nkululeko 0.88.5__py3-none-any.whl → 0.88.7__py3-none-any.whl

nkululeko/constants.py

@@ -1,2 +1,2 @@
-VERSION="0.88.5"
+VERSION="0.88.7"
 SAMPLING_RATE = 16000

nkululeko/data/dataset.py

@@ -423,9 +423,20 @@ class Dataset:
             self.util.debug(f"{self.name}: trying to reuse data splits")
             self.df_test = pd.read_pickle(storage_test)
             self.df_train = pd.read_pickle(storage_train)
+        elif isinstance(ast.literal_eval(split_strategy), list):
+            # treat this as a list of test speakers
+            self.assign_speakers(ast.literal_eval(split_strategy))
         else:
             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"))
+        if as_test:
+            self.df_train = pd.DataFrame()
+        as_train = eval(self.util.config_val_data(self.name, "as_train", "False"))
+        if as_train:
+            self.df_test = pd.DataFrame()
+
         if self.df_test.shape[0] > 0:
             self.df_test = self.finish_up(self.df_test, storage_test)
         if self.df_train.shape[0] > 0:
@@ -515,6 +526,19 @@ class Dataset:
         # because this generates new train/test sample quantaties, the feature extraction has to be done again
         glob_conf.config["FEATS"]["needs_feature_extraction"] = "True"
 
+    def assign_speakers(self, speakers):
+        """One way to split train and eval sets: Specify test speaker names."""
+        self.df_test = self.df[self.df.speaker.isin(speakers)]
+        if len(self.df_test) == 0:
+            self.util.error(f"no speakers found in {speakers}")
+        self.df_train = self.df[~self.df.index.isin(self.df_test.index)]
+        self.util.debug(
+            f"{self.name} (speakers assigned): [{self.df_train.shape[0]}/{self.df_test.shape[0]}]"
+            " samples in train/test"
+        )
+        # because this generates new train/test sample quantaties, the feature extraction has to be done again
+        glob_conf.config["FEATS"]["needs_feature_extraction"] = "True"
+
     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))

nkululeko/experiment.py

@@ -5,13 +5,13 @@ import pickle
 import random
 import time
 
-import audeer
-import audformat
 import numpy as np
 import pandas as pd
 from sklearn.preprocessing import LabelEncoder
 
-import nkululeko.glob_conf as glob_conf
+import audeer
+import audformat
+
 from nkululeko.data.dataset import Dataset
 from nkululeko.data.dataset_csv import Dataset_CSV
 from nkululeko.demo_predictor import Demo_predictor
@@ -19,6 +19,7 @@ from nkululeko.feat_extract.feats_analyser import FeatureAnalyser
 from nkululeko.feature_extractor import FeatureExtractor
 from nkululeko.file_checker import FileChecker
 from nkululeko.filter_data import DataFilter
+import nkululeko.glob_conf as glob_conf
 from nkululeko.plots import Plots
 from nkululeko.reporting.report import Report
 from nkululeko.runmanager import Runmanager
@@ -185,9 +186,7 @@ class Experiment:
                 f"reusing previously stored {storage_test} and {storage_train}"
             )
             self.df_test = self._import_csv(storage_test)
-            # print(f"df_test: {self.df_test}")
             self.df_train = self._import_csv(storage_train)
-            # print(f"df_train: {self.df_train}")
         else:
             self.df_train, self.df_test = pd.DataFrame(), pd.DataFrame()
             for d in self.datasets.values():

