Update project structure. General folders with subfolders for methods

breimanntools · Oct 3, 2023 · ac524b6 · ac524b6
1 parent d7d7629
commit ac524b6
Show file tree

Hide file tree

Showing 462 changed files with 98 additions and 104 deletions.
diff --git a/aaanalysis/__init__.py b/aaanalysis/__init__.py
@@ -1,8 +1,6 @@
 from aaanalysis.data_handling import load_dataset, load_scales
-from aaanalysis.aaclust import AAclust
-from aaanalysis.aaclust_plot import AAclustPlot
-from aaanalysis.cpp import CPP, CPPPlot, SequenceFeature, SplitRange
-from aaanalysis.dpulearn import dPULearn
+from aaanalysis.feature_engineering import AAclust, AAclustPlot, CPP, CPPPlot, SequenceFeature, SplitRange
+from aaanalysis.pu_learning import dPULearn
 from aaanalysis.plotting import (plot_get_clist, plot_get_cmap, plot_get_cdict,
                                  plot_settings, plot_legend, plot_gcfs)
 from aaanalysis.config import options

diff --git a/aaanalysis/__pycache__/__init__.cpython-39.pyc b/aaanalysis/__pycache__/__init__.cpython-39.pyc
diff --git a/aaanalysis/aaclust/__init__.py b/aaanalysis/aaclust/__init__.py
diff --git a/aaanalysis/aaclust/__pycache__/__init__.cpython-39.pyc b/aaanalysis/aaclust/__pycache__/__init__.cpython-39.pyc
diff --git a/aaanalysis/aaclust/__pycache__/_aaclust_statics.cpython-39.pyc b/aaanalysis/aaclust/__pycache__/_aaclust_statics.cpython-39.pyc
diff --git a/aaanalysis/aaclust_plot/__init__.py b/aaanalysis/aaclust_plot/__init__.py
diff --git a/aaanalysis/aaclust_plot/__pycache__/__init__.cpython-39.pyc b/aaanalysis/aaclust_plot/__pycache__/__init__.cpython-39.pyc
diff --git a/aaanalysis/cpp/__init__.py b/aaanalysis/cpp/__init__.py
diff --git a/aaanalysis/cpp/__pycache__/__init__.cpython-39.pyc b/aaanalysis/cpp/__pycache__/__init__.cpython-39.pyc
diff --git a/aaanalysis/dpulearn/__init__.py b/aaanalysis/dpulearn/__init__.py
diff --git a/aaanalysis/dpulearn/__pycache__/__init__.cpython-39.pyc b/aaanalysis/dpulearn/__pycache__/__init__.cpython-39.pyc
diff --git a/aaanalysis/cpp_plot/__init__.py → aaanalysis/explainable_ai/__init__.py b/aaanalysis/cpp_plot/__init__.py → aaanalysis/explainable_ai/__init__.py
diff --git a/aaanalysis/explainer/tree_based.py → aaanalysis/explainable_ai/tree_based.py b/aaanalysis/explainer/tree_based.py → aaanalysis/explainable_ai/tree_based.py
@@ -17,7 +17,7 @@
 
 # II Main Functions
 class TreeModel:
-    """A wrapper for Tree based prediction models and Tree explainer from SHAP package to
+    """A wrapper for Tree based prediction models and Tree explainable_ai from SHAP package to
     explain prediction (typically binary classification) results at global and individual level"""
     def __init__(self, model=None):
         """"""
@@ -33,7 +33,7 @@ def add_feat_import(self, df_feat=None):
 
 
 class ShapModel:
-    """A wrapper for Tree explainer from SHAP package"""
+    """A wrapper for Tree explainable_ai from SHAP package"""
     def __init__(self, model=None):
         """"""
 

