Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add Pandas Groupby Functionality to Stats Functions #86

Merged
merged 13 commits into from
Sep 8, 2023
Merged

Add Pandas Groupby Functionality to Stats Functions #86

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
e45b783
Adding group by functionality to mean, median, minimum, and variance
Jul 20, 2023
bd5a898
Adding groupby functionlaity for std.py
Jul 29, 2023
ed61088
Adding groupby functionality to check_normality.py
Jul 29, 2023
eb21c50
Adding groupby functionality to percentiles.py
Jul 30, 2023
01fa7df
Updating correlation_nodewise.py to work with groupby functionality
Jul 31, 2023
aa3f428
Adding groupby functionality to calc_boxplot_statistics.py
Jul 31, 2023
75410e1
Black/flake8 formatting changes
Jul 31, 2023
a23ed09
Fixing check_normality.py
Jul 31, 2023
ff45257
Bugfix functions
Aug 30, 2023
201946e
Bugfix remaining functions
Aug 30, 2023
8265da6
Black
Aug 30, 2023
d94a79b
Only aggregate numerical columns
Sep 6, 2023
e3770e6
Black
Sep 6, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 6 additions & 4 deletions thicket/stats/calc_boxplot_statistics.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,8 +51,9 @@ def calc_boxplot_statistics(thicket, columns=[], quartiles=[0.25, 0.5, 0.75], **
col + "_outliers" + q_list: [],
}

for node in pd.unique(thicket.dataframe.reset_index()["node"].tolist()):
values = thicket.dataframe.loc[node][col].tolist()
df = thicket.dataframe.reset_index().groupby("node")
for node, item in df:
values = df.get_group(node)[col].tolist()

q = np.quantile(values, quartiles)
q1 = q[0]
Expand Down Expand Up @@ -107,8 +108,9 @@ def calc_boxplot_statistics(thicket, columns=[], quartiles=[0.25, 0.5, 0.75], **
}
}

for node in pd.unique(thicket.dataframe.reset_index()["node"].tolist()):
values = thicket.dataframe.loc[node][(idx, col)].tolist()
df = thicket.dataframe.reset_index().groupby("node")
for node, item in df:
values = df.get_group(node)[(idx, col)].tolist()

q = np.quantile(values, quartiles)
q1 = q[0]
Expand Down
62 changes: 40 additions & 22 deletions thicket/stats/check_normality.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,46 +35,64 @@ def check_normality(thicket, columns=None):

# thicket object without columnar index
if thicket.dataframe.columns.nlevels == 1:
df = (
thicket.dataframe.select_dtypes(include="number")
.reset_index()
.groupby("node")
.agg(stats.shapiro)
)
for column in columns:
normality = []
for node in pd.unique(thicket.dataframe.reset_index()["node"].tolist()):
pvalue = stats.shapiro(thicket.dataframe.loc[node][column])[1]
for i in range(0, len(df[column])):
pvalue = df[column][i].pvalue

if pvalue < 0.05:
normality.append("False")
thicket.statsframe.dataframe.loc[
df.index[i], column + "_normality"
] = "False"
elif pvalue > 0.05:
normality.append("True")
thicket.statsframe.dataframe.loc[
df.index[i], column + "_normality"
] = "True"
else:
normality.append(pd.NA)
# check to see if exclusive metric
if column in thicket.exc_metrics:
thicket.statsframe.exc_metrics.append(column + "_normality")
# check to see if inclusive metric
else:
thicket.statsframe.inc_metrics.append(column + "_normality")

thicket.statsframe.dataframe[column + "_normality"] = normality
thicket.stataframe.dataframe.loc[
df.index[i], column + "_normality"
] = pd.NA
# check to see if exclusive metric
if column in thicket.exc_metrics:
thicket.statsframe.exc_metrics.append(column + "_normality")
# check to see if inclusive metric
else:
thicket.statsframe.inc_metrics.append(column + "_normality")
# columnar joined thicket object
else:
df = (
thicket.dataframe.select_dtypes(include="number")
.reset_index(level=1)
.groupby("node")
.agg(stats.shapiro)
)
for idx, column in columns:
normality = []
for node in pd.unique(thicket.dataframe.reset_index()["node"].tolist()):
pvalue = stats.shapiro(thicket.dataframe.loc[node][(idx, column)])[1]
for i in range(0, len(df[(idx, column)])):
pvalue = df[(idx, column)][i].pvalue

if pvalue < 0.05:
normality.append("False")
thicket.statsframe.dataframe.loc[
df.index[i], (idx, column + "_normality")
] = "False"
elif pvalue > 0.05:
normality.append("True")
thicket.statsframe.dataframe.loc[
df.index[i], (idx, column + "_normality")
] = "True"
else:
normality.append(pd.NA)
thicket.statsframe.dataframe.loc[
df.index[i], (idx, column + "_normality")
] = pd.NA
# check to see if exclusive metric
if (idx, column) in thicket.exc_metrics:
thicket.statsframe.exc_metrics.append((idx, column + "_normality"))
# check to see if inclusive metric
else:
thicket.statsframe.inc_metrics.append((idx, column + "_normality"))

