Case insensitive

kraken/ketos/recognition.py

@@ -481,6 +481,7 @@ def test(ctx, batch_size, model, evaluation_files, device, pad, workers,
 
     cer_list = []
     wer_list = []
+    cer_case_insensitive_list=[]
 
     with threadpool_limits(limits=threads):
         for p, net in nn.items():
@@ -518,6 +519,7 @@ def test(ctx, batch_size, model, evaluation_files, device, pad, workers,
                                    collate_fn=collate_sequences)
 
             test_cer = CharErrorRate()
+            test_cer_case_insensitive = CharErrorRate()
             test_wer = WordErrorRate()
 
             with KrakenProgressBar() as progress:
@@ -537,6 +539,8 @@ def test(ctx, batch_size, model, evaluation_files, device, pad, workers,
                             algn_pred.extend(algn2)
                             error += c
                             test_cer.update(x, y)
+                            # Update case-insensitive CER metric
+                            test_cer_case_insensitive.update(x.lower(), y.lower())
                             test_wer.update(x, y)
 
                     except FileNotFoundError as e:
@@ -550,12 +554,14 @@ def test(ctx, batch_size, model, evaluation_files, device, pad, workers,
                     progress.update(pred_task, advance=1)
 
             cer_list.append(1.0 - test_cer.compute())
+            cer_case_insensitive_list.append(1.0 - test_cer_case_insensitive.compute())
             wer_list.append(1.0 - test_wer.compute())
             confusions, scripts, ins, dels, subs = compute_confusions(algn_gt, algn_pred)
             rep = render_report(p,
                                 chars,
                                 error,
                                 cer_list[-1],
+                                cer_case_insensitive_list[-1],
                                 wer_list[-1],
                                 confusions,
                                 scripts,

kraken/serialization.py

@@ -248,6 +248,7 @@ def render_report(model: str,
                   chars: int,
                   errors: int,
                   char_accuracy: float,
+                  char_CI_accucary: float, #Case insensitive
                   word_accuracy: float,
                   char_confusions: 'Counter',
                   scripts: 'Counter',
@@ -278,6 +279,7 @@ def render_report(model: str,
               'chars': chars,
               'errors': errors,
               'character_accuracy': char_accuracy * 100,
+              'character_CI_accucary': char_CI_accucary * 100,
               'word_accuracy': word_accuracy * 100,
               'insertions': sum(insertions.values()),
               'deletions': deletions,

kraken/templates/report

@@ -3,6 +3,7 @@
 {{ report.chars }}	Characters
 {{ report.errors }}	Errors
 {{ '%0.2f'| format(report.character_accuracy) }}%	Character Accuracy
+{{ '%0.2f'| format(report.character_CI_accucary) }}%	Character Accuracy (Case-insensitive)
 {{ '%0.2f'| format(report.word_accuracy) }}%	Word Accuracy
 
 {{ report.insertions }}	Insertions