diff --git a/aaanalysis/feature_engineering/__init__.py b/aaanalysis/feature_engineering/__init__.py
@@ -0,0 +1,9 @@
+from aaanalysis.feature_engineering.aaclust_ import AAclust
+from aaanalysis.feature_engineering.aaclust_plot_ import AAclustPlot
+from aaanalysis.feature_engineering.cpp.feature import SequenceFeature
+from aaanalysis.feature_engineering.cpp.feature import SplitRange
+from aaanalysis.feature_engineering.cpp_plot_ import CPPPlot
+from aaanalysis.feature_engineering.cpp_ import CPP
+
+
+__all__ = ["AAclust", "AAclustPlot", "CPP", "CPPPlot", "SequenceFeature", "SplitRange"]
diff --git a/aaanalysis/feature_engineering/__pycache__/__init__.cpython-39.pyc b/aaanalysis/feature_engineering/__pycache__/__init__.cpython-39.pyc
diff --git a/...aclust/__pycache__/aaclust.cpython-39.pyc → ...ering/__pycache__/aaclust_.cpython-39.pyc b/...aclust/__pycache__/aaclust.cpython-39.pyc → ...ering/__pycache__/aaclust_.cpython-39.pyc
diff --git a/...t/__pycache__/aaclust_plot.cpython-39.pyc → .../__pycache__/aaclust_plot_.cpython-39.pyc b/...t/__pycache__/aaclust_plot.cpython-39.pyc → .../__pycache__/aaclust_plot_.cpython-39.pyc
diff --git a/...alysis/cpp/__pycache__/cpp.cpython-39.pyc → ...gineering/__pycache__/cpp_.cpython-39.pyc b/...alysis/cpp/__pycache__/cpp.cpython-39.pyc → ...gineering/__pycache__/cpp_.cpython-39.pyc
diff --git a/aaanalysis/feature_engineering/__pycache__/cpp_plot_.cpython-39.pyc b/aaanalysis/feature_engineering/__pycache__/cpp_plot_.cpython-39.pyc
diff --git a/aaanalysis/dpulearn_plot/__init__.py → ...s/feature_engineering/aaclust/__init__.py b/aaanalysis/dpulearn_plot/__init__.py → ...s/feature_engineering/aaclust/__init__.py
diff --git a/aaanalysis/feature_engineering/aaclust/__pycache__/__init__.cpython-39.pyc b/aaanalysis/feature_engineering/aaclust/__pycache__/__init__.cpython-39.pyc
diff --git a/...clust/__pycache__/_aaclust.cpython-39.pyc → ...clust/__pycache__/_aaclust.cpython-39.pyc b/...clust/__pycache__/_aaclust.cpython-39.pyc → ...clust/__pycache__/_aaclust.cpython-39.pyc
diff --git a/...t/__pycache__/_aaclust_bic.cpython-39.pyc → ...t/__pycache__/_aaclust_bic.cpython-39.pyc b/...t/__pycache__/_aaclust_bic.cpython-39.pyc → ...t/__pycache__/_aaclust_bic.cpython-39.pyc
diff --git a/aaanalysis/feature_engineering/aaclust/__pycache__/_aaclust_statics.cpython-39.pyc b/aaanalysis/feature_engineering/aaclust/__pycache__/_aaclust_statics.cpython-39.pyc
diff --git a/aaanalysis/aaclust/_aaclust.py → ...s/feature_engineering/aaclust/_aaclust.py b/aaanalysis/aaclust/_aaclust.py → ...s/feature_engineering/aaclust/_aaclust.py
diff --git a/aaanalysis/aaclust/_aaclust_bic.py → ...ature_engineering/aaclust/_aaclust_bic.py b/aaanalysis/aaclust/_aaclust_bic.py → ...ature_engineering/aaclust/_aaclust_bic.py
@@ -1,12 +1,9 @@
 """
 This is a script for computing the Bayesian Information Criterion (BIC).
 """
-import time
-import pandas as pd
 import numpy as np
-from collections import OrderedDict
 from scipy.spatial import distance
-from aaanalysis.aaclust._aaclust import compute_centers
+from aaanalysis.feature_engineering.aaclust._aaclust import compute_centers
 
 # I Helper Functions
 

diff --git a/aaanalysis/aaclust/_aaclust_statics.py → ...e_engineering/aaclust/_aaclust_statics.py b/aaanalysis/aaclust/_aaclust_statics.py → ...e_engineering/aaclust/_aaclust_statics.py
@@ -5,7 +5,8 @@
 import pandas as pd
 
 import aaanalysis.utils as ut
-from aaanalysis.aaclust._aaclust import _cluster_center, compute_medoids,  min_cor_all
+from aaanalysis.feature_engineering.aaclust._aaclust import compute_medoids
+
 
 # I Helper Functions
 def _sort_X_labels_names(X, labels=None, names=None):

diff --git a/aaanalysis/aaclust/aaclust.py → aaanalysis/feature_engineering/aaclust_.py b/aaanalysis/aaclust/aaclust.py → aaanalysis/feature_engineering/aaclust_.py
@@ -3,8 +3,6 @@
 """
 import numpy as np
 from typing import Optional, Dict, Union, List, Tuple, Type
-import inspect
-from inspect import isclass
 from sklearn.cluster import KMeans
 from sklearn.metrics import silhouette_score, calinski_harabasz_score
 from sklearn.base import ClusterMixin
@@ -16,10 +14,10 @@
 from aaanalysis.template_classes import Wrapper
 import aaanalysis.utils as ut
 
