nkululeko 0.93.10__tar.gz → 0.93.12__tar.gz

{nkululeko-0.93.10 → nkululeko-0.93.12}/CHANGELOG.md

@@ -1,8 +1,16 @@
 Changelog
 =========
 
+Version 0.93.12
+---------------
+* bugfix: map_continuous_to_cat crashed on empty data
+
+Version 0.93.11
+---------------
+* bugfix: silero segmenter assigned file duration values
+
 Version 0.93.10
---------------
+---------------
 * added nan check for imported features
 * added LOGO result output
 

{nkululeko-0.93.10/nkululeko.egg-info → nkululeko-0.93.12}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: nkululeko
-Version: 0.93.10
+Version: 0.93.12
 Summary: Machine learning audio prediction experiments based on templates
 Home-page: https://github.com/felixbur/nkululeko
 Author: Felix Burkhardt
@@ -63,7 +63,7 @@ Requires-Dist: pylatex
 ## Overview
 A project to detect speaker characteristics by machine learning experiments with a high-level interface.
 
-The idea is to have a framework (based on e.g. sklearn and torch) that can be used to rapidly and automatically analyse audio data and explore machine learning models based on that data.
+The idea is to have a framework (based on e.g. sklearn and torch) that can be used to rapidly and automatically analyse audio data and explore machine learning models based on that data. 
 
 * NEW with nkululeko: [Ensemble learning](http://blog.syntheticspeech.de/2024/06/25/nkululeko-ensemble-classifiers-with-late-fusion/)
 * NEW: [Finetune transformer-models](http://blog.syntheticspeech.de/2024/05/29/nkululeko-how-to-finetune-a-transformer-model/)
@@ -353,8 +353,16 @@ If you use it, please mention the Nkululeko paper:
 Changelog
 =========
 
+Version 0.93.12
+---------------
+* bugfix: map_continuous_to_cat crashed on empty data
+
+Version 0.93.11
+---------------
+* bugfix: silero segmenter assigned file duration values
+
 Version 0.93.10
---------------
+---------------
 * added nan check for imported features
 * added LOGO result output
 

{nkululeko-0.93.10 → nkululeko-0.93.12}/README.md

@@ -22,7 +22,7 @@
 ## Overview
 A project to detect speaker characteristics by machine learning experiments with a high-level interface.
 
-The idea is to have a framework (based on e.g. sklearn and torch) that can be used to rapidly and automatically analyse audio data and explore machine learning models based on that data.
+The idea is to have a framework (based on e.g. sklearn and torch) that can be used to rapidly and automatically analyse audio data and explore machine learning models based on that data. 
 
 * NEW with nkululeko: [Ensemble learning](http://blog.syntheticspeech.de/2024/06/25/nkululeko-ensemble-classifiers-with-late-fusion/)
 * NEW: [Finetune transformer-models](http://blog.syntheticspeech.de/2024/05/29/nkululeko-how-to-finetune-a-transformer-model/)

{nkululeko-0.93.10 → nkululeko-0.93.12}/nkululeko/constants.py

@@ -1,2 +1,2 @@
-VERSION="0.93.10"
+VERSION="0.93.12"
 SAMPLING_RATE = 16000

{nkululeko-0.93.10 → nkululeko-0.93.12}/nkululeko/data/dataset.py

@@ -34,9 +34,11 @@ class Dataset:
         self.plot = Plots()
         self.limit = int(self.util.config_val_data(self.name, "limit", 0))
         self.target_tables_append = eval(
-            self.util.config_val_data(self.name, "target_tables_append", "False")
+            self.util.config_val_data(
+                self.name, "target_tables_append", "False")
         )
-        self.start_fresh = eval(self.util.config_val("DATA", "no_reuse", "False"))
+        self.start_fresh = eval(
+            self.util.config_val("DATA", "no_reuse", "False"))
         self.is_labeled, self.got_speaker, self.got_gender, self.got_age = (
             False,
             False,
@@ -70,7 +72,8 @@ class Dataset:
         try:
             self.db = audformat.Database.load(root)
         except FileNotFoundError:
-            self.util.error(f"{self.name}: no audformat database found at {root}")
+            self.util.error(
+                f"{self.name}: no audformat database found at {root}")
         return root
 
     def _check_cols(self, df):
@@ -92,7 +95,8 @@ class Dataset:
         )
         self.util.debug(r_string)
         if glob_conf.report.initial:
-            glob_conf.report.add_item(ReportItem("Data", "Load report", r_string))
+            glob_conf.report.add_item(ReportItem(
+                "Data", "Load report", r_string))
             glob_conf.report.initial = False
 
