test: allow for another output in test_flowchart

Author: percevalw

changelog.md

@@ -2,9 +2,10 @@
 
 ## Unreleased
 
-### Changed
+### Added
 
 - Support for pyarrow > 0.17.0
+- Support for Python 3.7 to 3.10 (3.11 or higher is not tested)
 - Support for pyspark 3 (to force pyspark 2, use `pip install eds-scikit[spark2]`)
 
 ### Fixed

eds_scikit/utils/test_utils.py

@@ -96,7 +96,7 @@ def assert_equal(
     return output
 
 
-def images_are_equal(image_1: str, image_2: str):
+def image_diff(image_1: str, image_2: str):
     img1 = Image.open(image_1)
     img2 = Image.open(image_2)
 
@@ -105,12 +105,11 @@ def images_are_equal(image_1: str, image_2: str):
     img2 = img2.resize(img1.size)
 
     sum_sq_diff = np.sum(
-        (np.asarray(img1).astype("float") - np.asarray(img2).astype("float")) ** 2
-    )
-
-    if sum_sq_diff == 0:
-        # Images are exactly the same
-        return True
-    else:
-        normalized_sum_sq_diff = sum_sq_diff / np.sqrt(sum_sq_diff)
-        return normalized_sum_sq_diff < 0.001
+        (
+            np.asarray(img1).astype("float") / 255
+            - np.asarray(img2).astype("float") / 255
+        )
+        ** 2
+    ) / np.prod(img1.size)
+
+    return sum_sq_diff

tests/flowchart/test_flowchart.py

@@ -5,7 +5,7 @@
 import pytest
 
 from eds_scikit.utils.flowchart import Flowchart
-from eds_scikit.utils.test_utils import images_are_equal
+from eds_scikit.utils.test_utils import image_diff
 
 data_as_df = pd.DataFrame(
     dict(
@@ -63,13 +63,13 @@ def test_flowchart(data, tmpdir_factory):
 
     _ = F.generate_flowchart(alternate=True, fontsize=10)
 
-    out_path = tmp_dir / "flowchart.png"
+    out_path = str(tmp_dir / "flowchart.png")
     F.save(out_path, dpi=72)
 
-    target_1 = Path(__file__).parent / "expected_flowchart.png"
-    target_2 = Path(__file__).parent / "expected_flowchart_bis.png"
+    tgt_1 = str(Path(__file__).parent / "expected_flowchart.png")
+    tgt_2 = str(Path(__file__).parent / "expected_flowchart_bis.png")
 
-    assert images_are_equal(out_path, target_1) or images_are_equal(out_path, target_2)
+    assert image_diff(out_path, tgt_1) < 0.05 or image_diff(out_path, tgt_2) < 0.05
 
 
 def test_incorrect_data():