for document.
:ivar str model_id:
The model ID that used the document to train.
.. versionadded:: v2.1-preview
The *model_id* property
"""
def __init__(self, **kwargs):
self.name = kwargs.get("name", None)
self.status = kwargs.get("status", None)
self.page_count = kwargs.get("page_count", None)
self.errors = kwargs.get("errors", None)
self.model_id = kwargs.get("model_id", None)
@classmethod
def _from_generated(cls, train_result):
return [
cls(
name=doc.document_name,
status=doc.status,
page_count=doc.pages,
errors=FormRecognizerError._from_generated(doc.errors),
model_id=train_result.model_id if hasattr(train_result, "model_id") else None
) for doc in train_result.training_documents
] if train_result.training_documents else None
@classmethod
def _from_generated_composed(cls, model):
training_document_info = []
for train_result in model.composed_train_results:
for doc in train_result.training_documents:
training_document_info.append(
cls(
name=doc.document_name,
status=doc.status,
page_count=doc.pages,
errors=FormRecognizerError._from_generated(doc.errors),
model_id=train_result.model_id
)
)
return training_document_info
def __repr__(self):
return "TrainingDocumentInfo(name={}, status={}, page_count={}, errors={}, model_id={})" \
.format(
self.name,
self.status,
self.page_count,
repr(self.errors),
self.model_id
)[:1024]
[docs]class FormRecognizerError(object):
"""Represents an error that occurred while training.
:ivar str code: Error code.
:ivar str message: Error message.
"""
def __init__(self, **kwargs):
self.code = kwargs.get("code", None)
self.message = kwargs.get("message", None)
@classmethod
def _from_generated(cls, err):
return [
cls(code=error.code, message=error.message) for error in err
] if err else []
def __repr__(self):
return "FormRecognizerError(code={}, message={})".format(self.code, self.message)[:1024]
[docs]class CustomFormModelInfo(object):
"""Custom model information.
:ivar str model_id: The unique identifier of the model.
:ivar str status:
The status of the model, :class:`~azure.ai.formrecognizer.CustomFormModelStatus`.
Possible values include: 'creating', 'ready', 'invalid'.
:ivar ~datetime.datetime training_started_on:
Date and time (UTC) when model training was started.
:ivar ~datetime.datetime training_completed_on:
Date and time (UTC) when model training completed.
:ivar model_name: Optional user defined model name.
:vartype model_name: str
:ivar properties: Optional model properties.
:vartype properties: ~azure.ai.formrecognizer.CustomFormModelProperties
.. versionadded:: v2.1-preview
The *model_name* and *properties* properties
"""
def __init__(self, **kwargs):
self.model_id = kwargs.get("model_id", None)
self.status = kwargs.get("status", None)
self.training_started_on = kwargs.get("training_started_on", None)
self.training_completed_on = kwargs.get("training_completed_on", None)
self.model_name = kwargs.get("model_name", None)
self.properties = kwargs.get("properties", None)
@classmethod
def _from_generated(cls, model, model_id=None, **kwargs):
if model.status == "succeeded": # map copy status to model status
model.status = "ready"
model_name = None
if hasattr(model, "attributes"):
properties = CustomFormModelProperties._from_generated(model)
elif kwargs.pop("api_version", None) == "2.0":
properties = None
else:
properties = CustomFormModelProperties(is_composed_model=False)
if hasattr(model, "model_name"):
model_name = model.model_name
return cls(
model_id=model_id if model_id else model.model_id,
status=model.status,
training_started_on=model.created_date_time,
training_completed_on=model.last_updated_date_time,
properties=properties,
model_name=model_name
)
def __repr__(self):
return "CustomFormModelInfo(model_id={}, status={}, training_started_on={}, training_completed_on={}, " \
"properties={}, model_name={})" \
.format(
self.model_id,
self.status,
self.training_started_on,
self.training_completed_on,
repr(self.properties),
self.model_name
)[:1024]
[docs]class AccountProperties(object):
"""Summary of all the custom models on the account.
:ivar int custom_model_count: Current count of trained custom models.
:ivar int custom_model_limit: Max number of models that can be trained for this account.
"""
def __init__(self, **kwargs):
self.custom_model_count = kwargs.get("custom_model_count", None)
self.custom_model_limit = kwargs.get("custom_model_limit", None)
@classmethod
def _from_generated(cls, model):
return cls(
custom_model_count=model.count,
custom_model_limit=model.limit,
)
def __repr__(self):
return "AccountProperties(custom_model_count={}, custom_model_limit={})" \
.format(
self.custom_model_count,
self.custom_model_limit
)[:1024]
[docs]class CustomFormModelProperties(object):
"""Optional model properties.
:ivar bool is_composed_model: Is this model composed? (default: false).
"""
def __init__(
self,
**kwargs
):
self.is_composed_model = kwargs.get('is_composed_model', False)
@classmethod
def _from_generated(cls, model_info):
if model_info.attributes:
return cls(
is_composed_model=model_info.attributes.is_composed
)
return cls(
is_composed_model=False
)
def __repr__(self):
return "CustomFormModelProperties(is_composed_model={})".format(self.is_composed_model)