     def load(self):
@@ -103,7 +107,8 @@ class Dataset:
         store_file = f"{store}{self.name}.{store_format}"
         self.root = self._load_db()
         if not self.start_fresh and os.path.isfile(store_file):
-            self.util.debug(f"{self.name}: reusing previously stored file {store_file}")
+            self.util.debug(
+                f"{self.name}: reusing previously stored file {store_file}")
             self.df = self.util.get_store(store_file, store_format)
             self.is_labeled = self.target in self.df
             self.got_gender = "gender" in self.df
@@ -118,10 +123,12 @@ class Dataset:
         # map the audio file paths
         self.db.map_files(lambda x: os.path.join(self.root, x))
         # the dataframes (potentially more than one) with at least the file names
-        df_files = self.util.config_val_data(self.name, "files_tables", "['files']")
+        df_files = self.util.config_val_data(
+            self.name, "files_tables", "['files']")
         df_files_tables = ast.literal_eval(df_files)
         # The label for the target column
-        self.col_label = self.util.config_val_data(self.name, "label", self.target)
+        self.col_label = self.util.config_val_data(
+            self.name, "label", self.target)
         (
             df,
             self.is_labeled,
@@ -157,7 +164,8 @@ class Dataset:
                 self.got_age = got_age2 or self.got_age
                 if audformat.is_filewise_index(df_target.index):
                     try:
-                        df_target = df_target.loc[df.index.get_level_values("file")]
+                        df_target = df_target.loc[df.index.get_level_values(
+                            "file")]
                         df_target = df_target.set_index(df.index)
                     except KeyError:
                         # just a try...
@@ -206,7 +214,8 @@ class Dataset:
             end = self.df.index.get_level_values(2)
             self.df["duration"] = (end - start).total_seconds()
         elif self.df.duration.dtype == "timedelta64[ns]":
-            self.df["duration"] = self.df["duration"].map(lambda x: x.total_seconds())
+            self.df["duration"] = self.df["duration"].map(
+                lambda x: x.total_seconds())
         # Perform some filtering if desired
         required = self.util.config_val_data(self.name, "required", False)
         if required:
@@ -236,15 +245,18 @@ class Dataset:
                 res.append(abs(n - max))
             return res
 
-        reverse = eval(self.util.config_val_data(self.name, "reverse", "False"))
+        reverse = eval(self.util.config_val_data(
+            self.name, "reverse", "False"))
         if reverse:
-            max = eval(self.util.config_val_data(self.name, "reverse.max", "False"))
+            max = eval(self.util.config_val_data(
+                self.name, "reverse.max", "False"))
             if max:
                 max = float(max)
             else:
                 max = self.df[self.target].values.max()
             self.util.debug(f"reversing target numbers with max values: {max}")
-            self.df[self.target] = reverse_array(self.df[self.target].values, max)
+            self.df[self.target] = reverse_array(
+                self.df[self.target].values, max)
 
