nkululeko 0.94.2__py3-none-any.whl → 0.95.0__py3-none-any.whl

nkululeko/models/model_xgb.py

@@ -1,18 +1,164 @@
-# xgbmodel.py
+# model_xgb.py
 
+import os
 from xgboost import XGBClassifier
 
+import nkululeko.glob_conf as glob_conf
 from nkululeko.models.model import Model
 
 
 class XGB_model(Model):
-    """An XGBoost model"""
+    """An XGBoost model with early stopping support"""
 
     def __init__(self, df_train, df_test, feats_train, feats_test):
         super().__init__(df_train, df_test, feats_train, feats_test)
         self.name = "xgb"
         self.is_classifier = True
-        self.clf = XGBClassifier()  # set up the classifier
+
+        # Configure XGBoost parameters
+        xgb_params = {}
+
+        # Get early stopping configuration
+        self.early_stopping_rounds = self.util.config_val(
+            "MODEL", "early_stopping_rounds", False
+        )
+        self.eval_metric = self.util.config_val("MODEL", "eval_metric", "logloss")
+
+        # Set up other XGBoost parameters that can be configured
+        n_estimators = self.util.config_val("MODEL", "n_estimators", 100)
+        max_depth = self.util.config_val("MODEL", "max_depth", 6)
+        learning_rate = self.util.config_val("MODEL", "learning_rate", 0.3)
+        subsample = self.util.config_val("MODEL", "subsample", 1.0)
+
+        xgb_params["n_estimators"] = int(n_estimators)
+        xgb_params["max_depth"] = int(max_depth)
+        xgb_params["learning_rate"] = float(learning_rate)
+        xgb_params["subsample"] = float(subsample)
+
+        # Set random state for reproducibility
+        xgb_params["random_state"] = 42
+
+        # Add early stopping parameters to model initialization if configured
+        if self.early_stopping_rounds:
+            xgb_params["early_stopping_rounds"] = int(self.early_stopping_rounds)
+            xgb_params["eval_metric"] = self.eval_metric
+
+        # Initialize classifier with parameters
+        self.clf = XGBClassifier(**xgb_params)
+
+    def train(self):
+        """Train the XGBoost model with optional early stopping."""
+        # Check if NANs in features and handle them
+        if self.feats_train.isna().to_numpy().any():
+            self.util.debug(
+                "Model, train: replacing"
+                f" {self.feats_train.isna().sum().sum()} NANs with 0"
+            )
+            self.feats_train = self.feats_train.fillna(0)
+
+        feats = self.feats_train.to_numpy()
+        labels = self.df_train[self.target]
+
+        # Configure fitting parameters
+        fit_params = {}
+
+        # Check if early stopping is configured
+        if self.early_stopping_rounds:
+            # Check if we're in split3 mode (train/dev/test) where validation data is available
+            import ast
+
+            split3 = ast.literal_eval(
+                self.util.config_val("EXP", "traindevtest", "False")
+            )
+
+            if split3 and self.feats_test is not None and self.df_test is not None:
+                # In split3 mode, self.feats_test and self.df_test are actually the dev set
+                feats_dev = self.feats_test.to_numpy()
+                labels_dev = self.df_test[self.target]
+
+                # Handle NANs in dev features
+                if self.feats_test.isna().to_numpy().any():
+                    self.util.debug(
+                        "Model, dev: replacing"
+                        f" {self.feats_test.isna().sum().sum()} NANs with 0"
+                    )
+                    feats_dev = self.feats_test.fillna(0).to_numpy()
+
+                # Set up early stopping with validation data
+                eval_set = [(feats, labels), (feats_dev, labels_dev)]
+                fit_params["eval_set"] = eval_set
+                fit_params["verbose"] = True
+
+                self.util.debug(
+                    f"Training XGBoost with early stopping (using dev set):"
+                )
+                self.util.debug(
+                    f"  - early_stopping_rounds: {self.early_stopping_rounds}"
+                )
+                self.util.debug(f"  - eval_metric: {self.eval_metric}")
+                self.util.debug(f"  - validation set size: {feats_dev.shape[0]}")
+            else:
+                # For train/test split only: use a portion of training data for validation
+                from sklearn.model_selection import train_test_split
+
+                # Get validation split ratio (default 0.2 = 20% of training data)
+                val_split = float(
+                    self.util.config_val("MODEL", "validation_split", 0.2)
+                )
+
+                # Split training data into train and validation
+                feats_train_split, feats_val, labels_train_split, labels_val = (
+                    train_test_split(
+                        feats,
+                        labels,
+                        test_size=val_split,
+                        random_state=42,
+                        stratify=labels,
+                    )
+                )
+
+                # Set up early stopping with validation split
+                eval_set = [
+                    (feats_train_split, labels_train_split),
+                    (feats_val, labels_val),
+                ]
+                fit_params["eval_set"] = eval_set
+                fit_params["verbose"] = True
+
+                # Use the split training data for actual training
+                feats = feats_train_split
+                labels = labels_train_split
+
+                self.util.debug(
+                    f"Training XGBoost with early stopping (using validation split):"
+                )
+                self.util.debug(
+                    f"  - early_stopping_rounds: {self.early_stopping_rounds}"
+                )
+                self.util.debug(f"  - eval_metric: {self.eval_metric}")
+                self.util.debug(f"  - validation_split: {val_split}")
+                self.util.debug(f"  - training set size: {feats_train_split.shape[0]}")
+                self.util.debug(f"  - validation set size: {feats_val.shape[0]}")
+
+        # Handle class weights if configured
+        class_weight = self.util.config_val("MODEL", "class_weight", False)
+        if class_weight:
+            import sklearn.utils.class_weight
+
+            self.util.debug("using class weight")
+            classes_weights = sklearn.utils.class_weight.compute_sample_weight(
+                class_weight="balanced", y=labels
+            )
+            fit_params["sample_weight"] = classes_weights
+
+        # Train the model
+        self.clf.fit(feats, labels, **fit_params)
+
+        # Log information about the trained model
+        if hasattr(self.clf, "best_iteration"):
+            self.util.debug(f"Best iteration: {self.clf.best_iteration}")
+        if hasattr(self.clf, "best_score"):
+            self.util.debug(f"Best score: {self.clf.best_score}")
 
