Ocr support (#255)

* Prioritizing OCR release, this api spec doesn't match any BE version. It skips the Actions update but contains the update for OCR * running the generator * Adding test with new functionality * Automatically reformatting code --------- Co-authored-by: Auto-format Bot <[email protected]>
groundlight · Sep 25, 2024 · 3e377f5 · 3e377f5
1 parent 329d86b
commit 3e377f5
Show file tree

Hide file tree

Showing 12 changed files with 64 additions and 18 deletions.
diff --git a/generated/.openapi-generator/FILES b/generated/.openapi-generator/FILES
@@ -104,5 +104,4 @@ setup.cfg
 setup.py
 test-requirements.txt
 test/__init__.py
-test/test_detector_reset_api.py
 tox.ini
diff --git a/generated/docs/ImageQuery.md b/generated/docs/ImageQuery.md
@@ -16,6 +16,7 @@ Name | Type | Description | Notes
 **patience_time** | **float** | How long to wait for a confident response. | [readonly] 
 **confidence_threshold** | **float** | Min confidence needed to accept the response of the image query. | [readonly] 
 **rois** | [**[ROI], none_type**](ROI.md) | An array of regions of interest (bounding boxes) collected on image | [readonly] 
+**text** | **str, none_type** | A text field on image query. | [readonly] 
 **any string name** | **bool, date, datetime, dict, float, int, list, str, none_type** | any string name can be used but the value must be the correct type | [optional]
 
 [[Back to Model list]](../README.md#documentation-for-models) [[Back to API list]](../README.md#documentation-for-api-endpoints) [[Back to README]](../README.md)

diff --git a/generated/docs/LabelValue.md b/generated/docs/LabelValue.md
@@ -5,11 +5,12 @@
 Name | Type | Description | Notes
 ------------ | ------------- | ------------- | -------------
 **confidence** | **float, none_type** |  | [readonly] 
-**class_name** | **str, none_type** | A human-readable class name for this label (e.g. YES/NO) | [readonly] 
+**class_name** | **str, none_type** | Return a human-readable class name for this label (e.g. YES/NO) | [readonly] 
 **annotations_requested** | **[bool, date, datetime, dict, float, int, list, str, none_type]** |  | [readonly] 
 **created_at** | **datetime** |  | [readonly] 
 **detector_id** | **int, none_type** |  | [readonly] 
 **source** | **bool, date, datetime, dict, float, int, list, str, none_type** |  | [readonly] 
+**text** | **str, none_type** | Text annotations | [readonly] 
 **rois** | [**[ROI], none_type**](ROI.md) |  | [optional] 
 **any string name** | **bool, date, datetime, dict, float, int, list, str, none_type** | any string name can be used but the value must be the correct type | [optional]
 

diff --git a/generated/docs/ResultTypeEnum.md b/generated/docs/ResultTypeEnum.md
@@ -4,7 +4,7 @@
 ## Properties
 Name | Type | Description | Notes
 ------------ | ------------- | ------------- | -------------
-**value** | **str** |  |  must be one of ["binary_classification", "counting", ]
+**value** | **str** |  |  must be one of ["binary_classification", "counting", "multi_classification", ]
 
 [[Back to Model list]](../README.md#documentation-for-models) [[Back to API list]](../README.md#documentation-for-api-endpoints) [[Back to README]](../README.md)
 

diff --git a/generated/groundlight_openapi_client/model/image_query.py b/generated/groundlight_openapi_client/model/image_query.py
@@ -151,6 +151,10 @@ def openapi_types():
                 [ROI],
                 none_type,
             ),  # noqa: E501
+            "text": (
+                str,
+                none_type,
+            ),  # noqa: E501
         }
 
     @cached_property
@@ -169,6 +173,7 @@ def discriminator():
         "patience_time": "patience_time",  # noqa: E501
         "confidence_threshold": "confidence_threshold",  # noqa: E501
         "rois": "rois",  # noqa: E501
+        "text": "text",  # noqa: E501
     }
 
     read_only_vars = {
@@ -183,6 +188,7 @@ def discriminator():
         "patience_time",  # noqa: E501
         "confidence_threshold",  # noqa: E501
         "rois",  # noqa: E501
+        "text",  # noqa: E501
     }
 
     _composed_schemas = {}
@@ -202,6 +208,7 @@ def _from_openapi_data(
         patience_time,
         confidence_threshold,
         rois,
+        text,
         *args,
         **kwargs,
     ):  # noqa: E501
@@ -219,6 +226,7 @@ def _from_openapi_data(
             patience_time (float): How long to wait for a confident response.
             confidence_threshold (float): Min confidence needed to accept the response of the image query.
             rois ([ROI], none_type): An array of regions of interest (bounding boxes) collected on image
+            text (str, none_type): A text field on image query.
 