         # check if the target variable should be scaled (z-transformed)
         scale = self.util.config_val_data(self.name, "scale", False)
@@ -317,13 +329,15 @@ class Dataset:
                 pass
             try:
                 # also it might be possible that the age is part of the speaker description
-                df_local["age"] = db[table]["speaker"].get(map="age").astype(int)
+                df_local["age"] = db[table]["speaker"].get(
+                    map="age").astype(int)
                 got_age = True
             except (ValueError, audformat.errors.BadKeyError):
                 pass
             try:
                 # same for the target, e.g. "age"
-                df_local[self.target] = db[table]["speaker"].get(map=self.target)
+                df_local[self.target] = db[table]["speaker"].get(
+                    map=self.target)
                 is_labeled = True
             except (ValueError, audformat.core.errors.BadKeyError):
                 pass
@@ -384,8 +398,10 @@ class Dataset:
             testdf = self.db.tables[self.target + ".test"].df
             traindf = self.db.tables[self.target + ".train"].df
             # use only the train and test samples that were not perhaps filtered out by an earlier processing step
-            self.df_test = self.df.loc[self.df.index.intersection(testdf.index)]
-            self.df_train = self.df.loc[self.df.index.intersection(traindf.index)]
+            self.df_test = self.df.loc[self.df.index.intersection(
+                testdf.index)]
+            self.df_train = self.df.loc[self.df.index.intersection(
+                traindf.index)]
         elif split_strategy == "train":
             self.df_train = self.df
             self.df_test = pd.DataFrame()
@@ -408,18 +424,23 @@ class Dataset:
             if entry_train_tables:
                 train_tables = ast.literal_eval(entry_train_tables)
                 for train_table in train_tables:
-                    traindf = pd.concat([traindf, self.db.tables[train_table].df])
+                    traindf = pd.concat(
+                        [traindf, self.db.tables[train_table].df])
             # use only the train and test samples that were not perhaps filtered out by an earlier processing step
             # testdf.index.map(lambda x: os.path.join(self.root, x))
             #            testdf.index = testdf.index.to_series().apply(lambda x: self.root+x)
             testdf = testdf.set_index(
-                audformat.utils.to_segmented_index(testdf.index, allow_nat=False)
+                audformat.utils.to_segmented_index(
+                    testdf.index, allow_nat=False)
             )
             traindf = traindf.set_index(
-                audformat.utils.to_segmented_index(traindf.index, allow_nat=False)
+                audformat.utils.to_segmented_index(
+                    traindf.index, allow_nat=False)
             )
-            self.df_test = self.df.loc[self.df.index.intersection(testdf.index)]
-            self.df_train = self.df.loc[self.df.index.intersection(traindf.index)]
+            self.df_test = self.df.loc[self.df.index.intersection(
+                testdf.index)]
+            self.df_train = self.df.loc[self.df.index.intersection(
+                traindf.index)]
             # it might be necessary to copy the target values
             try:
                 self.df_test[self.target] = testdf[self.target]
@@ -446,10 +467,12 @@ class Dataset:
             self.util.error(f"unknown split strategy: {split_strategy}")
 
         # check if train or test set should be ignored
-        as_test = eval(self.util.config_val_data(self.name, "as_test", "False"))
+        as_test = eval(self.util.config_val_data(
+            self.name, "as_test", "False"))
         if as_test:
             self.df_train = pd.DataFrame()
-        as_train = eval(self.util.config_val_data(self.name, "as_train", "False"))
+        as_train = eval(self.util.config_val_data(
+            self.name, "as_train", "False"))
         if as_train:
             self.df_test = pd.DataFrame()
 
@@ -480,7 +503,8 @@ class Dataset:
 
         seed = 42
         k = 30
-        test_size = int(self.util.config_val_data(self.name, "test_size", 20)) / 100.0
+        test_size = int(self.util.config_val_data(
+            self.name, "test_size", 20)) / 100.0
         df = self.df
         # split target
         targets = df[self.target].to_numpy()
@@ -496,7 +520,8 @@ class Dataset:
         stratif_vars = self.util.config_val("DATA", "balance", False)
         stratif_vars_array = {}
         if not stratif_vars:
-            self.util.error("balanced split needs stratif_vars to stratify the splits")
+            self.util.error(
+                "balanced split needs stratif_vars to stratify the splits")
         else:
             stratif_vars = ast.literal_eval(stratif_vars)
             for stratif_var in stratif_vars.keys():
@@ -505,7 +530,8 @@ class Dataset:
                     continue
                 else:
                     data = df[stratif_var].to_numpy()
-                    bins = self.util.config_val("DATA", f"{stratif_var}_bins", False)
+                    bins = self.util.config_val(
+                        "DATA", f"{stratif_var}_bins", False)
                     if bins:
                         data = binning(data, nbins=int(bins))
                     stratif_vars_array[stratif_var] = data
@@ -556,7 +582,8 @@ class Dataset:
 
     def split_speakers(self):
         """One way to split train and eval sets: Specify percentage of evaluation speakers"""
-        test_percent = int(self.util.config_val_data(self.name, "test_size", 20))
+        test_percent = int(self.util.config_val_data(
+            self.name, "test_size", 20))
         df = self.df
         s_num = df.speaker.nunique()
         test_num = int(s_num * (test_percent / 100))
@@ -575,7 +602,8 @@ class Dataset:
 
     def random_split(self):
         """One way to split train and eval sets: Specify percentage of random samples"""
-        test_percent = int(self.util.config_val_data(self.name, "test_size", 20))
+        test_percent = int(self.util.config_val_data(
+            self.name, "test_size", 20))
         df = self.df
         s_num = len(df)
         test_num = int(s_num * (test_percent / 100))
@@ -676,8 +704,11 @@ class Dataset:
 
     def map_continuous_classification(self, df):
         """Map labels to bins for continuous data that should be classified"""
+        if df.empty:
+            return
         if self.check_continuous_classification():
-            self.util.debug(f"{self.name}: binning continuous variable to categories")
+            self.util.debug(
+                f"{self.name}: binning continuous variable to categories")
             cat_vals = self.util.continuous_to_categorical(df[self.target])
             df[self.target] = cat_vals.values
             labels = ast.literal_eval(glob_conf.config["DATA"]["labels"])