     def get_type(self):
         return "xgb"

nkululeko/plots.py

@@ -1,5 +1,6 @@
 # plots.py
 import ast
+import os
 
 import matplotlib.pyplot as plt
 import numpy as np
@@ -87,9 +88,10 @@ class Plots:
 
     def plot_distributions(self, df, type_s="samples"):
         class_label, df = self._check_binning("class_label", df)
-        attributes = ast.literal_eval(
-            self.util.config_val("EXPL", "value_counts", False)
-        )
+        value_counts_conf = self.util.config_val("EXPL", "value_counts", False)
+        if not isinstance(value_counts_conf, str):
+            value_counts_conf = str(value_counts_conf)
+        attributes = ast.literal_eval(value_counts_conf)
         # always plot the distribution of the main attribute
         filename = f"{class_label}_distribution"
         if self.util.is_categorical(df[class_label]):
@@ -216,11 +218,11 @@ class Plots:
 
     def save_plot(self, ax, caption, header, filename, type_s):
         # one up because of the runs
-        fig_dir = self.util.get_path("fig_dir") + "../"
+        fig_dir = os.path.dirname(self.util.get_path("fig_dir"))
         fig_plots = ax.figure
         # avoid warning
         # plt.tight_layout()
-        img_path = f"{fig_dir}{filename}_{type_s}.{self.format}"
+        img_path = os.path.join(fig_dir, f"{filename}_{type_s}.{self.format}")
         plt.savefig(img_path)
         plt.close(fig_plots)
         self.util.debug(f"Saved plot to {img_path}")
@@ -359,7 +361,7 @@ class Plots:
 
     def plot_durations(self, df, filename, sample_selection, caption=""):
         # one up because of the runs
-        fig_dir = self.util.get_path("fig_dir") + "../"
+        fig_dir = os.path.join(self.util.get_path("fig_dir"), "..")
         try:
             ax = sns.histplot(df, x="duration", hue="class_label", kde=True)
         except AttributeError as ae:
@@ -376,7 +378,7 @@ class Plots:
         ax.set_ylabel("number of samples")
         fig = ax.figure
         # plt.tight_layout()
-        img_path = f"{fig_dir}{filename}_{sample_selection}.{self.format}"
+        img_path = os.path.join(fig_dir, f"{filename}_{sample_selection}.{self.format}")
         plt.savefig(img_path)
         plt.close(fig)
         self.util.debug(f"plotted durations to {img_path}")
@@ -393,14 +395,14 @@ class Plots:
         filename = "speakers"
         caption = "speakers"
         # one up because of the runs
-        fig_dir = self.util.get_path("fig_dir") + "../"
+        fig_dir = os.path.join(self.util.get_path("fig_dir"), "..")
         sns.set_style("whitegrid")  # Set style for chart
         ax = df["speaker"].value_counts().plot(kind="pie", autopct="%1.1f%%")
         title = f"Speaker distr. for {sample_selection} {df.shape[0]}."
         ax.set_title(title)
         fig = ax.figure
         # plt.tight_layout()
-        img_path = f"{fig_dir}{filename}_{sample_selection}.{self.format}"
+        img_path = os.path.join(fig_dir, f"{filename}_{sample_selection}.{self.format}")
         plt.savefig(img_path)
         plt.close(fig)
         self.util.debug(f"plotted speakers to {img_path}")
@@ -415,7 +417,7 @@ class Plots:
 