nkululeko/multidb.py

@@ -36,6 +36,10 @@ def main(src_dir):
     config.read(config_file)
     datasets = config["EXP"]["databases"]
     datasets = ast.literal_eval(datasets)
+    try:
+        use_splits = eval(config["EXP"]["use_splits"])
+    except KeyError:
+        use_splits = False
     dim = len(datasets)
     results = np.zeros(dim * dim).reshape([dim, dim])
     last_epochs = np.zeros(dim * dim).reshape([dim, dim])
@@ -72,15 +76,23 @@ def main(src_dir):
                     config["DATA"][
                         "databases"
                     ] = f"['{train}', '{test}', {extra_trains_1}]"
-                    config["DATA"][f"{test}.split_strategy"] = "test"
-                    config["DATA"][f"{train}.split_strategy"] = "train"
+                    if use_splits:
+                        config["DATA"][f"{test}.as_test"] = "True"
+                        config["DATA"][f"{train}.as_train"] = "True"
+                    else:
+                        config["DATA"][f"{test}.split_strategy"] = "test"
+                        config["DATA"][f"{train}.split_strategy"] = "train"
                     extra_trains_2 = ast.literal_eval(extra_trains)
                     for extra_train in extra_trains_2:
                         config["DATA"][f"{extra_train}.split_strategy"] = "train"
                 else:
                     config["DATA"]["databases"] = f"['{train}', '{test}']"
-                    config["DATA"][f"{test}.split_strategy"] = "test"
-                    config["DATA"][f"{train}.split_strategy"] = "train"
+                    if use_splits:
+                        config["DATA"][f"{test}.as_test"] = "True"
+                        config["DATA"][f"{train}.as_train"] = "True"
+                    else:
+                        config["DATA"][f"{test}.split_strategy"] = "test"
+                        config["DATA"][f"{train}.split_strategy"] = "train"
                 config["EXP"]["name"] = f"{train}_vs_{test}"
 
             tmp_config = "tmp.ini"
@@ -116,6 +128,8 @@ def plot_heatmap(results, last_epochs, labels, name, config, datasets):
     colsums = results.mean(axis=0)
     vfunc = np.vectorize(trunc_to_three)
     colsums = vfunc(colsums)
+    rowsums = results.mean(axis=1)
+    rowsums = vfunc(rowsums)
     colsums_epochs = last_epochs.mean(axis=0)
     colsums_epochs = vfunc(colsums_epochs)
     res_dir = config["EXP"]["root"]
@@ -127,7 +141,9 @@ def plot_heatmap(results, last_epochs, labels, name, config, datasets):
         data_s = ", ".join(datasets)
         text_file.write(f"{data_s}\n")
         colsums = np.array2string(colsums, separator=", ")
-        text_file.write(f"column sums\n{colsums}\n")
+        text_file.write(f"column means\n{colsums}\n")
+        rowsums = np.array2string(rowsums, separator=", ")
+        text_file.write(f"rows means\n{rowsums}\n")
         text_file.write("all results\n")
         text_file.write(repr(results))
         text_file.write("\n")

nkululeko/utils/util.py

@@ -6,13 +6,12 @@ import os.path
 import pickle
 import sys
 
-# from sysconfig import get_config_h_filename
-# from turtle import setup
-import audeer
-import audformat
 import numpy as np
 import pandas as pd
 
+import audeer
+import audformat
+
 
 class Util:
     # a list of words that need not to be warned upon if default values are
@@ -116,8 +115,8 @@ class Util:
         return dir_name
 
     def config_val_data(self, dataset, key, default):
-        """
-        Retrieve a configuration value for datasets.
+        """Retrieve a configuration value for datasets.
+
         If the value is present in the experiment configuration it will be used, else
         we look in a global file specified by the root_folders value.
         """

{nkululeko-0.88.5.dist-info → nkululeko-0.88.7.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: nkululeko
-Version: 0.88.5
+Version: 0.88.7
 Summary: Machine learning audio prediction experiments based on templates
 Home-page: https://github.com/felixbur/nkululeko
 Author: Felix Burkhardt
@@ -360,6 +360,14 @@ F. Burkhardt, Johannes Wagner, Hagen Wierstorf, Florian Eyben and Björn Schulle
 Changelog
 =========
 
+Version 0.88.7
+--------------
+* added use_splits for multidb
+
+Version 0.88.6
+--------------
+* added test speaker assign
+
 Version 0.88.5
 --------------
 * add a unique name to the uncertainty plot

