✨ add support for V2 Classification, Crop and OCR utilities

.github/workflows/_test-code-samples.yml

@@ -40,7 +40,7 @@ jobs:
 
       - name: Tests code samples
         run: |
-            ./tests/test_code_samples.sh ${{ secrets.MINDEE_ACCOUNT_SE_TESTS }} ${{ secrets.MINDEE_ENDPOINT_SE_TESTS }} ${{ secrets.MINDEE_API_KEY_SE_TESTS }}  ${{ secrets.MINDEE_V2_SE_TESTS_API_KEY }}  ${{ secrets.MINDEE_V2_SE_TESTS_FINDOC_MODEL_ID }}  ${{ secrets.MINDEE_V2_SE_TESTS_SPLIT_MODEL_ID }}
+            ./tests/test_code_samples.sh ${{ secrets.MINDEE_ACCOUNT_SE_TESTS }} ${{ secrets.MINDEE_ENDPOINT_SE_TESTS }} ${{ secrets.MINDEE_API_KEY_SE_TESTS }}  ${{ secrets.MINDEE_V2_SE_TESTS_API_KEY }}  ${{ secrets.MINDEE_V2_SE_TESTS_FINDOC_MODEL_ID }} ${{ secrets.MINDEE_V2_SE_TESTS_CLASSIFICATION_MODEL_ID }} ${{ secrets.MINDEE_V2_SE_TESTS_CROP_MODEL_ID }} ${{ secrets.MINDEE_V2_SE_TESTS_OCR_MODEL_ID }} ${{ secrets.MINDEE_V2_SE_TESTS_SPLIT_MODEL_ID }}
 
       - name: Notify Slack Action on Failure
         uses: ravsamhq/notify-slack-action@2.3.0

.github/workflows/_test-integrations.yml

@@ -49,7 +49,10 @@ jobs:
           MINDEE_V2_API_KEY: ${{ secrets.MINDEE_V2_SE_TESTS_API_KEY }}
           MINDEE_V2_FINDOC_MODEL_ID: ${{ secrets.MINDEE_V2_SE_TESTS_FINDOC_MODEL_ID }}
           MINDEE_V2_SE_TESTS_BLANK_PDF_URL: ${{ secrets.MINDEE_V2_SE_TESTS_BLANK_PDF_URL }}
+          MINDEE_V2_SE_TESTS_CLASSIFICATION_MODEL_ID: ${{ secrets.MINDEE_V2_SE_TESTS_CLASSIFICATION_MODEL_ID }}
+          MINDEE_V2_SE_TESTS_CROP_MODEL_ID: ${{ secrets.MINDEE_V2_SE_TESTS_CROP_MODEL_ID }}
           MINDEE_V2_SE_TESTS_SPLIT_MODEL_ID: ${{ secrets.MINDEE_V2_SE_TESTS_SPLIT_MODEL_ID }}
+          MINDEE_V2_SE_TESTS_OCR_MODEL_ID: ${{ secrets.MINDEE_V2_SE_TESTS_OCR_MODEL_ID }}
       run: |
         pytest --cov mindee -m integration
 

docs/extras/code_samples/v2_classification.txt

@@ -0,0 +1,35 @@
+from mindee import (
+    ClientV2,
+    PathInput,
+    ClassificationParameters,
+    ClassificationResponse,
+)
+
+input_path = "/path/to/the/file.ext"
+api_key = "MY_API_KEY"
+model_id = "MY_CLASSIFICATION_MODEL_ID"
+
+# Init a new client
+mindee_client = ClientV2(api_key)
+
+# Set parameters
+params = ClassificationParameters(
+    # ID of the model, required.
+    model_id=model_id,
+)
+
+# Load a file from disk
+input_source = PathInput(input_path)
+
+# Send for processing using polling
+response = mindee_client.enqueue_and_get_result(
+    ClassificationResponse,
+    input_source,
+    params,
+)
+
+# Print a brief summary of the parsed data
+print(response.inference)
+
+# Access the classification result
+classification: str = response.inference.result.classification.document_type