     def describe_df(self, name, df, target, filename):
         """Make a stacked barplot of samples and speakers per sex and target values. speaker, gender and target columns must be present"""
-        fig_dir = self.util.get_path("fig_dir") + "../"  # one up because of the runs
+        fig_dir = self.util.get_path("fig_dir")  # + "../"  # one up because of the runs
         sampl_num = df.shape[0]
         sex_col = "gender"
         if target == "gender":
@@ -447,7 +449,7 @@ class Plots:
                     kind="bar", ax=axes, title=f"samples ({sampl_num})"
                 )
             # plt.tight_layout()
-            img_path = f"{fig_dir}{filename}.{self.format}"
+            img_path = os.path.join(fig_dir, f"{filename}.{self.format}")
             plt.savefig(img_path)
             fig.clear()
             plt.close(fig)
@@ -462,11 +464,12 @@ class Plots:
 
     def scatter_plot(self, feats, label_df, label, dimred_type):
         dim_num = int(self.util.config_val("EXPL", "scatter.dim", 2))
-        # one up because of the runs
-        fig_dir = self.util.get_path("fig_dir") + "../"
+        # one up because of the runs (for explore module)
+        fig_dir = os.path.join(self.util.get_path("fig_dir"), "..")
         sample_selection = self.util.config_val("EXPL", "sample_selection", "all")
-        filename = f"{label}_{self.util.get_feattype_name()}_{sample_selection}_{dimred_type}_{str(dim_num)}d"
-        filename = f"{fig_dir}{filename}.{self.format}"
+        exp_name = self.util.get_name()
+        filename = f"{label}_{exp_name}_{self.util.get_feattype_name()}_{sample_selection}_{dimred_type}_{str(dim_num)}d"
+        filename = os.path.join(fig_dir, f"{filename}.{self.format}")
         self.util.debug(f"computing {dimred_type}, this might take a while...")
         data = None
         labels = label_df[label]
@@ -573,6 +576,7 @@ class Plots:
             self.util.error(f"wrong dimension number: {dim_num}")
         fig = ax.figure
         plt.savefig(filename)
+        self.util.debug(f"plotted {dimred_type} scatter plot to {filename}")
         fig.clear()
         plt.close(fig)
         glob_conf.report.add_item(
@@ -599,8 +603,10 @@ class Plots:
         # remove fullstops in the name
         feature_name = feature.replace(".", "-")
         # one up because of the runs
-        fig_dir = self.util.get_path("fig_dir") + "../"
-        filename = f"{fig_dir}feat_dist_{title}_{feature_name}.{self.format}"
+        fig_dir = os.path.join(self.util.get_path("fig_dir"), "..")
+        filename = os.path.join(
+            fig_dir, f"feat_dist_{title}_{feature_name}.{self.format}"
+        )
         if self.util.is_categorical(df_labels[label]):
             df_plot = pd.DataFrame(
                 {label: df_labels[label], feature: df_features[feature]}
@@ -647,9 +653,9 @@ class Plots:
         # plt.tight_layout()
         # print(ax)
         # one up because of the runs
-        fig_dir = self.util.get_path("fig_dir") + "../"
+        fig_dir = os.path.join(self.util.get_path("fig_dir"), "..")
         exp_name = self.util.get_exp_name(only_data=True)
-        filename = f"{fig_dir}{exp_name}EXPL_tree-plot.{self.format}"
+        filename = os.path.join(fig_dir, f"{exp_name}EXPL_tree-plot.{self.format}")
         fig = ax.figure
         fig.savefig(filename)
         fig.clear()

nkululeko/predict.py

@@ -1,8 +1,8 @@
 # predict.py
-# use some model and add automatically predicted labels to train and test splits
-# then save as a new dataset
+# use some model and add automatically predicted labels 
+# also can labels train and test splits then save as a new dataset
 
-r"""This script is used to call the nkululeko PREDICT framework.
+r"""This script is used to call the nkululeko PREDICT module.
 
 It loads a configuration file, creates a new experiment,
 and performs automatic prediction on the train and test datasets. The predicted labels are added to the datasets and
@@ -60,9 +60,10 @@ def main():
     if "class_label" in df.columns:
         df = df.drop(columns=[target])
         df = df.rename(columns={"class_label": target})
-    name = util.get_data_name() + "_predicted"
+    sample_selection = util.config_val("PREDICT", "sample_selection", "all")
+    name = f"{sample_selection}_predicted"
     df.to_csv(f"{expr.data_dir}/{name}.csv")
-    util.debug(f"saved {name}.csv to {expr.data_dir}")
+    util.debug(f"saved {os.path.join(expr.data_dir, name)}.csv")
     print("DONE")
 
 

nkululeko/reporting/report.py

@@ -5,7 +5,6 @@ Collector class for report items collected during module processing.
 
