Merge pull request #77 from DIDONEproject/tutorials

Tutorials
DIDONEproject · Oct 11, 2024 · 0df74ca · 0df74ca
2 parents 30ea4c7 + c82a5f8
commit 0df74ca
Show file tree

Hide file tree

Showing 7 changed files with 982 additions and 1,092 deletions.
diff --git a/README.md b/README.md
@@ -42,6 +42,9 @@ https://github.com/DIDONEproject/music_symbolic_features
 
 ## Changelog
 
+#### v1.2.4
+* Fix on lyrics module. Implemeted error output file for error registration.
+
 #### v1.2.3
 * Minifix on lyrics module
 

diff --git a/docs/source/Tutorial poprock.ipynb b/docs/source/Tutorial poprock.ipynb
diff --git a/docs/source/Tutorial.ipynb b/docs/source/Tutorial.ipynb
diff --git a/docs/source/Tutorial_poprock.ipynb b/docs/source/Tutorial_poprock.ipynb
diff --git a/musif/extract/extract.py b/musif/extract/extract.py
@@ -310,6 +310,22 @@ def extract(self) -> DataFrame:
 
         return score_df
 
+    def _check_for_error_file(self):
+        # Check for error file
+        try:
+            df = pd.read_csv(f'{self._cfg.output_dir}/error_files.csv', low_memory=False)
+            df['ErrorFiles'] = df['ErrorFiles'].astype(str)
+            df['ErrorFiles'] = df['ErrorFiles'].str.rsplit('/', 1).str[-1]
+            errored_files = list(df['ErrorFiles'])
+            print(errored_files)
+            print("CSV loaded successfully.")
+        except Exception:
+            # Handle the case where the file is empty
+            print("There is no error_files.csv, it will be created and loaded error files are included manually in it.")
+            import os
+            if not os.path.exists(f'{self._cfg.output_dir}'):
+                os.makedirs(f'{self._cfg.output_dir}')
+
     def _process_corpus(
         self, filenames: List[PurePath]
     ) -> Tuple[List[dict], List[dict]]:
@@ -322,6 +338,7 @@ def process_corpus_par(idx, filename):
                 else:
                     score_features = self._process_score(idx, filename)
             except Exception as e:
+                self._check_for_error_file()
                 print(f"Error found on {filename}. Saving the filename and error print to {str(self._cfg.output_dir)}/error_files.csv for latter tracking")
                 error_files.append(filename)
                 errors.append(e)

diff --git a/musif/extract/features/lyrics/handler.py b/musif/extract/features/lyrics/handler.py
@@ -90,7 +90,7 @@ def update_score_objects(
 
             features[get_part_feature(part, VOICE_PRESENCE)] = len(
                 part_data[DATA_SOUNDING_MEASURES]
-            ) / len(part_data[DATA_MEASURES])
+            ) / len(part_data[DATA_MEASURES]) if part_data[DATA_MEASURES] else 0
 
             features[get_part_feature(part, SYLLABIC_RATIO)] = get_syllabic_ratio(
                 part_data[DATA_NOTES], part_data[DATA_LYRICS]

diff --git a/pyproject.toml b/pyproject.toml
@@ -13,7 +13,7 @@ dependencies = [
     "deepdiff>=6.2.1",
 ]
 name = "musif"
-version = "1.2.3"
+version = "1.2.4"
 description = "Music feature extraction library from the DIDONE project"
 authors = [{name = "Didone Project", email = "[email protected]"}]
 requires-python = ">=3.10"