-from aaanalysis.aaclust._aaclust import (estimate_lower_bound_n_clusters, optimize_n_clusters, merge_clusters,
-                                         compute_centers, compute_medoids)
-from aaanalysis.aaclust._aaclust_bic import bic_score
-from aaanalysis.aaclust._aaclust_statics import compute_correlation, name_clusters
+from aaanalysis.feature_engineering.aaclust._aaclust import (estimate_lower_bound_n_clusters, optimize_n_clusters, merge_clusters,
+                                                             compute_centers, compute_medoids)
+from aaanalysis.feature_engineering.aaclust._aaclust_bic import bic_score
+from aaanalysis.feature_engineering.aaclust._aaclust_statics import compute_correlation, name_clusters
 
 
 # I Helper Functions

diff --git a/aaanalysis/explainer/__init__.py → ...ture_engineering/aaclust_plot/__init__.py b/aaanalysis/explainer/__init__.py → ...ture_engineering/aaclust_plot/__init__.py
diff --git a/aaanalysis/aaclust_plot/aaclust_plot.py → ...ysis/feature_engineering/aaclust_plot_.py b/aaanalysis/aaclust_plot/aaclust_plot.py → ...ysis/feature_engineering/aaclust_plot_.py
diff --git a/...unit/plotting_functions_tests/__init__.py → ...lysis/feature_engineering/cpp/__init__.py b/...unit/plotting_functions_tests/__init__.py → ...lysis/feature_engineering/cpp/__init__.py
diff --git a/aaanalysis/feature_engineering/cpp/__pycache__/__init__.cpython-39.pyc b/aaanalysis/feature_engineering/cpp/__pycache__/__init__.cpython-39.pyc
diff --git a/...lysis/cpp/__pycache__/_cpp.cpython-39.pyc → ...ering/cpp/__pycache__/_cpp.cpython-39.pyc b/...lysis/cpp/__pycache__/_cpp.cpython-39.pyc → ...ering/cpp/__pycache__/_cpp.cpython-39.pyc
diff --git a/...p/__pycache__/_feature_pos.cpython-39.pyc → ...p/__pycache__/_feature_pos.cpython-39.pyc b/...p/__pycache__/_feature_pos.cpython-39.pyc → ...p/__pycache__/_feature_pos.cpython-39.pyc
diff --git a/.../__pycache__/_feature_stat.cpython-39.pyc → .../__pycache__/_feature_stat.cpython-39.pyc b/.../__pycache__/_feature_stat.cpython-39.pyc → .../__pycache__/_feature_stat.cpython-39.pyc
diff --git a/...ysis/cpp/__pycache__/_part.cpython-39.pyc → ...ring/cpp/__pycache__/_part.cpython-39.pyc b/...ysis/cpp/__pycache__/_part.cpython-39.pyc → ...ring/cpp/__pycache__/_part.cpython-39.pyc
diff --git a/...sis/cpp/__pycache__/_split.cpython-39.pyc → ...ing/cpp/__pycache__/_split.cpython-39.pyc b/...sis/cpp/__pycache__/_split.cpython-39.pyc → ...ing/cpp/__pycache__/_split.cpython-39.pyc
diff --git a/...is/cpp/__pycache__/feature.cpython-39.pyc → ...ng/cpp/__pycache__/feature.cpython-39.pyc b/...is/cpp/__pycache__/feature.cpython-39.pyc → ...ng/cpp/__pycache__/feature.cpython-39.pyc
diff --git a/aaanalysis/cpp/_cpp.py → aaanalysis/feature_engineering/cpp/_cpp.py b/aaanalysis/cpp/_cpp.py → aaanalysis/feature_engineering/cpp/_cpp.py
diff --git a/aaanalysis/cpp/_feature_pos.py → ...s/feature_engineering/cpp/_feature_pos.py b/aaanalysis/cpp/_feature_pos.py → ...s/feature_engineering/cpp/_feature_pos.py
@@ -3,8 +3,8 @@
 for given parts and features .
 """
 
-from aaanalysis.cpp._part import Parts
-from aaanalysis.cpp._split import Split
+from aaanalysis.feature_engineering.cpp._part import Parts
+from aaanalysis.feature_engineering.cpp._split import Split
 import aaanalysis.utils as ut
 