 """
 
-from nkululeko.reporting.latex_writer import LatexWriter
 from nkululeko.utils.util import Util
 
 
@@ -31,7 +30,10 @@ class Report:
                 print("\t" + c.contents)
 
     def export_latex(self):
-        lw = LatexWriter()
-        for topic in self.report_items:
-            lw.add_items_for_section(topic, self.report_items[topic])
-        lw.finish_doc()
+        if str(self.util.config_val("REPORT", "show", "False")).lower() == "true":
+            from nkululeko.reporting.latex_writer import LatexWriter
+
+            lw = LatexWriter()
+            for topic in self.report_items:
+                lw.add_items_for_section(topic, self.report_items[topic])
+            lw.finish_doc()

nkululeko/reporting/reporter.py

@@ -152,11 +152,14 @@ class Reporter:
             probas["truth"] = self.truths
             try:
                 le = glob_conf.label_encoder
-                mapping = dict(zip(le.classes_, range(len(le.classes_))))
-                mapping_reverse = {value: key for key, value in mapping.items()}
-                probas = probas.rename(columns=mapping_reverse)
-                probas["predicted"] = probas["predicted"].map(mapping_reverse)
-                probas["truth"] = probas["truth"].map(mapping_reverse)
+                if le is not None:
+                    mapping = dict(zip(le.classes_, range(len(le.classes_))))
+                    mapping_reverse = {value: key for key, value in mapping.items()}
+                    probas = probas.rename(columns=mapping_reverse)
+                    probas["predicted"] = probas["predicted"].map(mapping_reverse)
+                    probas["truth"] = probas["truth"].map(mapping_reverse)
+                else:
+                    self.util.debug("Label encoder is None, skipping label mapping")
             except AttributeError as ae:
                 self.util.debug(f"Can't label categories: {ae}")
             # compute entropy per sample

nkululeko/runmanager.py

@@ -181,7 +181,7 @@ class Runmanager:
         """
         # self.load_model(report)
         # report = self.model.predict()
-        self.util.debug(f"plotting conf matrix to {plot_name}")
+        self.util.debug(f"plotting conf matrix as {plot_name}")
         report.plot_confmatrix(plot_name, epoch=report.epoch)
         report.print_results(report.epoch, file_name=plot_name)
         report.print_probabilities(file_name=plot_name)

nkululeko/utils/util.py

@@ -1,16 +1,18 @@
 # util.py
 import ast
 import configparser
+import json
 import logging
 import os.path
 import pickle
 import sys
 
-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
@@ -92,6 +94,8 @@ class Util:
                 dir_name = "./results/"
             elif entry == "model_dir":
                 dir_name = "./models/"
+            elif entry == "cache":
+                dir_name = "./cache/"
             else:
                 dir_name = "./store/"
         else:
@@ -107,6 +111,8 @@ class Util:
                     entryn = "./results/"
                 elif entry == "model_dir":
                     entryn = "./models/"
+                elif entry == "cache":
+                    entryn = "./cache/"
                 else:
                     entryn = "./store/"
 
@@ -328,6 +334,7 @@ class Util:
             self.logger.warning(f"WARNING: {self.caller}: {message}")
         else:
             print(f"WARNING: {message}")
+
     def debug(self, message):
         if self.logger is not None:
             self.logger.debug(f"DEBUG: {self.caller}: {message}")
@@ -505,3 +512,28 @@ class Util:
     def to_3_digits_str(self, x):
         """Given a float, return this to 3 digits as string without integer number."""
         return str(self.to_3_digits(x))[1:]
+
+    def save_json(self, file: str, var: dict):
+        """Save variable to json file.
+
+        Args:
+            file: path to json file
+            var: dictionary to store
+
+        """
+        with open(file, "w", encoding="utf-8") as fp:
+            json.dump(var, fp, ensure_ascii=False, indent=2)
+
+    def read_json(self, file: str) -> object:
+        """Read variable from json file.
+
+        Args:
+            file: path to json file
+
+        Returns:
+            content of json file
+
+        """
+        with open(file, "r") as fp:
+            return json.load(fp)
+

{nkululeko-0.94.2.dist-info → nkululeko-0.95.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: nkululeko
-Version: 0.94.2
+Version: 0.95.0
 Summary: Machine learning audio prediction experiments based on templates
 Home-page: https://github.com/felixbur/nkululeko
 Author: Felix Burkhardt