{nkululeko-0.88.5.dist-info → nkululeko-0.88.7.dist-info}/RECORD

@@ -2,12 +2,12 @@ nkululeko/__init__.py,sha256=62f8HiEzJ8rG2QlTFJXUCMpvuH3fKI33DoJSj33mscc,63
 nkululeko/aug_train.py,sha256=YhuZnS_WVWnun9G-M6g5n6rbRxoVREz6Zh7k6qprFNQ,3194
 nkululeko/augment.py,sha256=4MG0apTAG5RgkuJrYEjGgDdbodZWi_HweSPNI1JJ5QA,3051
 nkululeko/cacheddataset.py,sha256=lIJ6hUo5LoxSrzXtWV8mzwO7wRtUETWnOQ4ws2XfL1E,969
-nkululeko/constants.py,sha256=ir__UePsDhzi0X5DuhvHtXtpIXtZaMIpjU97A7Ljv54,39
+nkululeko/constants.py,sha256=p-kvGUZX0J2JPXoROES9PcftVSZ1B1GfzkBt6d8MJhY,39
 nkululeko/demo.py,sha256=bLuHkeEl5rOfm7ecGHCcWATiPK7-njNbtrGljxzNzFs,5088
 nkululeko/demo_feats.py,sha256=sAeGFojhEj9WEDFtG3SzPBmyYJWLF2rkbpp65m8Ujo4,2025
 nkululeko/demo_predictor.py,sha256=zs1bjhpnKuNCPLJeiyDm19ME1NEDOQT3QNeyVKJq9Yc,4882
 nkululeko/ensemble.py,sha256=rUHg8YmD6L8Ktt2T5M6iwsWVWbpCnfiynhHdN22bLRQ,11873
-nkululeko/experiment.py,sha256=wXZnb_cfOqF8b0Zqzu2bbrEgCCpG_zPkDbD-Usw5sRs,31283
+nkululeko/experiment.py,sha256=L4PzoScPLG2xTyniVy9evcBy_8CIe3RTeTEUVTqiuvQ,31186
 nkululeko/explore.py,sha256=lDzRoW_Taa5u4BBABZLD89BcQWnYlrftJR4jgt1yyj0,2609
 nkululeko/export.py,sha256=mHeEAAmtZuxdyebLlbSzPrHSi9OMgJHbk35d3DTxRBc,4632
 nkululeko/feature_extractor.py,sha256=UnspIWz3XrNhKnBBhWZkH2bHvD-sROtrQVqB1JvkUyw,4088
@@ -15,7 +15,7 @@ nkululeko/file_checker.py,sha256=LoLnL8aHpW-axMQ46qbqrManTs5otG9ShpEZuz9iRSk,347
 nkululeko/filter_data.py,sha256=w-X2mhKdYr5DxDIz50E5yzO6Jmzk4jjDBoXsgOOVtcA,7222
 nkululeko/glob_conf.py,sha256=KL9YJQTHvTztxo1vr25qRRgaPnx4NTg0XrdbovKGMmw,525
 nkululeko/modelrunner.py,sha256=cKYD9a7MRoBxfqUy3X8kf6rGTYho-33In8I9YkzMOo8,11196
-nkululeko/multidb.py,sha256=fG3VukEWP1vreVN4gB1IRXxwwg4jLftsSEYtu0o1f78,5634
+nkululeko/multidb.py,sha256=1X2vZwDHf6HuYKCoIGDP34FECMZ2mcGNZ6-cFYZFnIQ,6332
 nkululeko/nkuluflag.py,sha256=PGWSmZz-PiiHLgcZJAoGOI_Y-sZDVI1ksB8p5r7riWM,3725
 nkululeko/nkululeko.py,sha256=Kn3s2E3yyH8cJ7z6lkMxrnqtCxTu7-qfe9Zr_ONTD5g,1968
 nkululeko/plots.py,sha256=WsI_dtPKfrYPsKymHRmIhqj33aZzTcE8fF_EwLkm_5A,22899
@@ -46,7 +46,7 @@ nkululeko/autopredict/ap_stoi.py,sha256=It0Lk-ki-gohA2AzD8nkLAN2WahYvD9rPDGTQuvd