docs/extras/code_samples/v2_crop.txt

@@ -0,0 +1,35 @@
+from mindee import (
+    ClientV2,
+    PathInput,
+    CropParameters,
+    CropResponse,
+)
+
+input_path = "/path/to/the/file.ext"
+api_key = "MY_API_KEY"
+model_id = "MY_CROP_MODEL_ID"
+
+# Init a new client
+mindee_client = ClientV2(api_key)
+
+# Set parameters
+params = CropParameters(
+    # ID of the model, required.
+    model_id=model_id,
+)
+
+# Load a file from disk
+input_source = PathInput(input_path)
+
+# Send for processing using polling
+response = mindee_client.enqueue_and_get_result(
+    CropResponse,
+    input_source,
+    params,
+)
+
+# Print a brief summary of the parsed data
+print(response.inference)
+
+# Access the crop result
+crops: list = response.inference.result.crops

docs/extras/code_samples/v2_ocr.txt

@@ -0,0 +1,35 @@
+from mindee import (
+    ClientV2,
+    PathInput,
+    OCRParameters,
+    OCRResponse,
+)
+
+input_path = "/path/to/the/file.ext"
+api_key = "MY_API_KEY"
+model_id = "MY_OCR_MODEL_ID"
+
+# Init a new client
+mindee_client = ClientV2(api_key)
+
+# Set parameters
+params = OCRParameters(
+    # ID of the model, required.
+    model_id=model_id,
+)
+
+# Load a file from disk
+input_source = PathInput(input_path)
+
+# Send for processing using polling
+response = mindee_client.enqueue_and_get_result(
+    OCRResponse,
+    input_source,
+    params,
+)
+
+# Print a brief summary of the parsed data
+print(response.inference)
+
+# Access the ocr result
+pages: list = response.inference.result.pages

mindee/__init__.py

@@ -22,6 +22,16 @@
 from mindee.parsing.common.predict_response import PredictResponse
 from mindee.parsing.common.workflow_response import WorkflowResponse
 from mindee.parsing.v2 import InferenceResponse, JobResponse
+from mindee.v2.product.classification.classification_parameters import (
+    ClassificationParameters,
+)
+from mindee.v2.product.classification.classification_response import (
+    ClassificationResponse,
+)
+from mindee.v2.product.crop.crop_parameters import CropParameters
+from mindee.v2.product.crop.crop_response import CropResponse
+from mindee.v2.product.ocr.ocr_parameters import OCRParameters
+from mindee.v2.product.ocr.ocr_response import OCRResponse
 from mindee.v2.product.split.split_parameters import SplitParameters
 from mindee.v2.product.split.split_response import SplitResponse
 
@@ -30,8 +40,12 @@
     "AsyncPredictResponse",
     "Base64Input",
     "BytesInput",
+    "ClassificationResponse",
+    "ClassificationParameters",
     "Client",
     "ClientV2",
+    "CropParameters",
+    "CropResponse",
     "DataSchema",
     "DataSchemaField",
     "DataSchemaReplace",
@@ -42,6 +56,8 @@
     "Job",
     "JobResponse",
     "LocalResponse",
+    "OCRParameters",
+    "OCRResponse",
     "PageOptions",
     "PathInput",
     "PollingOptions",

mindee/input/__init__.py

@@ -1,7 +1,6 @@
 from mindee.input.local_response import LocalResponse
 from mindee.input.base_parameters import BaseParameters
 from mindee.input.inference_parameters import InferenceParameters
-from mindee.v2.product.split.split_parameters import SplitParameters
 from mindee.input.page_options import PageOptions
 from mindee.input.polling_options import PollingOptions
 from mindee.input.sources.base_64_input import Base64Input
@@ -26,6 +25,5 @@
     "PathInput",
     "PollingOptions",
     "UrlInputSource",
-    "SplitParameters",
     "WorkflowOptions",
 ]