{nkululeko-0.94.2.dist-info → nkululeko-0.95.0.dist-info}/RECORD

@@ -3,28 +3,28 @@ nkululeko/__init__.py,sha256=62f8HiEzJ8rG2QlTFJXUCMpvuH3fKI33DoJSj33mscc,63
 nkululeko/aug_train.py,sha256=wpiHCJ7zsW38kumg3ypwXZe2HQrhUblAnv7P2QeJnAc,3525
 nkululeko/augment.py,sha256=3RzaxB3gRxovgJVjHXi0glprW01J7RaHhUkqotW2T3U,2955
 nkululeko/cacheddataset.py,sha256=XFpWZmbJRg0pvhnIgYf0TkclxllD-Fctu-Ol0PF_00c,969
-nkululeko/constants.py,sha256=1uXTovFvHp9PGrpkAyOYxPPZcMM5ojIHW2L7u92ADUo,39
+nkululeko/constants.py,sha256=kSkvbiOI3g6sa27nNrD9sHxZMswo_qr0qKYSc_hZQLY,39
 nkululeko/demo-ft.py,sha256=iD9Pzp9QjyAv31q1cDZ75vPez7Ve8A4Cfukv5yfZdrQ,770
 nkululeko/demo.py,sha256=tu7Al2l5MCLVegkDC-NE2wcuc_YE7NRbgOlPW3yhGEs,4940
 nkululeko/demo_feats.py,sha256=BvZjeNFTlERIRlq34OHM4Z96jdDQAhB01BGQAUcX9dM,2026
 nkululeko/demo_predictor.py,sha256=lDF-xOxRdEAclOmbepAYg-BQXQdGkHfq2n74PTIoop8,4872
 nkululeko/ensemble.py,sha256=71V-rre61H3J4sh7lu-OTo4I2_g7mm_rQxwW1ARDHgY,12782
-nkululeko/experiment.py,sha256=xZQ3SpFhH4QByRzVBCO4Ps84KDXKuVPZ_qUzLUPgN5g,36221
-nkululeko/explore.py,sha256=FPM2CS-LKgcDV-LnjYlD6pEv7HuCQpH_C3KyyiOCdk4,3589
+nkululeko/experiment.py,sha256=hdFvRA7EoQz10nId9MwcbYOTz2ifYeGrFKVJOv9a88Q,38394
+nkululeko/explore.py,sha256=aDVHwuo-lkih7VZrbb_zFKg5fowSrAIcx0V9wf0SRGo,4175
 nkululeko/export.py,sha256=U-V4acxtuL6qKt6oAsVcM5TTeWogYUJ3GU-lA6rq6d4,4336
 nkululeko/feature_extractor.py,sha256=X6ZWDjGwUMVwnP6TkCEnw8B4xo8eWvJa1QT9-0WUuvA,4102
 nkululeko/file_checker.py,sha256=xJY0Q6w47pnmgJVK5rcAKPYBrCpV7eBT4_3YBzTx-H8,3454
 nkululeko/filter_data.py,sha256=4sGrKvMZ_hLnJPrHm_CqjDPKIRV8REWoT7nfSYGXbwo,7305
 nkululeko/fixedsegment.py,sha256=Tb92QiuiyMsOO3WRWwuGjZGibS8hbHHCrcWAXGk7g04,2868
-nkululeko/glob_conf.py,sha256=KL9YJQTHvTztxo1vr25qRRgaPnx4NTg0XrdbovKGMmw,525
-nkululeko/modelrunner.py,sha256=NpDgXfKkn8dOrQzhUiEfGI56Qrb1sOtWTD31II4Zgbk,11550
+nkululeko/glob_conf.py,sha256=NLFh-1_I0Wdfo2EnSq1Oppx23AX6jAUpgFbk2zqZJ24,659
+nkululeko/modelrunner.py,sha256=mnM5JBTiTVH4BOGZBepOuwEzAgrZDmoejhpVEvzq35s,11970
 nkululeko/multidb.py,sha256=sO6OwJn8sn1-C-ig3thsIL8QMWHdV9SnJhDodKjeKrI,6876
 nkululeko/nkuluflag.py,sha256=PGWSmZz-PiiHLgcZJAoGOI_Y-sZDVI1ksB8p5r7riWM,3725
 nkululeko/nkululeko.py,sha256=6ALPMMIz6l0O3IRaP0q4b59ZUxpfzNqLQUqZMf5t3Zo,1976
-nkululeko/plots.py,sha256=i9VIkviBWLgncfnyK44TUMzg2Xa0_UhfL0LnMF1vHTw,27022
-nkululeko/predict.py,sha256=MLnHEyFmSiHLLs-HDczag8Vu3zKF5T1rXLKdZZJ6py8,2083
+nkululeko/plots.py,sha256=lUxgyoriYTwdpHZvBBQ4e41v77deQrt0PcRDLJWijys,27503
+nkululeko/predict.py,sha256=VSF42CqJvyiwAnvoThV0oNNJLHhTRMS0cI6lICQeptc,2177
 nkululeko/resample.py,sha256=rn3-M1A-iwVGibfQNGyeYNa7briD24lIN9Szq_1uTJo,5194
-nkululeko/runmanager.py,sha256=-QI7pGLVnNAPMAIobcEip9zQNQdO2u0sp0Yd4XH4mBE,8902
