Sorting audio names

+ added sorting of names at importation to improve stability

BBPs/BBPUtils.py

@@ -13,7 +13,7 @@ import pandas as pd
 from scipy.signal import butter, filtfilt
 
 #%% Functions process
-def import_csv(csv_name, folder, separator, useless=True):
+def import_csv(csv_name, folder, useless=True, slash="/"):
     """
     Parameters
     ----------
@@ -35,7 +35,7 @@ def import_csv(csv_name, folder, separator, useless=True):
     data = []
     
     # read data csv
-    with open(folder + separator + csv_name, newline="") as csv_file:
+    with open(folder + slash + csv_name, newline="") as csv_file:
         lines = csv.reader(csv_file, delimiter=',')
         for row in lines:
             data = data + [row]
@@ -72,7 +72,7 @@ def get_csv(csv_folder, slash="\\"):
     -------
     data_20_21 : DATAFRAME
         Contains the inforamtion inside the .CSVs.
-    audio_paths : LIST
+    sorted_names : LIST
         Names corresponding to audio files in the first column of the .CSVs.
     """
     csv_names = [a for a in os.listdir(csv_folder) if a.endswith('.csv')]
@@ -93,7 +93,18 @@ def get_csv(csv_folder, slash="\\"):
         if filename.endswith(".wav"):
             audio_names = np.append(audio_names, filename.replace('/', slash))
 
-    return data_frame, audio_names
+    # sort audio files names
+    # We want to sort it by year
+    years = np.unique([path[4:8] for path in audio_names])
+    sorted_names = np.array([], dtype='<U49')
+    # and inside each year, sort it by date.
+    for year in years:
+        idx_to_sort = np.where(np.array([path[4:8] for path in audio_names]) == year)[0]
+        temp_path = np.copy(audio_names[idx_to_sort])
+        temp_path.sort()
+        sorted_names = np.append(temp_path, sorted_names)
+            
+    return data_frame, sorted_names
 
 def butter_pass_filter(data, cutoff, fs, order=1, mode='high'):
     """

Clicks/ClickUtils.py

@@ -76,7 +76,7 @@ def get_csv(csv_folder, slash="\\"):
     -------
     data_20_21 : DATAFRAME
         Contains the inforamtion inside the .CSVs.
-    audio_paths : LIST
+    sorted_names : LIST
         Names corresponding to audio files in the first column of the .CSVs.
     """
     csv_names = [a for a in os.listdir(csv_folder) if a.endswith('.csv')]
@@ -97,7 +97,18 @@ def get_csv(csv_folder, slash="\\"):
         if filename.endswith(".wav"):
             audio_names = np.append(audio_names, filename.replace('/', slash))
 
-    return data_frame, audio_names
+    # sort audio files names
+    # We want to sort it by year
+    years = np.unique([path[4:8] for path in audio_names])
+    sorted_names = np.array([], dtype='<U49')
+    # and inside each year, sort it by date.
+    for year in years:
+        idx_to_sort = np.where(np.array([path[4:8] for path in audio_names]) == year)[0]
+        temp_path = np.copy(audio_names[idx_to_sort])
+        temp_path.sort()
+        sorted_names = np.append(temp_path, sorted_names)
+            
+    return data_frame, sorted_names
 
 def butter_pass_filter(data, cutoff, fs, order=1, mode='high'):
     """

Whistles/WhistleUtils.py

@@ -77,15 +77,15 @@ def get_csv(csv_folder, slash="\\"):
     -------
     data_20_21 : DATAFRAME
         Contains the inforamtion inside the .CSVs.
-    audio_paths : LIST
+    sorted_names : LIST
         Names corresponding to audio files in the first column of the .CSVs.
     """
     csv_names = [a for a in os.listdir(csv_folder) if a.endswith('.csv')]
 
     # import data
-    data = import_csv(csv_names[0], csv_folder, slash=slash)
+    data = import_csv(csv_names[0], csv_folder, separator=slash)
     for i in range(1,len(csv_names)):
-        data = data + import_csv(csv_names[i], csv_folder, slash=slash)[1:]
+        data = data + import_csv(csv_names[i], csv_folder, separator=slash)[1:]
     data_frame = pd.DataFrame(data=data[:][1:], columns=data[:][0])
 
     # change dtype for selected columns
@@ -98,7 +98,18 @@ def get_csv(csv_folder, slash="\\"):
         if filename.endswith(".wav"):
             audio_names = np.append(audio_names, filename.replace('/', slash))
 
-    return data_frame, audio_names
+    # sort audio files names
+    # We want to sort it by year
+    years = np.unique([path[4:8] for path in audio_names])
+    sorted_names = np.array([], dtype='<U49')
+    # and inside each year, sort it by date.
+    for year in years:
+        idx_to_sort = np.where(np.array([path[4:8] for path in audio_names]) == year)[0]
+        temp_path = np.copy(audio_names[idx_to_sort])
+        temp_path.sort()
+        sorted_names = np.append(temp_path, sorted_names)
+            
+    return data_frame, sorted_names
 
 #%% Trajectories algorithms
 def get_local_maxima(spectrogram, spectrogram2, hardness, threshold=10e-5):