 

diff --git a/aaanalysis/cpp/_feature_stat.py → .../feature_engineering/cpp/_feature_stat.py b/aaanalysis/cpp/_feature_stat.py → .../feature_engineering/cpp/_feature_stat.py
@@ -11,7 +11,7 @@
 from statsmodels.stats.multitest import multipletests
 import warnings
 
-from aaanalysis.cpp._split import SplitRange
+from aaanalysis.feature_engineering.cpp._split import SplitRange
 import aaanalysis.utils as ut
 
 

diff --git a/aaanalysis/cpp/_part.py → aaanalysis/feature_engineering/cpp/_part.py b/aaanalysis/cpp/_part.py → aaanalysis/feature_engineering/cpp/_part.py
diff --git a/aaanalysis/cpp/_split.py → aaanalysis/feature_engineering/cpp/_split.py b/aaanalysis/cpp/_split.py → aaanalysis/feature_engineering/cpp/_split.py
diff --git a/aaanalysis/cpp/feature.py → ...alysis/feature_engineering/cpp/feature.py b/aaanalysis/cpp/feature.py → ...alysis/feature_engineering/cpp/feature.py
@@ -10,9 +10,9 @@
 import multiprocessing as mp
 import warnings
 
-from aaanalysis.cpp._feature_pos import SequenceFeaturePositions
-from aaanalysis.cpp._split import Split, SplitRange
-from aaanalysis.cpp._part import Parts
+from aaanalysis.feature_engineering.cpp._feature_pos import SequenceFeaturePositions
+from aaanalysis.feature_engineering.cpp._split import Split, SplitRange
+from aaanalysis.feature_engineering.cpp._part import Parts
 
 import aaanalysis as aa
 import aaanalysis.utils as ut

diff --git a/aaanalysis/cpp/cpp.py → aaanalysis/feature_engineering/cpp_.py b/aaanalysis/cpp/cpp.py → aaanalysis/feature_engineering/cpp_.py
@@ -3,8 +3,8 @@
 """
 import pandas as pd
 
-from aaanalysis.cpp.feature import SequenceFeature
-from aaanalysis.cpp._feature_stat import SequenceFeatureStatistics
+from aaanalysis.feature_engineering.cpp.feature import SequenceFeature
+from aaanalysis.feature_engineering.cpp._feature_stat import SequenceFeatureStatistics
 
 import aaanalysis as aa
 import aaanalysis.utils as ut

diff --git a/...xamples/a7ef104b449e953f/38b060a751ac9638 → .../feature_engineering/cpp_plot/__init__.py b/...xamples/a7ef104b449e953f/38b060a751ac9638 → .../feature_engineering/cpp_plot/__init__.py
diff --git a/aaanalysis/cpp_plot/cpp_plot.py → aaanalysis/feature_engineering/cpp_plot_.py b/aaanalysis/cpp_plot/cpp_plot.py → aaanalysis/feature_engineering/cpp_plot_.py
@@ -6,7 +6,7 @@
 import seaborn as sns
 import inspect
 
-from aaanalysis.cpp._cpp import CPPPlots, get_optimal_fontsize
+from aaanalysis.feature_engineering.cpp._cpp import CPPPlots, get_optimal_fontsize
 import aaanalysis as aa
 
