Merge pull request #29 from michelin/yarrow-add-metrics-optional-fiel…

…d-to-yarrow

Add metrics optional field to yarrow

.github/workflows/test-deploy.yaml

@@ -9,7 +9,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.7", "3.8", "3.9", "3.10", "3.11"]
+        python-version: ["3.8", "3.9", "3.10", "3.11"]
     steps:
       - uses: actions/checkout@v3
       - name: Set up Python ${{ matrix.python-verion }}
@@ -31,8 +31,7 @@ jobs:
         if: ${{ always() }}
       - name: lint
         run: |
-          black --check .
-          isort -c .
+          make check
 
   build:
     needs: test

Makefile

@@ -18,3 +18,21 @@ help:
 # "make mode" option.  $(O) is meant as a shortcut for $(SPHINXOPTS).
 %: Makefile
 	@$(SPHINXBUILD) -M $@ "$(SOURCEDIR)" "$(BUILDDIR)" $(SPHINXOPTS) $(O)
+
+VENV_NAME = .venv
+PIP = $(VENV_DIR)/bin/pip
+
+setup:
+	python -m venv $(VENV_NAME)
+	$(PIP) install --upgrade pip
+	$(PIP) install .[dev]
+
+check:
+	black . --check
+	isort . --check --gitignore
+	autoflake --check .
+
+lint:
+	black .
+	isort . --gitignore
+	autoflake -i --remove-all-unused-imports .

README.md

@@ -19,6 +19,12 @@ The full description can be found [here](schema/description.md) with the rules o
 ```sh
 pip install yarrowformat
 ```
+## Install it locally
+
+```sh
+make setup
+source .venv/bin/activate
+```
 
 ## How to use
 
@@ -41,7 +47,7 @@ yar_set.add_annotation(annot)
 
 # now save it somewhere else
 with open("path/to/other/file.yarrow.json", "w") as fp:
-    json.dump(yar_set.pydantic().dict(), fp, default=str)
+    json.dump(yar_set.pydantic().model_dump(exclude_none=True), fp, default=str)
 
 ```
 

examples/generate_overlays/generate_overlays_classes.py

@@ -45,4 +45,9 @@ class FakeMeta(BaseModel):
     with open(
         "examples/generate_overlays/example_overlays_classes.yarrow.json", "w"
     ) as fp:
-        json.dump(yarrowset.pydantic().dict(), fp, default=str, indent=4)
+        json.dump(
+            yarrowset.pydantic().model_dump(exclude_none=True),
+            fp,
+            default=str,
+            indent=4,
+        )

examples/generate_overlays/generate_overlays_multilayer.py

@@ -2,8 +2,6 @@
 from datetime import datetime
 from uuid import uuid4
 
-from pydantic import BaseModel
-
 import yarrow
 
 if __name__ == "__main__":
@@ -40,12 +38,15 @@
 
     yarrowset = yarrow.YarrowDataset(info=info)
 
-    # You shoud use the add_* method insert elements into the dataset
+    # You should use the add_* method insert elements into the dataset
     yarrowset.add_multilayer_image(multilayer=multilayer)
 
     with open(
         "examples/generate_overlays/example_overlays_multilayer.yarrow.json", "w"
     ) as fp:
         json.dump(
-            yarrowset.pydantic().dict(exclude_none=True), fp, default=str, indent=4
+            yarrowset.pydantic().model_dump(eexclude_none=True),
+            fp,
+            default=str,
+            indent=4,
         )

examples/generate_simple/generate_simple.py

@@ -1,4 +1,3 @@
-import json
 from datetime import datetime
 
 import yarrow
@@ -18,4 +17,5 @@
 
     yarrowset.pydantic().save_to_file(
         "examples/generate_simple/example_simple.yarrow.json",
+        exclude_none=True,
     )

examples/object_classes/README.md

@@ -142,7 +142,7 @@ All previously mentioned classes (Image, Annotation and YarrowDataset) implement
 ```python
 import json
 
