Merge pull request #2 from selimfirat/develop

v0.1.1 update

Author: selimfirat

deploy.sh

@@ -0,0 +1,4 @@
+#!/usr/bin/env bash
+rm -rf ./build/ ./dist/ ./pysad.egg-info/
+python3 setup.py sdist bdist_wheel
+python3 -m twine upload dist/*

docs/features.rst

@@ -16,14 +16,13 @@ Streaming methods efficiently handle the limitied memory and processing time req
 Streaming Anomaly Detection Tools
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-`PySAD` contains stream simulators, evaluators, preprocessors, statistic trackers, postprocessors, probability calibrators and more.
-In addition to streaming models, `PySAD` also provides integrations for batch anomaly detectors of the `PyOD framework <https://github.com/yzhao062/pyod/>`_ so that they can be used in the streaming setting.
+`PySAD` contains stream simulators, evaluators, preprocessors, statistic trackers, postprocessors, probability calibrators and more. In addition to streaming models, `PySAD` also provides integrations for batch anomaly detectors of the `PyOD <https://github.com/yzhao062/pyod/>`_ so that they can be used in the streaming setting.
 
 
 Comprehensiveness
 ^^^^^^^^^^^^^^^^^
 
-`PySAD` provides models that are specifically designed for multivariate and univariate data. One can experiment via `PySAD` in supervised, semi-supervised and unsupervised setting.
+`PySAD` serves models that are specifically designed for both univariate and multivariate data. Furthermore, one can experiment via `PySAD` in supervised, semi-supervised and unsupervised setting.
 
 
 User Friendly
@@ -35,4 +34,4 @@ Users with any experience level can easily use `PySAD`. One can easily design ex
 Free and Open Source Software (FOSS)
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-`PySAD` is distributed under `3-Clause BSD License` and favors FOSS principles.
+`PySAD` is distributed under `BSD License 2.0 <https://github.com/selimfirat/pysad/blob/master/LICENSE>`_ and favors FOSS principles.

docs/license.rst

@@ -1,6 +1,6 @@
 License
 =======
 
-This project is licensed under the `3-Clause BSD License <../LICENSE>`_.
+This project is licensed under the `BSD License 2.0 <https://github.com/selimfirat/pysad/blob/master/LICENSE>`_.
 
 .. literalinclude:: ../LICENSE

examples/example_probability_calibration.py

@@ -9,14 +9,14 @@
     np.random.seed(61)  # Fix seed.
 
     model = xStream()  # Init model.
-    calibrator = ConformalProbabilityCalibrator(windowed=True, window_size=300) # Init probability calibrator.
-    streaming_data = Data().get_iterator("arrhythmia.mat") # Get streamer.
+    calibrator = ConformalProbabilityCalibrator(windowed=True, window_size=300)  # Init probability calibrator.
+    streaming_data = Data().get_iterator("arrhythmia.mat")  # Get streamer.
 
     for i, (x, y_true) in enumerate(streaming_data):  # Stream data.
         anomaly_score = model.fit_score_partial(x)  # Fit to an instance x and score it.
 
         calibrated_score = calibrator.fit_transform(anomaly_score)  # Fit & calibrate score.
 
         # Output if the instance is anomalous.
-        if calibrated_score < 0.05:  # If probability is less than 5%.
+        if calibrated_score > 0.95:  # If probability of being normal is less than 5%.
             print(f"Alert: {i}th data point is anomalous.")

examples/example_usage_short.py

@@ -3,9 +3,10 @@
 from pysad.models import LODA
 from pysad.utils import Data
 
+
 model = LODA()  # Init model
 metric = AUROCMetric()  # Init area under receiver-operating- characteristics curve metric
-streaming_data = Data().get_iterator("arrhythmia.mat") # Get data streamer.
+streaming_data = Data().get_iterator("arrhythmia.mat")  # Get data streamer.
 
 for x, y_true in streaming_data:  # Stream data.
     anomaly_score = model.fit_score_partial(x)  # Fit the instance to model and score the instance.

pysad/models/xstream.py

@@ -77,8 +77,8 @@ def score_partial(self, X):
         return score
 
     def _compute_deltamax(self):
-        #mx = np.max(np.concatenate(self.ref_window, axis=0), axis=0)
-        #mn = np.min(np.concatenate(self.ref_window, axis=0), axis=0)
+        # mx = np.max(np.concatenate(self.ref_window, axis=0), axis=0)
+        # mn = np.min(np.concatenate(self.ref_window, axis=0), axis=0)
         mn, mx = get_minmax_array(np.concatenate(self.ref_window, axis=0))
 
         deltamax = (mx - mn) / 2.0

pysad/transform/ensemble/__init__.py

@@ -1,6 +1,6 @@
 """
 The :mod:`pysad.transform.ensemble` module consist of ensemblers to combine scores from multiple anomaly detectors.
 """
-from .ensemblers import MaximumScoreEnsembler,AverageScoreEnsembler,MedianScoreEnsembler, PYODScoreEnsembler, AverageOfMaximumScoreEnsembler, MaximumOfAverageScoreEnsembler
+from .ensemblers import MaximumScoreEnsembler, AverageScoreEnsembler, MedianScoreEnsembler, AverageOfMaximumScoreEnsembler, MaximumOfAverageScoreEnsembler
 
 __all__ = ["MaximumScoreEnsembler", "MedianScoreEnsembler", "AverageScoreEnsembler", "MaximumOfAverageScoreEnsembler", "AverageOfMaximumScoreEnsembler"]

pysad/transform/probability_calibration/conformal_prediction.py

@@ -39,6 +39,4 @@ def transform_partial(self, score):
         Returns:
             float: Processed score.
         """
-
-        return (np.sum(np.array(self.window.get()) <= score))/ (len(self.window.get()))
-
+        return (np.sum(np.array(self.window.get()) > score)) / (len(self.window.get()))

pysad/transform/probability_calibration/gaussian_tail.py

@@ -54,7 +54,7 @@ def transform_partial(self, score):
         else:
             std = 1.0
 
-        return self._qfunction(score, mean, std)
+        return 1 - self._qfunction(score, mean, std)
 
     def _qfunction(self, x, mean, std):
         """

pysad/version.py

@@ -21,4 +21,4 @@
 # 'X.Y.dev0' is the canonical version of 'X.Y.dev'
 #
 
-__version__ = '0.1.0'
+__version__ = '0.1.1'