🎨 removing unnecessary underscores

Author: rbturnbull

hespi/hespi.py

@@ -286,9 +286,9 @@ def institutional_label_detect(self, component, stub, output_dir) -> Dict:
             counter = Counter(best_engine_results)
             preferred_engine = counter.most_common(1)[0][0]
         elif detection_results.get('label_classification', None) in ["printed", "typewriter"]:
-            preferred_engine = '_Tesseract'
+            preferred_engine = 'Tesseract'
         else:
-            preferred_engine = '_TrOCR'
+            preferred_engine = 'TrOCR'
 
         # Determining Recognised Text for fields not in the reference database
         for key, detection_result in detection_results.items():
@@ -362,7 +362,7 @@ def read_field_file(
 
                 detection_results[f"{field}_ocr_results"].append(
                     {
-                        'ocr': '_TrOCR', 
+                        'ocr': 'TrOCR', 
                         'original_text_detected': htr_text, 
                         'adjusted_text': adjusted_text,
                         'match_score': match_score,
@@ -386,7 +386,7 @@ def read_field_file(
 
             detection_results[f"{field}_ocr_results"].append(
                 {
-                    'ocr': '_Tesseract', 
+                    'ocr': 'Tesseract', 
                     'original_text_detected': tesseract_text, 
                     'adjusted_text': adjusted_text, 
                     'match_score': match_score,

hespi/util.py

@@ -62,11 +62,11 @@ def process_row_ocr_results(row, field_name):
     tesseract_match_score = []
 
     for d in row:
-        if d['ocr'] == '_TrOCR':
+        if d['ocr'] == 'TrOCR':
             trocr_original.append(d['original_text_detected'])
             trocr_adjusted.append(d['adjusted_text'])
             trocr_match_score.append(d['match_score'])
-        elif d['ocr'] == '_Tesseract':
+        elif d['ocr'] == 'Tesseract':
             tesseract_original.append(d['original_text_detected'])
             tesseract_adjusted.append(d['adjusted_text'])
             tesseract_match_score.append(d['match_score'])

tests/test_hespi.py

@@ -131,7 +131,7 @@ def test_read_field_file_tesseract_only():
     assert len(result["species_image"]) == 1
     assert result["species_image"][0] == image
     assert len(result['species_ocr_results']) == 1
-    assert result["species_ocr_results"][0]['ocr'] == '_Tesseract'
+    assert result["species_ocr_results"][0]['ocr'] == 'Tesseract'
     assert result["species_ocr_results"][0]['original_text_detected'] == 'zOstericolaXX'
     assert result["species_ocr_results"][0]['adjusted_text'] == 'zostericola'
     assert result["species_ocr_results"][0]['match_score'] == 0.917
@@ -150,12 +150,12 @@ def test_read_field_file_htr():
     assert result["species_image"][0] == image
     assert len(result["species_ocr_results"]) == 2
 
-    assert result["species_ocr_results"][0]['ocr'] == '_TrOCR'
+    assert result["species_ocr_results"][0]['ocr'] == 'TrOCR'
     assert result["species_ocr_results"][0]['original_text_detected'] == 'zostericolaX'
     assert result["species_ocr_results"][0]['adjusted_text'] == 'zostericolax'
     assert result["species_ocr_results"][0]['match_score'] == ""
 
-    assert result["species_ocr_results"][1]['ocr'] == '_Tesseract'
+    assert result["species_ocr_results"][1]['ocr'] == 'Tesseract'
     assert result["species_ocr_results"][1]['original_text_detected'] == 'zOstericolaXX'
     assert result["species_ocr_results"][1]['adjusted_text'] == 'zostericolaxx'
     assert result["species_ocr_results"][1]['match_score'] == ""
@@ -174,12 +174,12 @@ def test_read_field_file_fuzzy():
     assert result["species_image"][0] == image
     assert len(result["species_ocr_results"]) == 2
 