-yar_set_dict = yar_set.pydantic().dict(exclude_none=True)
+yar_set_dict = yar_set.pydantic().model_dump(exclude_none=True)
 
 with open("path/to/file.yarrow.json", "w") as fp:
     json.dump(yar_set_dict, fp, default=str)

examples/object_classes/demo.py

@@ -31,5 +31,5 @@
     yar_set.add_annotation(annot)
 
     yar_set_dict = yar_set.pydantic().save_to_file(
-        "examples/object_classes/demo.yarrow.json"
+        "examples/object_classes/demo.yarrow.json", exclude_none=True
     )

schema/yarrow_schema.json

@@ -46,6 +46,14 @@
             "items": {
                 "$ref": "#/definitions/MultilayerImage_pydantic"
             }
+        },
+        "metrics": {
+            "title": "Metrics",
+            "type": "dict",
+            "items": {
+                "$ref": "#/definitions/Metrics"
+            }
+
         }
     },
     "required": [
@@ -502,6 +510,11 @@
                     "type": "string"
                 }
             }
+        },
+        "Metrics": {
+            "title": "Metrics",
+            "description": "Defining custom metrics links to image",
+            "type": "dict"
         }
     }
 }

setup.cfg

@@ -11,7 +11,7 @@ author_email = mathieu.pichon_ext@michelin.com
 package_dir =
     = src
 packages = find:
-python_requires = >= 3.7
+python_requires = >= 3.8
 install_requires =
     setuptools
     wheel
@@ -25,6 +25,7 @@ include_package_data = True
 dev =
     black == 22.8.0
     isort
+    autoflake
     pytest
     pytest-cov
     sphinx
@@ -44,3 +45,8 @@ versionfile_source = src/yarrow/_version.py
 versionfile_build = yarrow/_version.py
 tag_prefix = v
 parentdir_prefix = yarrow-
+
+[autoflake]
+ignore-init-module-imports=true
+exclude=.venv
+recursive=true

src/yarrow/yarrow.py

@@ -1,7 +1,7 @@
 import json
 import uuid
 from datetime import datetime
-from typing import Any, List, Optional, Union
+from typing import Any, Dict, List, Optional, Union
 from warnings import warn
 
 import numpy as np
@@ -70,6 +70,7 @@ def __eq__(self, other) -> bool:
         if isinstance(other, Image_pydantic):
             return all(
                 (
+                    self.azure_url == other.azure_url,
                     self.file_name == other.file_name,
                     self.width == other.width,
                     self.height == other.height,
@@ -298,6 +299,7 @@ class YarrowDataset_pydantic(BaseModel):
     contributors     : Optional[List[Contributor]] = Field(default_factory=list)
     categories       : Optional[List[Category]] = Field(default_factory=list)
     multilayer_images: Optional[List[MultilayerImage_pydantic]] = Field(default_factory=list)
+    metrics          : Optional[Dict[str, float]] = None
     # fmt: on
 
     def __eq__(self, other: Any) -> bool:

src/yarrow/yarrow_cls.py

@@ -55,7 +55,7 @@ def __init__(
         self.date_captured = date_captured
         self.azure_url = azure_url
         self.confidential = confidential
-        self.meta = meta or {}
+        self.meta = meta
         self.comment = comment
         self.asset_id = asset_id
         self.split = split
@@ -199,7 +199,7 @@ def __init__(
         self.num_keypoints = num_keypoints
         self.weight = weight
         self.date_captured = date_captured
-        self.meta = meta or {}
+        self.meta = meta
 
         self._pydantic_self = None
 
@@ -322,7 +322,7 @@ def __init__(
         self.id = id or uuid_init()
         self.images = images or []
         self.name = name or ""
-        self.meta = meta or {}
+        self.meta = meta
         self.split = split
 
         self._pydantic = None
@@ -385,6 +385,7 @@ def __init__(
         confidential: List[Clearance] = None,
         categories: List[Category] = None,
         multilayer_images: List[MultilayerImage] = None,
+        metrics: Dict[str, float] = None,
     ) -> None:
         """Entry point for a YarrowDataset, can be created empty of images or annotations.\
             Use the appropriate functions `add_annotations()` and `add_images()` to insert \
@@ -407,6 +408,7 @@ def __init__(
             confidential (List[Clearance], optional): Defaults to None.
             categories (List[Category], optional): Defaults to None.
             multilayer_images (List[MultilayerImage], optional): Defaults to None.
+            metrics (Dict[str, float], optional): Defaults to None.
         """
         self.info = info
         self.images = images or []
@@ -415,6 +417,7 @@ def __init__(
         self.confidential = confidential or []
         self.categories = categories or []
         self.multilayer_images = multilayer_images or []
+        self.metrics = metrics or {}
 