mindee/v2/__init__.py

@@ -1,7 +1,23 @@
+from mindee.v2.product.classification.classification_parameters import (
+    ClassificationParameters,
+)
+from mindee.v2.product.classification.classification_response import (
+    ClassificationResponse,
+)
+from mindee.v2.product.crop.crop_parameters import CropParameters
+from mindee.v2.product.crop.crop_response import CropResponse
+from mindee.v2.product.ocr.ocr_parameters import OCRParameters
+from mindee.v2.product.ocr.ocr_response import OCRResponse
 from mindee.v2.product.split.split_parameters import SplitParameters
 from mindee.v2.product.split.split_response import SplitResponse
 
 __all__ = [
+    "ClassificationResponse",
+    "ClassificationParameters",
+    "CropResponse",
+    "CropParameters",
+    "OCRResponse",
+    "OCRParameters",
     "SplitResponse",
     "SplitParameters",
 ]

mindee/v2/product/__init__.py

@@ -1,7 +1,23 @@
+from mindee.v2.product.classification.classification_parameters import (
+    ClassificationParameters,
+)
+from mindee.v2.product.classification.classification_response import (
+    ClassificationResponse,
+)
+from mindee.v2.product.crop.crop_parameters import CropParameters
+from mindee.v2.product.crop.crop_response import CropResponse
+from mindee.v2.product.ocr.ocr_parameters import OCRParameters
+from mindee.v2.product.ocr.ocr_response import OCRResponse
 from mindee.v2.product.split.split_parameters import SplitParameters
 from mindee.v2.product.split.split_response import SplitResponse
 
 __all__ = [
+    "ClassificationParameters",
+    "ClassificationResponse",
+    "CropResponse",
+    "CropParameters",
+    "OCRResponse",
+    "OCRParameters",
     "SplitResponse",
     "SplitParameters",
 ]

mindee/v2/product/classification/__init__.py

@@ -0,0 +1,21 @@
+from mindee.v2.product.classification.classification_classifier import (
+    ClassificationClassifier,
+)
+from mindee.v2.product.classification.classification_inference import (
+    ClassificationInference,
+)
+from mindee.v2.product.classification.classification_parameters import (
+    ClassificationParameters,
+)
+from mindee.v2.product.classification.classification_response import (
+    ClassificationResponse,
+)
+from mindee.v2.product.classification.classification_result import ClassificationResult
+
+__all__ = [
+    "ClassificationClassifier",
+    "ClassificationInference",
+    "ClassificationParameters",
+    "ClassificationResponse",
+    "ClassificationResult",
+]

mindee/v2/product/classification/classification_classifier.py

@@ -0,0 +1,14 @@
+from mindee.parsing.common.string_dict import StringDict
+
+
+class ClassificationClassifier:
+    """Document level classification."""
+
+    document_type: str
+    """The document type, as identified on given classification values."""
+
+    def __init__(self, server_response: StringDict):
+        self.document_type = server_response["document_type"]
+
+    def __str__(self) -> str:
+        return f":Document Type: {self.document_type}"

mindee/v2/product/classification/classification_inference.py

@@ -0,0 +1,19 @@
+from mindee.parsing.common.string_dict import StringDict
+from mindee.v2.parsing.inference.base_inference import BaseInference
+from mindee.v2.product.classification.classification_result import ClassificationResult
+
+
+class ClassificationInference(BaseInference):
+    """The inference result for a classification utility request."""
+
+    result: ClassificationResult
+    """Result of a classification inference."""
+    _slug: str = "classification"
+    """Slug of the endpoint."""
+
+    def __init__(self, raw_response: StringDict) -> None:
+        super().__init__(raw_response)
+        self.result = ClassificationResult(raw_response["result"])
+
+    def __str__(self) -> str:
+        return f"Inference\n#########\n{self.model}\n{self.file}\n{self.result}\n"