 import aaanalysis.utils as ut

diff --git a/aaanalysis/pertubation/__init__.py b/aaanalysis/pertubation/__init__.py
diff --git a/aaanalysis/pertubation/aamut/__init__.py b/aaanalysis/pertubation/aamut/__init__.py
diff --git a/aaanalysis/pertubation/aamut_.py b/aaanalysis/pertubation/aamut_.py
diff --git a/aaanalysis/pu_learning/__init__.py b/aaanalysis/pu_learning/__init__.py
@@ -0,0 +1,3 @@
+from aaanalysis.pu_learning.dpulearn_ import dPULearn
+
+__all__ = ["dPULearn"]
diff --git a/aaanalysis/pu_learning/__pycache__/__init__.cpython-39.pyc b/aaanalysis/pu_learning/__pycache__/__init__.cpython-39.pyc
diff --git a/...learn/__pycache__/dpulearn.cpython-39.pyc → ...ning/__pycache__/dpulearn_.cpython-39.pyc b/...learn/__pycache__/dpulearn.cpython-39.pyc → ...ning/__pycache__/dpulearn_.cpython-39.pyc
diff --git a/aaanalysis/pu_learning/dpulearn/__init__.py b/aaanalysis/pu_learning/dpulearn/__init__.py
diff --git a/aaanalysis/dpulearn/dpulearn.py → aaanalysis/pu_learning/dpulearn_.py b/aaanalysis/dpulearn/dpulearn.py → aaanalysis/pu_learning/dpulearn_.py
diff --git a/aaanalysis/pu_learning/dpulearn_plot/__init__.py b/aaanalysis/pu_learning/dpulearn_plot/__init__.py
diff --git a/aaanalysis/dpulearn_plot/dpulearn_plot.py → aaanalysis/pu_learning/dpulearn_plot_.py b/aaanalysis/dpulearn_plot/dpulearn_plot.py → aaanalysis/pu_learning/dpulearn_plot_.py
diff --git a/docs/build/doctrees/environment.pickle b/docs/build/doctrees/environment.pickle
diff --git a/docs/build/doctrees/generated/aaanalysis.AAclust.doctree b/docs/build/doctrees/generated/aaanalysis.AAclust.doctree
diff --git a/docs/build/doctrees/generated/aaanalysis.AAclustPlot.doctree b/docs/build/doctrees/generated/aaanalysis.AAclustPlot.doctree
diff --git a/docs/build/doctrees/generated/aaanalysis.CPP.doctree b/docs/build/doctrees/generated/aaanalysis.CPP.doctree
diff --git a/docs/build/doctrees/generated/aaanalysis.CPPPlot.doctree b/docs/build/doctrees/generated/aaanalysis.CPPPlot.doctree
diff --git a/docs/build/doctrees/generated/aaanalysis.SequenceFeature.doctree b/docs/build/doctrees/generated/aaanalysis.SequenceFeature.doctree
diff --git a/docs/build/doctrees/generated/aaanalysis.dPULearn.doctree b/docs/build/doctrees/generated/aaanalysis.dPULearn.doctree
diff --git a/docs/build/doctrees/generated/tutorial1_quick_start.doctree b/docs/build/doctrees/generated/tutorial1_quick_start.doctree
diff --git a/docs/build/html/_downloads/004048c0cbb6684bdb9047282ab71735/aaanalysis-plot_settings-2.pdf b/docs/build/html/_downloads/004048c0cbb6684bdb9047282ab71735/aaanalysis-plot_settings-2.pdf
diff --git a/docs/build/html/_downloads/163aacac4bd235c9af7a62d7b4d0c89f/aaanalysis-plot_get_cdict-1.pdf b/docs/build/html/_downloads/163aacac4bd235c9af7a62d7b4d0c89f/aaanalysis-plot_get_cdict-1.pdf
diff --git a/docs/build/html/_downloads/1f3abea1675a65bb341756c52c9927f4/aaanalysis-plot_gcfs-1.pdf b/docs/build/html/_downloads/1f3abea1675a65bb341756c52c9927f4/aaanalysis-plot_gcfs-1.pdf
diff --git a/docs/build/html/_downloads/72c2e4be500ecf10c85a4e6f81c365fc/aaanalysis-plot_legend-1.pdf b/docs/build/html/_downloads/72c2e4be500ecf10c85a4e6f81c365fc/aaanalysis-plot_legend-1.pdf
diff --git a/docs/build/html/_downloads/795a736e5af756908120c8bda412fd28/aaanalysis-plot_get_cmap-1.pdf b/docs/build/html/_downloads/795a736e5af756908120c8bda412fd28/aaanalysis-plot_get_cmap-1.pdf
diff --git a/docs/build/html/_downloads/88d7f3f7cb5a284c0bfaa377fb4ce1d8/aaanalysis-plot_get_clist-1.pdf b/docs/build/html/_downloads/88d7f3f7cb5a284c0bfaa377fb4ce1d8/aaanalysis-plot_get_clist-1.pdf
diff --git a/docs/build/html/_downloads/fb29bffb69140db4f68c4eb913c6f7d3/aaanalysis-plot_settings-1.pdf b/docs/build/html/_downloads/fb29bffb69140db4f68c4eb913c6f7d3/aaanalysis-plot_settings-1.pdf
diff --git a/docs/build/html/_images/output_13_1.png b/docs/build/html/_images/output_13_1.png
diff --git a/docs/build/html/_sources/generated/tutorial1_quick_start.rst.txt b/docs/build/html/_sources/generated/tutorial1_quick_start.rst.txt
@@ -131,7 +131,7 @@ A feature matrix from a given set of CPP features can be created using
 
 .. parsed-literal::
 
-    Mean accuracy of 0.58
+    Mean accuracy of 0.57
 
 
 Creating more features with CPP will take a little time, but improve
@@ -160,7 +160,7 @@ prediction performance:
 
 .. parsed-literal::
 
-    Mean accuracy of 0.9
+    Mean accuracy of 0.88