+nkululeko/runmanager.py,sha256=YtGQP0UyyQTKkilncB1XYM-T8oatzGcZEOcj5SorjJw,8902
 nkululeko/scaler.py,sha256=a4lKwWT436TV4VEvqtP1uQ58Yz67XVHr1HjO5gp3xLI,5109
 nkululeko/segment.py,sha256=7UrJEwdLmh9wDL5iBwpdJyJm9dwSxidHrHt-_D2qtxw,4949
 nkululeko/syllable_nuclei.py,sha256=5w_naKxNxz66a_qLkraemi2fggM-gWesiiBPS47iFcE,9931
@@ -35,11 +35,12 @@ nkululeko/augmenting/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hS
 nkululeko/augmenting/augmenter.py,sha256=TUUznEz0pe9DSMC9r7LoBckuvsJTprvypeV5-8zLn20,2846
 nkululeko/augmenting/randomsplicer.py,sha256=TQTy4RBt6XbWiuUu5Ic913DMvmwTUwEufldBJjo7i1s,2801
 nkululeko/augmenting/randomsplicing.py,sha256=GXCpCDdOsOyWACDJ3ujmFZBVe6ISvkoQLefBNPgxxow,1750
-nkululeko/augmenting/resampler.py,sha256=gcjyyTD6QtJK6s_xoOQpsu5adpn0uSJwHxJTHMskfOM,3541
+nkululeko/augmenting/resampler.py,sha256=c5AjohxomX8ujOoJRnLZoNe1fxY8Fdw6LAdFm9KDy78,4020
 nkululeko/autopredict/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nkululeko/autopredict/ap_age.py,sha256=yzd8sF6gi0hnqNawyLBCIkt-pKgl9gYPlZHsrLGfz0U,1098
 nkululeko/autopredict/ap_arousal.py,sha256=lpv3jTSVEVCcR226JevNM6S7e0_uMZXHb_8Wpup1yj8,1027
 nkululeko/autopredict/ap_dominance.py,sha256=Ltq5x0ralxU1758_e-nNKvzexiPUM66xLAm3Wo2B07c,1040
+nkululeko/autopredict/ap_emotion.py,sha256=1efW3cQjwc804Pf2aYU-XfjYtYXtZdyeiXtWL439x6o,1030
 nkululeko/autopredict/ap_gender.py,sha256=RjLv9YxY9OPHT_gnd6htjKQzQA4DSKcbjipKGjHHx2A,1011
 nkululeko/autopredict/ap_mos.py,sha256=PMLU67JDgYQMobRSR2vW9cWoL3QK5JbhLM65fVsRGkc,1108
 nkululeko/autopredict/ap_pesq.py,sha256=EuJ9u6oaSPWdYsaU8q3t8tiFKhfW1qdqgO-cySpfxw0,1141
@@ -47,20 +48,22 @@ nkululeko/autopredict/ap_sdr.py,sha256=xYCy4M_aWzoFiYD_KOK9ys2P2v0bfxNkLcIRdi5z2
 nkululeko/autopredict/ap_sid.py,sha256=b_JwVWlqcwdC7acU9Q7mExuOJKUn6qdlmQTm8pmmptk,2642
 nkululeko/autopredict/ap_snr.py,sha256=cjc0pUsCN_RacTw1UBR2cGY9t_um48f2cjo3QJDn7bw,1111
 nkululeko/autopredict/ap_stoi.py,sha256=csv9qCcRmieHAhypszqGoGt9r3biM8IYPgcTwp9GIFM,1188
+nkululeko/autopredict/ap_text.py,sha256=zaz9qIg90-ghZhBe1ka0HoUnap6s6RyopUKoCpttHOU,1333
 nkululeko/autopredict/ap_valence.py,sha256=9S06SpO_zXKSpkf0InHYYXZcD9HDGoCJ6UPkn__eBAg,1027
 nkululeko/autopredict/estimate_snr.py,sha256=1k9-XadABudnsNOeFZD_Fg0E64-GUQVS7JEp82MLQS4,4995
+nkululeko/autopredict/whisper_transcriber.py,sha256=DWDvpRaV5KmUF18ojPEvxnVXm_h_nWyY-TfW2Ngd5N8,2941
 nkululeko/data/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nkululeko/data/dataset.py,sha256=JLbBYGniUrjwxs-HtbIyhqO3Cv-ELfpmlq7jzij4dBc,41759
 nkululeko/data/dataset_csv.py,sha256=AIbtB6pGk5BSQGIgfokZ7tEGFjmuOq5w2XumRSimVWs,4833
 nkululeko/feat_extract/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nkululeko/feat_extract/feats_agender.py,sha256=onfAQ6-xx_mFMJXEF1IX8cHBmGtGeX6weJmxbkfh1_o,3184