{nkululeko-0.93.10 → nkululeko-0.93.12}/nkululeko/plots.py

@@ -628,8 +628,7 @@ class Plots:
         # one up because of the runs
         fig_dir = self.util.get_path("fig_dir") + "../"
         exp_name = self.util.get_exp_name(only_data=True)
-        format = self.util.config_val("PLOT", "format", "png")
-        filename = f"{fig_dir}{exp_name}EXPL_tree-plot.{format}"
+        filename = f"{fig_dir}{exp_name}EXPL_tree-plot.{self.format}"
         fig = ax.figure
         fig.savefig(filename)
         fig.clear()

{nkululeko-0.93.10 → nkululeko-0.93.12}/nkululeko/segment.py

@@ -62,6 +62,11 @@ def main():
     expr.fill_train_and_tests()
     util.debug(f"train shape : {expr.df_train.shape}, test shape:{expr.df_test.shape}")
 
+    def calc_dur(x):
+        starts = x[1]
+        ends = x[2]
+        return (ends - starts).total_seconds()
+
     # segment
     segmented_file = util.config_val("SEGMENT", "result", "segmented.csv")
 
@@ -104,16 +109,11 @@ def main():
             df_seg = df_seg.drop(columns=[target])
             df_seg = df_seg.rename(columns={"class_label": target})
         # save file
+        df_seg["duration"] = df_seg.index.to_series().map(lambda x: calc_dur(x))
         df_seg.to_csv(f"{expr.data_dir}/{segmented_file}")
 
-    def calc_dur(x):
-        starts = x[1]
-        ends = x[2]
-        return (ends - starts).total_seconds()
-
     if "duration" not in df.columns:
         df["duration"] = df.index.to_series().map(lambda x: calc_dur(x))
-    df_seg["duration"] = df_seg.index.to_series().map(lambda x: calc_dur(x))
     num_before = df.shape[0]
     num_after = df_seg.shape[0]
     util.debug(

{nkululeko-0.93.10 → nkululeko-0.93.12/nkululeko.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: nkululeko
-Version: 0.93.10
+Version: 0.93.12
 Summary: Machine learning audio prediction experiments based on templates
 Home-page: https://github.com/felixbur/nkululeko
 Author: Felix Burkhardt
@@ -63,7 +63,7 @@ Requires-Dist: pylatex
 ## Overview
 A project to detect speaker characteristics by machine learning experiments with a high-level interface.
 
-The idea is to have a framework (based on e.g. sklearn and torch) that can be used to rapidly and automatically analyse audio data and explore machine learning models based on that data.
+The idea is to have a framework (based on e.g. sklearn and torch) that can be used to rapidly and automatically analyse audio data and explore machine learning models based on that data. 
 
 * NEW with nkululeko: [Ensemble learning](http://blog.syntheticspeech.de/2024/06/25/nkululeko-ensemble-classifiers-with-late-fusion/)
 * NEW: [Finetune transformer-models](http://blog.syntheticspeech.de/2024/05/29/nkululeko-how-to-finetune-a-transformer-model/)
@@ -353,8 +353,16 @@ If you use it, please mention the Nkululeko paper:
 Changelog
 =========
 
+Version 0.93.12
+---------------
+* bugfix: map_continuous_to_cat crashed on empty data
+
+Version 0.93.11
+---------------
+* bugfix: silero segmenter assigned file duration values
+
 Version 0.93.10
---------------
+---------------
 * added nan check for imported features
 * added LOGO result output
 