thicket.statsframe.dataframe[(idx, column + "_normality")] = normality

# sort columns in index
thicket.statsframe.dataframe = thicket.statsframe.dataframe.sort_index(axis=1)
30 changes: 16 additions & 14 deletions thicket/stats/correlation_nodewise.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,27 +38,28 @@ def correlation_nodewise(thicket, column1=None, column2=None, correlation="pears
)
# thicket object without columnar index
if thicket.dataframe.columns.nlevels == 1:
df = thicket.dataframe.reset_index().groupby("node")
correlated = []
for node in thicket.statsframe.dataframe.index.tolist():
for node, item in df:
if correlation == "pearson":
correlated.append(
stats.pearsonr(
thicket.dataframe.loc[node][column1],
thicket.dataframe.loc[node][column2],
df.get_group(node)[column1],
df.get_group(node)[column2],
)[0]
)
elif correlation == "spearman":
correlated.append(
stats.spearmanr(
thicket.dataframe.loc[node][column1],
thicket.dataframe.loc[node][column2],
df.get_group(node)[column1],
df.get_group(node)[column2],
)[0]
)
elif correlation == "kendall":
correlated.append(
stats.kendalltau(
thicket.dataframe.loc[node][column1],
thicket.dataframe.loc[node][column2],
df.get_group(node)[column1],
df.get_group(node)[column2],
)[0]
)
else:
Expand All @@ -70,27 +71,28 @@ def correlation_nodewise(thicket, column1=None, column2=None, correlation="pears
] = correlated
# columnar joined thicket object
else:
df = thicket.dataframe.reset_index().groupby("node")
correlated = []
for node in thicket.statsframe.dataframe.index.tolist():
for node, item in df:
if correlation == "pearson":
correlated.append(
stats.pearsonr(
thicket.dataframe.loc[node][column1],
thicket.dataframe.loc[node][column2],
df.get_group(node)[column1],
df.get_group(node)[column2],
)[0]
)
elif correlation == "spearman":
correlated.append(
stats.spearmanr(
thicket.dataframe.loc[node][column1],
thicket.dataframe.loc[node][column2],
df.get_group(node)[column1],
df.get_group(node)[column2],
)[0]
)
elif correlation == "kendall":
correlated.append(
stats.kendalltau(
thicket.dataframe.loc[node][column1],
thicket.dataframe.loc[node][column2],
df.get_group(node)[column1],
df.get_group(node)[column2],
)[0]
)
else:
Expand Down
15 changes: 4 additions & 11 deletions thicket/stats/maximum.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,8 +3,6 @@
#
# SPDX-License-Identifier: MIT

import pandas as pd

from ..utils import verify_thicket_structures


Expand All @@ -31,32 +29,27 @@ def maximum(thicket, columns=None):

# thicket object without columnar index
if thicket.dataframe.columns.nlevels == 1:
df = thicket.dataframe[columns].reset_index().groupby("node").agg(max)
for column in columns:
maximum = []
for node in pd.unique(thicket.dataframe.reset_index()["node"].tolist()):
maximum.append(max(thicket.dataframe.loc[node][column]))
thicket.statsframe.dataframe[column + "_max"] = df[column]
# check to see if exclusive metric
if column in thicket.exc_metrics:
thicket.statsframe.exc_metrics.append(column + "_max")
# check to see if inclusive metric
else:
thicket.statsframe.inc_metrics.append(column + "_max")

thicket.statsframe.dataframe[column + "_max"] = maximum
# columnar joined thicket object
else:
df = thicket.dataframe[columns].reset_index(level=1).groupby("node").agg(max)
for idx, column in columns:
maximum = []
for node in pd.unique(thicket.dataframe.reset_index()["node"].tolist()):
maximum.append(max(thicket.dataframe.loc[node][(idx, column)]))
thicket.statsframe.dataframe[(idx, column + "_max")] = df[(idx, column)]
# check to see if exclusive metric
if (idx, column) in thicket.exc_metrics:
thicket.statsframe.exc_metrics.append((idx, column + "_max"))
# check to see if inclusive metric
else:
thicket.statsframe.inc_metrics.append((idx, column + "_max"))

thicket.statsframe.dataframe[(idx, column + "_max")] = maximum

# sort columns in index
thicket.statsframe.dataframe = thicket.statsframe.dataframe.sort_index(axis=1)
17 changes: 6 additions & 11 deletions thicket/stats/mean.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,6 @@
# SPDX-License-Identifier: MIT

import numpy as np
import pandas as pd

from ..utils import verify_thicket_structures

Expand All @@ -30,32 +29,28 @@ def mean(thicket, columns=None):