 nkululeko/feat_extract/feats_agender_agender.py,sha256=_YQv1qw--3uQfnyTQDCwlmPRnrhdMhgXbYK2yQtseW0,3464
-nkululeko/feat_extract/feats_analyser.py,sha256=txuIEgO4uprle35RzBczvZm5Hc7iUl2p9oBEfdrvg_I,13506
+nkululeko/feat_extract/feats_analyser.py,sha256=lodim7qQ8M7c3iMeJ5bHQ-nCy9Cehx1Xl5K3leii6-w,14768
 nkululeko/feat_extract/feats_ast.py,sha256=w62xEoLiFtU-rj6SXkqXAktmoFaXcAcAWpUyEjp8JWo,4652
 nkululeko/feat_extract/feats_auddim.py,sha256=CGLp_aYhudfwoU5522vjrvjPxfZcyw593A8xLjYefV8,3134
 nkululeko/feat_extract/feats_audmodel.py,sha256=OsZyB1rdcG0Fai2gAwBlbuubmWor1_-P4IDkZLqgPKE,3161
 nkululeko/feat_extract/feats_clap.py,sha256=1tttpfm2SJmQgYm2u8eUVpDiDOpWdKqFChpY3ZZokNs,3395
-nkululeko/feat_extract/feats_emotion2vec.py,sha256=ObVlqbsJsw-hWGsUOXY68Ebynt5Bn4Xtlu_Gvq3XJI4,8728
+nkululeko/feat_extract/feats_emotion2vec.py,sha256=LnV8xEg7L7HIDqz0ulqUNoaAHBU0d5gyQPb2_32T_18,9694
 nkululeko/feat_extract/feats_hubert.py,sha256=F3vrPCkx8EimJjFWYCZ7Yg9uo1G3NjYt4UKrGIUev8k,5172
 nkululeko/feat_extract/feats_import.py,sha256=cPi4XRuRs71npB8YGXr7rYOvkeTU_oZEl3GrGncdiqY,2222
 nkululeko/feat_extract/feats_mld.py,sha256=5aRoYiGDm5ApoFntxAMQYPjEelXHHRBHZcAJR9dxaeI,1945
@@ -78,7 +81,7 @@ nkululeko/feat_extract/feats_wav2vec2.py,sha256=q1QzMD3KbhF2SOmxdwI7CiViRmhlFRyg
 nkululeko/feat_extract/feats_wavlm.py,sha256=O9cfc39VF5aPJRRATKb37pHT4W11i2cu5O1mY9LOjIA,4755
 nkululeko/feat_extract/feats_whisper.py,sha256=n3ESZtva7wshs8E8diBlQYa9xCH_P0UY1DncSrxz-FY,4508
 nkululeko/feat_extract/featureset.py,sha256=clcBv9rzBRW-bfw7JC_FYTjU5uUS-c0UE1XtQLYYRiE,1615
-nkululeko/feat_extract/feinberg_praat.py,sha256=bgzWtQkKbgcygrzwAxDXosui1rcc38qhWuJq9GLr0z8,21308
+nkululeko/feat_extract/feinberg_praat.py,sha256=mMin5V-Kmx24oYJT_miNFN4t-tEVEF3Cd0969xiVV0E,28573
 nkululeko/feat_extract/transformer_feature_extractor.py,sha256=LaXuW-AJZ931ttLis0J5h9N3RtiiE51BnkxJR-bubfY,5837
 nkululeko/losses/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nkululeko/losses/loss_ccc.py,sha256=NOK0y0fxKUnU161B5geap6Fmn8QzoPl2MqtPiV8IuJE,976
@@ -97,17 +100,17 @@ nkululeko/models/model_svm.py,sha256=zP8ykLhCZTYvwSqw06XHuzq9qMBtsiYpxjUpWDAnMyA
 nkululeko/models/model_svr.py,sha256=FEwYRdgqwgGhZdkpRnT7Ef12lklWi6GZL28PyV99xWs,726
 nkululeko/models/model_tree.py,sha256=6L3PD3aIiiQz1RPWS6z3Edx4f0gnR7AOfBKOJzf0BNU,433
 nkululeko/models/model_tree_reg.py,sha256=IMaQpNImoRqP8Biw1CsJevxpV_PVpKblsKtYlMW5d_U,429
-nkululeko/models/model_tuned.py,sha256=VuRyNqw3XTpQ2eHsWOJN8X-V98AN8Wqiq7UgwT5BQRU,23763
-nkululeko/models/model_xgb.py,sha256=ytBaSHZH8r7VvRYdmrBrQnzRM6V4HyCJ8O-v20J8G_g,448
+nkululeko/models/model_tuned.py,sha256=74c_pQUtpx_x8bM3r5ufuqhaaQxfy6KRUqirdzSac-Q,35999
+nkululeko/models/model_xgb.py,sha256=_VxFFP1QcoyxrwvJSrzdIwwDt85IulUWvg1BxXBgN1Y,6616