{nkululeko-0.93.10 → nkululeko-0.93.12}/nkululeko.egg-info/SOURCES.txt

@@ -4,51 +4,6 @@ README.md
 pyproject.toml
 setup.cfg
 setup.py
-data/aesdd/process_database.py
-data/androids/process_database.py
-data/ased/process_database.py
-data/asvp-esd/process_database.py
-data/baved/process_database.py
-data/cafe/process_database.py
-data/clac/process_database.py
-data/cmu-mosei/process_database.py
-data/demos/process_database.py
-data/ekorpus/process_database.py
-data/emns/process_database.py
-data/emofilm/convert_to_16k.py
-data/emofilm/process_database.py
-data/emorynlp/process_database.py
-data/emov-db/process_database.py
-data/emovo/process_database.py
-data/emozionalmente/create.py
-data/enterface/process_database.py
-data/esd/process_database.py
-data/gerparas/process_database.py
-data/iemocap/process_database.py
-data/jl/process_database.py
-data/jtes/process_database.py
-data/meld/process_database.py
-data/mesd/process_database.py
-data/mess/process_database.py
-data/mlendsnd/process_database.py
-data/msp-improv/process_database2.py
-data/msp-podcast/process_database.py
-data/oreau2/process_database.py
-data/portuguese/process_database.py
-data/ravdess/process_database.py
-data/ravdess/process_database_speaker.py
-data/savee/process_database.py
-data/shemo/process_database.py
-data/subesco/process_database.py
-data/tess/process_database.py
-data/thorsten-emotional/process_database.py
-data/urdu/process_database.py
-data/vivae/process_database.py
-docs/source/conf.py
-meta/demos/demo_best_model.py
-meta/demos/my_experiment.py
-meta/demos/my_experiment_local.py
-meta/demos/plot_faster_anim.py
 nkululeko/__init__.py
 nkululeko/aug_train.py
 nkululeko/augment.py
@@ -168,5 +123,4 @@ nkululeko/segmenting/seg_silero.py
 nkululeko/utils/__init__.py
 nkululeko/utils/files.py
 nkululeko/utils/stats.py
-nkululeko/utils/util.py
-venv/bin/activate_this.py
+nkululeko/utils/util.py

nkululeko-0.93.10/data/aesdd/process_database.py

@@ -1,50 +0,0 @@
-import argparse
-from pathlib import Path
-
-import pandas as pd
-
-emotion_map = {
-    "a": "anger",
-    "d": "disgust",
-    "h": "happiness",
-    "f": "fear",
-    "s": "sadness",
-}
-
-
-def main():
-    parser = argparse.ArgumentParser()
-    parser.add_argument("--data_dir", type=str, default="AESDD", help="Path of AESDD directory")
-    parser.add_argument("--out_dir", type=str, default=".")
-    args = parser.parse_args()
-
-    data_dir = Path(args.data_dir)
-    out_dir = Path(args.out_dir)
-
-    paths = list(data_dir.glob("**/*.wav"))
-    files = [file for file in paths  if file.stem != "s05 (3)"]
-    names = [file.stem for file in files]
-    emotion = [emotion_map[file.stem[0]] for file in files]
-    speaker = [str(int(x[x.find("(") + 1: x.find(")")])) for x in names]
-    gender = [["female", "male"][int(v) % 2] for v in speaker]
-    language =["greek" for file in files]
-
-    # convert to df
-    df = pd.DataFrame({"file": files, "speaker": speaker, "emotion": emotion, "gender": gender})
-
-    # print distribution per emotion
-    # print(df.groupby("emotion").count()['file'])
-
-    # allocate speaker 5 for test set
-    train_df = df[df["speaker"] != "5"]
-    test_df = df.drop(train_df.index)
-
-    # save to CSV
-    df.to_csv(out_dir / "aesdd.csv", index=False)
-    train_df.to_csv(out_dir / "aesdd_train.csv", index=False)
-    test_df.to_csv(out_dir / "aesdd_test.csv", index=False)
-
-    print(f"Total: {len(df)}, Train: {len(train_df)}, Test: {len(test_df)}")
-    
-if __name__ == "__main__":
-    main()