-    assert result["species_ocr_results"][0]['ocr'] == '_TrOCR'
+    assert result["species_ocr_results"][0]['ocr'] == 'TrOCR'
     assert result["species_ocr_results"][0]['original_text_detected'] == 'zostericolaX'
     assert result["species_ocr_results"][0]['adjusted_text'] == 'zostericola'
     assert result["species_ocr_results"][0]['match_score'] == 0.957
 
-    assert result["species_ocr_results"][1]['ocr'] == '_Tesseract'
+    assert result["species_ocr_results"][1]['ocr'] == 'Tesseract'
     assert result["species_ocr_results"][1]['original_text_detected'] == 'zOstericolaXX'
     assert result["species_ocr_results"][1]['adjusted_text'] == 'zostericola'
     assert result["species_ocr_results"][1]['match_score'] == 0.917
@@ -216,12 +216,12 @@ def test_institutional_label_detect(mock_yolo_output):
         assert len(result["species_image"]) == 1
         assert len(result["species_ocr_results"]) == 2
 
-        assert result["species_ocr_results"][0]['ocr'] == '_TrOCR'
+        assert result["species_ocr_results"][0]['ocr'] == 'TrOCR'
         assert result["species_ocr_results"][0]['original_text_detected'] == 'zostericolaX'
         assert result["species_ocr_results"][0]['adjusted_text'] == 'zostericola'
         assert result["species_ocr_results"][0]['match_score'] == 0.957
 
-        assert result["species_ocr_results"][1]['ocr'] == '_Tesseract'
+        assert result["species_ocr_results"][1]['ocr'] == 'Tesseract'
         assert result["species_ocr_results"][1]['original_text_detected'] == 'zOstericolaXX'
         assert result["species_ocr_results"][1]['adjusted_text'] == 'zostericola'
         assert result["species_ocr_results"][1]['match_score'] == 0.917
@@ -329,7 +329,7 @@ def test_determine_best_ocr_result_non_reference():
     assert best_match_score == ""
     assert best_engine == ""
 
-    best_text, best_match_score, best_engine = hespi.determine_best_ocr_result(result['location_ocr_results'], preferred_engine="_TrOCR")
+    best_text, best_match_score, best_engine = hespi.determine_best_ocr_result(result['location_ocr_results'], preferred_engine="TrOCR")
     assert best_text == "Queenscliff"
     assert best_match_score == ""
     assert best_engine == ""
@@ -345,12 +345,12 @@ def test_determine_best_ocr_result_reference():
     best_text, best_match_score, best_engine = hespi.determine_best_ocr_result(result['species_ocr_results'])
     assert best_text == "zostericola"
     assert best_match_score == 0.957
-    assert best_engine == "_TrOCR"
+    assert best_engine == "TrOCR"
 
-    best_text, best_match_score, best_engine = hespi.determine_best_ocr_result(result['species_ocr_results'], preferred_engine="_Tesseract")
+    best_text, best_match_score, best_engine = hespi.determine_best_ocr_result(result['species_ocr_results'], preferred_engine="Tesseract")
     assert best_text == "zostericola"
     assert best_match_score == 0.957
-    assert best_engine == "_TrOCR"
+    assert best_engine == "TrOCR"
 
 
 def test_determine_best_ocr_result_single():

tests/test_util.py

@@ -128,9 +128,9 @@ def test_ocr_data_df_ocr_results():
                 "family":"family",
                 "id":"id",
                 "species_ocr_results": [
-                    dict(ocr="_TrOCR", original_text_detected="zostericolumXX", adjusted_text="zostericolum", match_score=0.9),
-                    dict(ocr="_TrOCR", original_text_detected="z", adjusted_text="z", match_score=0),
-                    dict(ocr="_Tesseract", original_text_detected="zasdfoppasf", adjusted_text="zasdfoppasf", match_score=''),
+                    dict(ocr="TrOCR", original_text_detected="zostericolumXX", adjusted_text="zostericolum", match_score=0.9),
+                    dict(ocr="TrOCR", original_text_detected="z", adjusted_text="z", match_score=0),
+                    dict(ocr="Tesseract", original_text_detected="zasdfoppasf", adjusted_text="zasdfoppasf", match_score=''),
                 ],
                 "extra": [],
             }