 nkululeko/models/model_xgr.py,sha256=H01FJCRgmX2unvambMs5TTCS9sI6VDB9ip9G6rVGt2c,419
 nkululeko/models/tests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nkululeko/models/tests/test_model_svm.py,sha256=spDlZmeBKBdK4EFBpOgEkaAfGeGH9kau6CqSWOY6Uag,1856
 nkululeko/reporting/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nkululeko/reporting/defines.py,sha256=0vh-Tlx4fAPpk1o6mP_4x3EkIoqzYMr38IZnj-JM5z4,641
 nkululeko/reporting/latex_writer.py,sha256=NGwSIfd4nfslDkNUOSZSdqY_VDLA8634thyhe-vj1bY,1824
-nkululeko/reporting/report.py,sha256=bYN8B66gg3IWHAyfd6uIVjpYKy3rOI6aEwgfXU0LSAY,1006
+nkululeko/reporting/report.py,sha256=B5eoIKMz46VKDBsi7M9u_iegzAD-E3eGCmolzSFjZ3c,1118
 nkululeko/reporting/report_item.py,sha256=drkknsyFhGviaPJNmPQtCXJmRhTSSfjNcJt0Bls6JAA,533
-nkululeko/reporting/reporter.py,sha256=-VyV0TZ0vBAx6UZNegnKS3i3WpkF27ntBRlYvp9NNiQ,20174
+nkululeko/reporting/reporter.py,sha256=bfRMN-8nAHAx3iWvjumQsqS-7DHdv7fgAzaUkdi-_qo,20336
 nkululeko/reporting/result.py,sha256=G63a2tHCwHhM6NBJgYzsWKWJm4Yu3r4hsCHA2Km7eHU,1073
 nkululeko/segmenting/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nkululeko/segmenting/seg_inaspeechsegmenter.py,sha256=b3t0zdpJYofKWMyKRMtMMX91xeR-k8d5pbnNaQHcsOE,1902
@@ -117,10 +120,10 @@ nkululeko/utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 nkululeko/utils/files.py,sha256=SrrYaU7AB80MZHiV1jcB0h_zigvYLYgSVNTXV4ao38g,4593
 nkululeko/utils/stats.py,sha256=3Fyx8q8BSKYmiufT6OkRug9RATWmGrr9BaX_y8jziWo,3074
 nkululeko/utils/unzip.py,sha256=G68f5120TjwACZC3bQcneMniddnwubPbBdMc2L5KBOo,1206
-nkululeko/utils/util.py,sha256=6NDKhOx0fV5fKyhSoY4hem96p7OuPcmhCDQR9EzkQhw,17829
-nkululeko-0.94.2.dist-info/licenses/LICENSE,sha256=0zGP5B_W35yAcGfHPS18Q2B8UhvLRY3dQq1MhpsJU_U,1076
-nkululeko-0.94.2.dist-info/METADATA,sha256=YmvStXvJdODIXXgDRQSmzhmJzl_SuX7donWfg66KZBI,2874
-nkululeko-0.94.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-nkululeko-0.94.2.dist-info/entry_points.txt,sha256=lNTkFEdh6Kjo5o95ZAWf_0Lq-4ztGoAoMVSDuPtuyS0,442
-nkululeko-0.94.2.dist-info/top_level.txt,sha256=bf1k1YKkqcXemNX_cUgoyKqQ3_GVErPqAY-53J36jkM,19
-nkululeko-0.94.2.dist-info/RECORD,,
+nkululeko/utils/util.py,sha256=b4wolRb-9eCgfUDaxedcG3ys6HwgJrO5uoYvGSvi1nc,18571
+nkululeko-0.95.0.dist-info/licenses/LICENSE,sha256=0zGP5B_W35yAcGfHPS18Q2B8UhvLRY3dQq1MhpsJU_U,1076
+nkululeko-0.95.0.dist-info/METADATA,sha256=wVnk9XYBqWK3W55yNDp7-AK8h4Xpqgd1sRuWBrDGvV8,2874
+nkululeko-0.95.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+nkululeko-0.95.0.dist-info/entry_points.txt,sha256=lNTkFEdh6Kjo5o95ZAWf_0Lq-4ztGoAoMVSDuPtuyS0,442
+nkululeko-0.95.0.dist-info/top_level.txt,sha256=bf1k1YKkqcXemNX_cUgoyKqQ3_GVErPqAY-53J36jkM,19
+nkululeko-0.95.0.dist-info/RECORD,,