     def __eq__(self, other: "YarrowDataset"):
         if isinstance(other, YarrowDataset):
@@ -426,6 +429,7 @@ def __eq__(self, other: "YarrowDataset"):
                     set(self.confidential) == set(other.confidential),
                     set(self.categories) == set(other.categories),
                     set(self.multilayer_images) == set(other.multilayer_images),
+                    set(self.metrics) == set(other.metrics),
                 )
             )
         return NotImplemented
@@ -603,6 +607,7 @@ def pydantic(
             ]
             if len(self.multilayer_images) > 0
             else None,
+            metrics=self.metrics if self.metrics else None,
         )
 
     def set_split(self, split: str) -> None:
@@ -780,6 +785,7 @@ def from_yarrow(cls, yarrow: YarrowDataset_pydantic) -> "YarrowDataset":
             confidential=conf_list,
             categories=cat_list,
             multilayer_images=multilayer_list,
+            metrics=yarrow.metrics,
         )
 
     @classmethod

tests/test_append.py

@@ -2,6 +2,7 @@
 from copy import deepcopy
 
 import pytest
+from pydantic import ValidationError
 
 from yarrow import *
 
@@ -46,7 +47,7 @@ def test_append_empty(yar_dataset: YarrowDataset):
 def test_save_and_load_file(yar_dataset: YarrowDataset, tmp_path):
     # We save the yarrow
     yar_path = os.path.join(tmp_path, "test.yarrow.json")
-    yar_dataset.pydantic().save_to_file(yar_path)
+    yar_dataset.pydantic().save_to_file(yar_path, exclude_none=True)
 
     new_dataset = YarrowDataset.parse_file(yar_path)
     compare_yarrow_datasets_pydantic(yar_dataset, new_dataset)
@@ -55,8 +56,47 @@ def test_save_and_load_file(yar_dataset: YarrowDataset, tmp_path):
 def test_save_and_load_raw(yar_dataset: YarrowDataset, tmp_path):
     # We save the yarrow
     yar_path = os.path.join(tmp_path, "test.yarrow.json")
-    yar_dataset.pydantic().save_to_file(yar_path)
+    yar_dataset.pydantic().save_to_file(yar_path, exclude_none=True)
 
     with open(yar_path, "rb") as jsf:
         new_dataset = YarrowDataset.parse_raw(jsf)
     compare_yarrow_datasets_pydantic(yar_dataset, new_dataset)
+
+
+def test_pass_wrong_dict_to_metrics_should_raise():
+    # Given
+    excepted_error_msg = (
+        "Input should be a valid number, unable to parse string as a number"
+    )
+    excepted_number_errors = 1
+
+    # When
+    with pytest.raises(ValidationError) as excinfo:
+        YarrowDataset_pydantic(
+            images=[],
+            info=Info(source="common_flow", date_created="2021-01-01"),
+            metrics={"key": "wrong_value"},
+        )
+    actual_error_msg = excinfo.value.errors()[0].get("msg")
+    actual_number_errors = len(excinfo.value.errors())
+
+    # Then
+    assert len(excinfo.value.errors()) == excepted_number_errors == actual_number_errors
+    assert excepted_error_msg == actual_error_msg
+
+
+def test_pass_good_dict_to_metrics_should_create_YarrowDataset():
+    # Given
+    excepted_object = YarrowDataset_pydantic
+    excepted_metrics = {"iou": 0.99}
+
+    # When
+    actual_yar_dataset = YarrowDataset_pydantic(
+        images=[],
+        info=Info(source="common_flow", date_created="2021-01-01"),
+        metrics=excepted_metrics,
+    )
+
+    # Then
+    assert type(actual_yar_dataset) == excepted_object
+    assert actual_yar_dataset.metrics == excepted_metrics

tests/test_cls.py

@@ -147,7 +147,7 @@ def test_yar_cls_add_annotation(
 def test_poly_mask_validator(yar_dataset: YarrowDataset):
     annot = yar_dataset.annotations[0]
 
-    annot_param = annot.pydantic().dict()
+    annot_param = annot.pydantic().model_dump(exclude_none=True)
     polygon = [[0.1, 0.1], [0.2, 0.2], [0.3, 0.3]]
 
     annot_param["polygon"] = polygon
@@ -183,7 +183,7 @@ def test_image_id(yar_dataset: YarrowDataset):
 
 
 def test_serialize(yar_dataset: YarrowDataset):
-    str_res = yar_dataset.pydantic().json(exclude_none=True)
+    str_res = yar_dataset.pydantic().model_dump_json(exclude_none=True)
 
     dict_res = json.loads(str_res)