         Keyword Args:
             _check_type (bool): if True, values for parameters in openapi_types
@@ -290,6 +298,7 @@ def _from_openapi_data(
         self.patience_time = patience_time
         self.confidence_threshold = confidence_threshold
         self.rois = rois
+        self.text = text
         for var_name, var_value in kwargs.items():
             if (
                 var_name not in self.attribute_map

diff --git a/generated/groundlight_openapi_client/model/label_value.py b/generated/groundlight_openapi_client/model/label_value.py
@@ -125,6 +125,10 @@ def openapi_types():
                 str,
                 none_type,
             ),  # noqa: E501
+            "text": (
+                str,
+                none_type,
+            ),  # noqa: E501
             "rois": (
                 [ROI],
                 none_type,
@@ -142,6 +146,7 @@ def discriminator():
         "created_at": "created_at",  # noqa: E501
         "detector_id": "detector_id",  # noqa: E501
         "source": "source",  # noqa: E501
+        "text": "text",  # noqa: E501
         "rois": "rois",  # noqa: E501
     }
 
@@ -152,24 +157,26 @@ def discriminator():
         "created_at",  # noqa: E501
         "detector_id",  # noqa: E501
         "source",  # noqa: E501
+        "text",  # noqa: E501
     }
 
     _composed_schemas = {}
 
     @classmethod
     @convert_js_args_to_python_args
     def _from_openapi_data(
-        cls, confidence, class_name, annotations_requested, created_at, detector_id, source, *args, **kwargs
+        cls, confidence, class_name, annotations_requested, created_at, detector_id, source, text, *args, **kwargs
     ):  # noqa: E501
         """LabelValue - a model defined in OpenAPI
 
         Args:
             confidence (float, none_type):
-            class_name (str, none_type): A human-readable class name for this label (e.g. YES/NO)
+            class_name (str, none_type): Return a human-readable class name for this label (e.g. YES/NO)
             annotations_requested ([bool, date, datetime, dict, float, int, list, str, none_type]):
             created_at (datetime):
             detector_id (int, none_type):
             source (bool, date, datetime, dict, float, int, list, str, none_type):
+            text (str, none_type): Text annotations
 
         Keyword Args:
             _check_type (bool): if True, values for parameters in openapi_types
@@ -237,6 +244,7 @@ def _from_openapi_data(
         self.created_at = created_at
         self.detector_id = detector_id
         self.source = source
+        self.text = text
         for var_name, var_value in kwargs.items():
             if (
                 var_name not in self.attribute_map

diff --git a/generated/groundlight_openapi_client/model/result_type_enum.py b/generated/groundlight_openapi_client/model/result_type_enum.py
@@ -53,6 +53,7 @@ class ResultTypeEnum(ModelSimple):
         ("value",): {
             "BINARY_CLASSIFICATION": "binary_classification",
             "COUNTING": "counting",
+            "MULTI_CLASSIFICATION": "multi_classification",
         },
     }
 
@@ -102,10 +103,10 @@ def __init__(self, *args, **kwargs):
         Note that value can be passed either in args or in kwargs, but not in both.
 
         Args:
-            args[0] (str):, must be one of ["binary_classification", "counting", ]  # noqa: E501
+            args[0] (str):, must be one of ["binary_classification", "counting", "multi_classification", ]  # noqa: E501
 
         Keyword Args:
-            value (str):, must be one of ["binary_classification", "counting", ]  # noqa: E501
+            value (str):, must be one of ["binary_classification", "counting", "multi_classification", ]  # noqa: E501
             _check_type (bool): if True, values for parameters in openapi_types
                                 will be type checked and a TypeError will be
                                 raised if the wrong type is input.
@@ -194,10 +195,10 @@ def _from_openapi_data(cls, *args, **kwargs):
         Note that value can be passed either in args or in kwargs, but not in both.
 