mindee/v2/product/classification/classification_parameters.py

@@ -0,0 +1,9 @@
+from mindee.input.base_parameters import BaseParameters
+
+
+class ClassificationParameters(BaseParameters):
+    """
+    Parameters accepted by the classification utility v2 endpoint.
+    """
+
+    _slug: str = "utilities/classification"

mindee/v2/product/classification/classification_response.py

@@ -0,0 +1,19 @@
+from mindee.parsing.common.string_dict import StringDict
+from mindee.v2.parsing.inference import BaseResponse
+from mindee.v2.product.classification.classification_inference import (
+    ClassificationInference,
+)
+
+
+class ClassificationResponse(BaseResponse):
+    """Represent a classification inference response from Mindee V2 API."""
+
+    inference: ClassificationInference
+    """Inference object for classification inference."""
+
+    _slug: str = "utilities/classification"
+    """Slug of the inference."""
+
+    def __init__(self, raw_response: StringDict) -> None:
+        super().__init__(raw_response)
+        self.inference = ClassificationInference(raw_response["inference"])

mindee/v2/product/classification/classification_result.py

@@ -0,0 +1,16 @@
+from mindee.parsing.common.string_dict import StringDict
+from mindee.v2.product.classification.classification_classifier import (
+    ClassificationClassifier,
+)
+
+
+class ClassificationResult:
+    """Classification result info."""
+
+    classification: ClassificationClassifier
+
+    def __init__(self, raw_response: StringDict) -> None:
+        self.classification = ClassificationClassifier(raw_response["classification"])
+
+    def __str__(self) -> str:
+        return f"Classification\n======{self.classification}"

mindee/v2/product/crop/__init__.py

@@ -0,0 +1,13 @@
+from mindee.v2.product.crop.crop_box import CropBox
+from mindee.v2.product.crop.crop_inference import CropInference
+from mindee.v2.product.crop.crop_parameters import CropParameters
+from mindee.v2.product.crop.crop_response import CropResponse
+from mindee.v2.product.crop.crop_result import CropResult
+
+__all__ = [
+    "CropBox",
+    "CropInference",
+    "CropParameters",
+    "CropResponse",
+    "CropResult",
+]

mindee/v2/product/crop/crop_box.py

@@ -0,0 +1,18 @@
+from mindee.parsing.common.string_dict import StringDict
+from mindee.parsing.v2.field.field_location import FieldLocation
+
+
+class CropBox:
+    """Crop inference result."""
+
+    location: FieldLocation
+    """Location which includes cropping coordinates for the detected object, within the source document."""
+    object_type: str
+    """Type or classification of the detected object."""
+
+    def __init__(self, server_response: StringDict):
+        self.location = FieldLocation(server_response["location"])
+        self.object_type = server_response["object_type"]
+
+    def __str__(self) -> str:
+        return f"* :Location: {self.location}\n  :Object Type: {self.object_type}"

mindee/v2/product/crop/crop_inference.py

@@ -0,0 +1,19 @@
+from mindee.parsing.common.string_dict import StringDict
+from mindee.v2.parsing.inference.base_inference import BaseInference
+from mindee.v2.product.crop.crop_result import CropResult
+
+
+class CropInference(BaseInference):
+    """Crop inference result."""
+
+    result: CropResult
+    """Result of a crop inference."""
+    _slug: str = "crop"
+    """Slug of the endpoint."""
+
+    def __init__(self, raw_response: StringDict) -> None:
+        super().__init__(raw_response)
+        self.result = CropResult(raw_response["result"])
+
+    def __str__(self) -> str:
+        return f"Inference\n#########\n{self.model}\n{self.file}\n{self.result}\n"

mindee/v2/product/crop/crop_parameters.py

@@ -0,0 +1,9 @@
+from mindee.input.base_parameters import BaseParameters
+
+
+class CropParameters(BaseParameters):
+    """
+    Parameters accepted by the crop utility v2 endpoint.
+    """
+
+    _slug: str = "utilities/crop"