# thicket object without columnar index
if thicket.dataframe.columns.nlevels == 1:
df = thicket.dataframe[columns].reset_index().groupby("node").agg(np.mean)
for column in columns:
mean = []
for node in pd.unique(thicket.dataframe.reset_index()["node"].tolist()):
mean.append(np.mean(thicket.dataframe.loc[node][column]))
thicket.statsframe.dataframe[column + "_mean"] = df[column]
# check to see if exclusive metric
if column in thicket.exc_metrics:
thicket.statsframe.exc_metrics.append(column + "_mean")
# check to see if inclusive metric
else:
thicket.statsframe.inc_metrics.append(column + "_mean")

thicket.statsframe.dataframe[column + "_mean"] = mean
# columnar joined thicket object
else:
df = (
thicket.dataframe[columns].reset_index(level=1).groupby("node").agg(np.mean)
)
for idx, column in columns:
mean = []
for node in pd.unique(thicket.dataframe.reset_index()["node"].tolist()):
mean.append(np.mean(thicket.dataframe.loc[node][(idx, column)]))
thicket.statsframe.dataframe[(idx, column + "_mean")] = df[(idx, column)]
# check to see if exclusive metric
if (idx, column) in thicket.exc_metrics:
thicket.statsframe.exc_metrics.append((idx, column + "_mean"))
# check to see if inclusive metric
else:
thicket.statsframe.inc_metrics.append((idx, column + "_mean"))

thicket.statsframe.dataframe[(idx, column + "_mean")] = mean

# sort columns in index
thicket.statsframe.dataframe = thicket.statsframe.dataframe.sort_index(axis=1)
19 changes: 9 additions & 10 deletions thicket/stats/median.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,6 @@
# SPDX-License-Identifier: MIT

import numpy as np
import pandas as pd

from ..utils import verify_thicket_structures

Expand All @@ -30,32 +29,32 @@ def median(thicket, columns=None):

# thicket object without columnar index
if thicket.dataframe.columns.nlevels == 1:
df = thicket.dataframe[columns].reset_index().groupby("node").agg(np.median)
for column in columns:
median = []
for node in pd.unique(thicket.dataframe.reset_index()["node"].tolist()):
median.append(np.median(thicket.dataframe.loc[node][column]))
thicket.statsframe.dataframe[column + "_median"] = df[column]
# check to see if exclusive metric
if column in thicket.exc_metrics:
thicket.statsframe.exc_metrics.append(column + "_median")
# check to see if inclusive metric
else:
thicket.statsframe.inc_metrics.append(column + "_median")

thicket.statsframe.dataframe[column + "_median"] = median
# columnar joined thicket object
else:
df = (
thicket.dataframe[columns]
.reset_index(level=1)
.groupby("node")
.agg(np.median)
)
for idx, column in columns:
median = []
for node in pd.unique(thicket.dataframe.reset_index()["node"].tolist()):
median.append(np.median(thicket.dataframe.loc[node][(idx, column)]))
thicket.statsframe.dataframe[(idx, column + "_median")] = df[(idx, column)]
# check to see if exclusive metric
if (idx, column) in thicket.exc_metrics:
thicket.statsframe.exc_metrics.append((idx, column + "_median"))
# check to see if inclusive metric
else:
thicket.statsframe.inc_metrics.append((idx, column + "_median"))

thicket.statsframe.dataframe[(idx, column + "_median")] = median

# sort columns in index
thicket.statsframe.dataframe = thicket.statsframe.dataframe.sort_index(axis=1)
16 changes: 4 additions & 12 deletions thicket/stats/minimum.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,8 +3,6 @@
#
# SPDX-License-Identifier: MIT

import pandas as pd

from ..utils import verify_thicket_structures


Expand All @@ -31,32 +29,26 @@ def minimum(thicket, columns=None):

# thicket object without columnar index
if thicket.dataframe.columns.nlevels == 1:
df = thicket.dataframe[columns].reset_index().groupby("node").agg(min)
for column in columns:
minimum = []
for node in pd.unique(thicket.dataframe.reset_index()["node"].tolist()):
minimum.append(min(thicket.dataframe.loc[node][column]))
thicket.statsframe.dataframe[column + "_min"] = df[column]
# check to see if exclusive metric
if column in thicket.exc_metrics:
thicket.statsframe.exc_metrics.append(column + "_min")
# check to see if inclusive metric
else:
thicket.statsframe.inc_metrics.append(column + "_min")

thicket.statsframe.dataframe[column + "_min"] = minimum
# columnar joined thicket object
else:
df = thicket.dataframe[columns].reset_index(level=1).groupby("node").agg(min)
for idx, column in columns:
minimum = []
for node in pd.unique(thicket.dataframe.reset_index()["node"].tolist()):
minimum.append(min(thicket.dataframe.loc[node][(idx, column)]))
thicket.statsframe.dataframe[(idx, column + "_min")] = df[(idx, column)]
# check to see if exclusive metric
if (idx, column) in thicket.exc_metrics:
thicket.statsframe.exc_metrics.append((idx, column + "_min"))
# check to see if inclusive metric
else:
thicket.statsframe.inc_metrics.append((idx, column + "_min"))

thicket.statsframe.dataframe[(idx, column + "_min")] = minimum

# sort columns in index
thicket.statsframe.dataframe = thicket.statsframe.dataframe.sort_index(axis=1)
Loading
Loading