         Args:
-            args[0] (str):, must be one of ["binary_classification", "counting", ]  # noqa: E501
+            args[0] (str):, must be one of ["binary_classification", "counting", "multi_classification", ]  # noqa: E501
 
         Keyword Args:
-            value (str):, must be one of ["binary_classification", "counting", ]  # noqa: E501
+            value (str):, must be one of ["binary_classification", "counting", "multi_classification", ]  # noqa: E501
             _check_type (bool): if True, values for parameters in openapi_types
                                 will be type checked and a TypeError will be
                                 raised if the wrong type is input.

diff --git a/generated/model.py b/generated/model.py
@@ -1,6 +1,6 @@
 # generated by datamodel-codegen:
 #   filename:  public-api.yaml
-#   timestamp: 2024-08-26T21:12:31+00:00
+#   timestamp: 2024-09-25T21:50:15+00:00
 
 from __future__ import annotations
 
@@ -110,6 +110,7 @@ class ROIRequest(BaseModel):
 class ResultTypeEnum(Enum):
     binary_classification = "binary_classification"
     counting = "counting"
+    multi_classification = "multi_classification"
 
 
 class SnoozeTimeUnitEnum(Enum):
@@ -290,16 +291,20 @@ class ImageQuery(BaseModel):
     rois: Optional[List[ROI]] = Field(
         ..., description="An array of regions of interest (bounding boxes) collected on image"
     )
+    text: Optional[str] = Field(..., description="A text field on image query.")
 
 
 class LabelValue(BaseModel):
     confidence: Optional[float] = Field(...)
-    class_name: Optional[str] = Field(..., description="A human-readable class name for this label (e.g. YES/NO)")
+    class_name: Optional[str] = Field(
+        ..., description="Return a human-readable class name for this label (e.g. YES/NO)"
+    )
     rois: Optional[List[ROI]] = None
     annotations_requested: List[AnnotationsRequestedEnum]
     created_at: datetime
     detector_id: Optional[int] = Field(...)
     source: SourceEnum
+    text: Optional[str] = Field(..., description="Text annotations")
 
 
 class LabelValueRequest(BaseModel):

diff --git a/package-lock.json b/package-lock.json
diff --git a/package.json b/package.json
@@ -2,7 +2,7 @@
   "name": "groundlight-sdk-generator",
   "dependencies": {
     "@openapitools/openapi-generator-cli": "^2.9.0",
-    "rehype-katex": "^7.0.0",
+    "rehype-katex": "^7.0.1",
     "remark-math": "^6.0.0"
   }
-}
+}
diff --git a/spec/public-api.yaml b/spec/public-api.yaml
@@ -925,6 +925,11 @@ components:
           nullable: true
           description: An array of regions of interest (bounding boxes) collected
             on image
+        text:
+          type: string
+          nullable: true
+          readOnly: true
+          description: A text field on image query.
       required:
       - confidence_threshold
       - created_at
@@ -936,6 +941,7 @@ components:
       - result
       - result_type
       - rois
+      - text
       - type
       x-internal: true
     ImageQueryTypeEnum:
@@ -953,7 +959,7 @@ components:
         class_name:
           type: string
           nullable: true
-          description: A human-readable class name for this label (e.g. YES/NO)
+          description: Return a human-readable class name for this label (e.g. YES/NO)
           readOnly: true
         rois:
           type: array
@@ -983,13 +989,19 @@ components:
           allOf:
           - $ref: '#/components/schemas/SourceEnum'
           readOnly: true
+        text:
+          type: string
+          readOnly: true
+          nullable: true
+          description: Text annotations
       required:
       - annotations_requested
       - class_name
       - confidence
       - created_at
       - detector_id
       - source
+      - text
     LabelValueRequest:
       type: object
       properties:
@@ -1145,6 +1157,7 @@ components:
       enum:
       - binary_classification
       - counting
+      - multi_classification
       type: string
     Rule:
       type: object

diff --git a/test/integration/test_groundlight.py b/test/integration/test_groundlight.py
@@ -297,6 +297,15 @@ def test_submit_image_query_returns_yes(gl: Groundlight):
     assert image_query.result.label == Label.YES
 
 
+def test_submit_image_query_returns_text(gl: Groundlight):
+    # We use the "never-review" pipeline to guarantee a confident "yes" answer.
+    detector = gl.get_or_create_detector(
+        name="Always same text", query="Is there a dog?", pipeline_config="constant-text"
+    )
+    image_query = gl.submit_image_query(detector=detector, image="test/assets/dog.jpeg", wait=10, human_review="NEVER")
+    assert isinstance(image_query.text, str)
+
+
 def test_submit_image_query_filename(gl: Groundlight, detector: Detector):
     _image_query = gl.submit_image_query(detector=detector.id, image="test/assets/dog.jpeg", human_review="NEVER")
     assert str(_image_query)