nkululeko-0.93.10/data/androids/process_database.py

@@ -1,94 +0,0 @@
-"""
-This folder is to import the 
-Androids-corpus depression
-database to nkululeko.
-
-I used the version downloadable from [github](https://github.com/CheyneyComputerScience/CREMA-D)
-
-downloaded April 27th 2023
-
-I used the version downloadable from [Dropbox, mentioned in this github page](https://github.com/androidscorpus/data)
-
-Download and unzip the file Androids-corpus.zip to the current folder
-
-Usage: `python process_database.py`
-
-"""
-
-import os
-
-import audeer
-import pandas as pd
-
-dataset_name = 'androids'
-data_root = './Androids-Corpus/'
-
-# read in the fold list 
-fold_dict = {}
-df_fold = pd.read_csv(f'{data_root}fold-lists.csv')
-for i in range(df_fold.shape[0]): 
-    for j in range(df_fold.shape[1]): 
-        try:
-            value = df_fold.iloc[i, j].replace('\'', '')
-        except AttributeError:
-            value = 'na' 
-        if j in range(7, 12):
-            fold_dict[value] = j - 7
-        else:
-            fold_dict[value] = j 
-directory_list = audeer.list_file_names(data_root, filetype='wav', recursive=True, basenames=True)
-
-depressions, speakers, educations, genders, ages, tasks, folds = [], [], [], [], [], [], []
-file_paths = []
-print(len(directory_list))
-gender_map = {'F':'female', 'M':'male'}
-depression_map = {'P':'depressed', 'C':'control'}
-
-for file in directory_list:
-    # storing file paths
-    # file = file.replace(os.getcwd(), '.')
-    file_paths.append(data_root+file)
-    # storing file emotions
-    fn = audeer.basename_wo_ext(file)
-
-    # The naming convention of the audio files is as follows:
-    # nn_XGmm_t.wav
-    # where nn is a unique integer identifier such that, in a given group, files with the same nn contain the voice of the same speaker (there is a trailing 0 for numbers lower than 10), X is an alphabetic character corresponding to the speaker’s condition (P for depression patient and C for control), G is an alphabetic character that stands for the speaker’s gender (M for male and F for female), mm is a two-digits integer number corresponding to the speaker’s age, and t is an integer number between 1 and 4 accounting for the education level (1 corresponds to primary school and 4 corresponds to university). The letter X was used for the 2 participants who did not provide information about this aspect. There is no indication of the task because recordings corresponding to RT and IT are stored in different directories.
-    if 'Reading-Task'in file:
-        task = 'reading'
-    elif 'Interview-Task' in file:
-        task = 'interview'
-    else:
-        print('ERROR: task undefined')
-        exit(-1)
- 
-    part = fn.split('_')
-    dir_name = f'{part[0]}_{part[1]}_{part[2]}'
-    depression = part[1][0]
-    speaker = f'{depression}_{part[0]}'
-    gender = part[1][1]
-    age = part[1][2:4]
-    education = part[2]
-    depressions.append(depression_map[depression])
-    speakers.append(speaker)
-    genders.append(gender_map[gender])
-    ages.append(age)
-    tasks.append(task)
-    educations.append(education)
-    folds.append(fold_dict[dir_name])
-#    print(f'{file} {speaker}')
-
-
-# dataframe for emotion of files
-df = pd.DataFrame({'file':file_paths, 
-                   'speaker':speakers, 
-                   'gender':genders, 
-                   'age':ages, 
-                   'task':tasks, 
-                   'depression':depressions, 
-                   'education':educations, 
-                   'fold':folds})
-
-df = df.set_index('file')
-df.head()
-df.to_csv(f'{dataset_name}.csv')

nkululeko-0.93.10/data/ased/process_database.py

@@ -1,58 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-
-# @filename: process_database.py
-# @description: Process the ASED database 
-
-
-import argparse
-import shutil
-from importlib.resources import path
-from pathlib import Path
-
-import pandas as pd
-from joblib import delayed
-
-emotion_map = {
-    "a": "anger",
-    "h": "happiness",
-    "n": "neutral",
-    "f": "fear",
-    "s": "sadness",
-}
-
-
-def main():
-    parser = argparse.ArgumentParser(
-        usage="python3 process_database.py database output"
-    )
-    parser.add_argument("--data_dir", type=str, default="ASED_V1", help="Path to the ASED database")
-    parser.add_argument("--out_dir", type=str, default=".", help="Path to the output directory")
-    args = parser.parse_args()
-
-    data_dir = Path(args.data_dir)
-    out_dir = Path(args.out_dir)
-
-    paths = list(data_dir.glob("**/*.wav"))
-    emotions = [emotion_map[path.stem[0].lower()] for path in paths]
-    genders = [["female", "male"][int(p.stem[9:11]) - 1] for p in paths]
-    spekaers = [p.stem[-2:] for p in paths]
-    languages = ["amharic" for p in paths]
-
-    # convert to df
-    df = pd.DataFrame({"file": paths, "emotion": emotions, "gender": genders, "speaker": spekaers, "language": languages})
-
-    # allocate speakers >= 55 for test
-    df_test = df[df["speaker"] > "55"]
-    df_train = df.drop(df_test.index)
-
-
-    # save to csv
-    df_train.to_csv(out_dir / "ased_train.csv", index=False)
-    df_test.to_csv(out_dir / "ased_test.csv", index=False)
-    df.to_csv(out_dir / "ased.csv", index=False)
-
-
-if __name__ == "__main__":
-    main()
-