 nkululeko/autopredict/ap_valence.py,sha256=n-hctRKySzhmJtowuMOTUu0T_ld3uK5pnfOzWeWW4VM,1024
 nkululeko/autopredict/estimate_snr.py,sha256=S-bpS0xFkwWc4Ch75UrjbS8y538lQ0U3g_iLRFXureY,5048
 nkululeko/data/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-nkululeko/data/dataset.py,sha256=hUD0NqWCfRaSHG8JNs1MsPb0zjUZAf8FJkg_c0ebq0Q,28046
+nkululeko/data/dataset.py,sha256=xaawk5QthuVStWjHWTFBtorcIe71lbPQgC6mHzSXGeI,29286
 nkululeko/data/dataset_csv.py,sha256=UGEpi__eT2KFS6Fop6N4HkMrzO-u5VP71gt44kwZavo,4588
 nkululeko/feat_extract/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nkululeko/feat_extract/feats_agender.py,sha256=sHyvxxlWXv1QGYXHGHIYEQK7X39eifSVie0tu-zBG3M,3189
@@ -106,9 +106,9 @@ nkululeko/segmenting/seg_silero.py,sha256=lLytS38KzARS17omwv8VBw-zz60RVSXGSvZ5Ev
 nkululeko/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nkululeko/utils/files.py,sha256=UiGAtZRWYjHSvlmPaTMtzyNNGE6qaLaxQkybctS7iRM,4021
 nkululeko/utils/stats.py,sha256=eC9dMO-by6CDnGLHDBQu-2B4-BudZNJ0nnWGhKYdUMA,2968
-nkululeko/utils/util.py,sha256=y-pdrjovT8yGtBTJ3ifIpTcF0fPnoz8UKbuLIZ0efpc,16768
-nkululeko-0.88.5.dist-info/LICENSE,sha256=0zGP5B_W35yAcGfHPS18Q2B8UhvLRY3dQq1MhpsJU_U,1076
-nkululeko-0.88.5.dist-info/METADATA,sha256=6-AE25u5U0-7CldZfExbSfjBu3_Ey3K7kMEkaNBysTc,39896
-nkululeko-0.88.5.dist-info/WHEEL,sha256=FZ75kcLy9M91ncbIgG8dnpCncbiKXSRGJ_PFILs6SFg,91
-nkululeko-0.88.5.dist-info/top_level.txt,sha256=DPFNNSHPjUeVKj44dVANAjuVGRCC3MusJ08lc2a8xFA,10
-nkululeko-0.88.5.dist-info/RECORD,,
+nkululeko/utils/util.py,sha256=KMxPzb0HN3XuNzAd7Kn3M3Nq91-0sDrAAEBgDKryCdo,16688
+nkululeko-0.88.7.dist-info/LICENSE,sha256=0zGP5B_W35yAcGfHPS18Q2B8UhvLRY3dQq1MhpsJU_U,1076
+nkululeko-0.88.7.dist-info/METADATA,sha256=VKwlkHohr4PJezcmZ45fVykmKmh1T6d2LCDvjR8Ierw,40017
+nkululeko-0.88.7.dist-info/WHEEL,sha256=Wyh-_nZ0DJYolHNn1_hMa4lM7uDedD_RGVwbmTjyItk,91
+nkululeko-0.88.7.dist-info/top_level.txt,sha256=DPFNNSHPjUeVKj44dVANAjuVGRCC3MusJ08lc2a8xFA,10
+nkululeko-0.88.7.dist-info/RECORD,,

{nkululeko-0.88.5.dist-info → nkululeko-0.88.7.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (71.0.1)
+Generator: setuptools (71.1.0)
 Root-Is-Purelib: true
 Tag: py3-none-any