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

feat: more general hist comparer #235

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

feat: more general hist comparer #235

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
177 changes: 133 additions & 44 deletions popmon/analysis/comparison/hist_comparer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@
# COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER
# IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
# CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

from typing import Any, Callable, Dict, Optional

import numpy as np
import pandas as pd
Expand Down Expand Up @@ -98,58 +98,126 @@ def hist_compare(row, hist_name1="", hist_name2=""):
return pd.Series(x)


class HistComparer(Pipeline):
"""Base pipeline to compare histogram to previous rolling histograms"""

class GenericHistComparer(Pipeline):
def __init__(
self,
func_hist_collector,
read_key,
store_key,
assign_to_key=None,
hist_col="histogram",
suffix="comp",
store_key: str,
hist_col: str,
prefix: str,
assign_to_key: Optional[str] = None,
left_read_key: Optional[str] = None,
right_read_key: Optional[str] = None,
func_left: Optional[Callable] = None,
func_right: Optional[Callable] = None,
suffix1: Optional[str] = None,
suffix2: Optional[str] = None,
left_kwargs: Optional[Dict[Any, Any]] = None,
right_kwargs: Optional[Dict[Any, Any]] = None,
*args,
**kwargs,
):
"""Initialize an instance of RollingHistComparer.
# TODO: add reference type to datastore (along the lines of datastore["references"] =
# {"ref": "Self Reference", "prev1": Rolling Reference (window=1, shift=1)"})

:param func_hist_collector: histogram collection function
:param str read_key: key of input data to read from data store
:param str store_key: key of output data to store in data store
:param str assign_to_key: key of the input data to assign function applied-output to. (optional)
:param str hist_col: column/key in input df/dict that contains the histogram. default is 'histogram'
:param str suffix: column/key of rolling histogram. default is 'roll' -> column = 'histogram_roll'
:param args: (tuple, optional): residual args passed on to func_mean and func_std
:param kwargs: (dict, optional): residual kwargs passed on to func_mean and func_std
"""
if assign_to_key is None:
assign_to_key = read_key
raise ValueError("Ambiguous assign to key")
# assign_to_key = read_key

if left_kwargs is None:
left_kwargs = {}

if right_kwargs is None:
right_kwargs = {}

modules = []

# make left reference histograms
hist_name1 = hist_col
if func_left is not None and suffix1 is not None:
hist_name1 = f"{hist_name1}_{suffix1}"
hist_collector1 = ApplyFunc(
apply_to_key=left_read_key,
assign_to_key=assign_to_key,
)
hist_collector1.add_apply_func(
func=func_left,
entire=True,
suffix=suffix1,
hist_name=hist_col,
**left_kwargs,
)
modules.append(hist_collector1)

# right left reference histograms
hist_name2 = hist_col
if func_right is not None and suffix2 is not None:
hist_name2 = f"{hist_name2}_{suffix2}"
hist_collector2 = ApplyFunc(
apply_to_key=right_read_key,
assign_to_key=assign_to_key,
)
hist_collector2.add_apply_func(
func=func_right,
entire=True,
suffix=suffix2,
hist_name=hist_col,
**right_kwargs,
)
modules.append(hist_collector2)

# make reference histogram(s)
hist_collector = ApplyFunc(
apply_to_key=read_key,
assign_to_key=assign_to_key,
)
hist_collector.add_apply_func(
func=func_hist_collector, entire=True, suffix=suffix, *args, **kwargs
)
# do histogram comparison
hist_comparer = ApplyFunc(
apply_to_key=assign_to_key,
assign_to_key=store_key,
apply_funcs=[
{
"func": hist_compare,
"hist_name1": hist_col,
"hist_name2": hist_col + "_" + suffix,
"prefix": suffix,
"hist_name1": hist_name1,
"hist_name2": hist_name2,
"prefix": prefix,
"axis": 1,
}
],
)
modules.append(hist_comparer)

super().__init__(modules=[hist_collector, hist_comparer])
super().__init__(modules)


class HistComparer(GenericHistComparer):
"""Base pipeline to compare histogram to previous rolling histograms"""

def __init__(
self,
func_hist_collector: Callable,
read_key: str,
store_key: str,
hist_col: str,
suffix: str,
assign_to_key: Optional[str] = None,
**kwargs,
):
"""Initialize an instance of RollingHistComparer.

:param func_hist_collector: histogram collection function
:param str read_key: key of input data to read from data store
:param str store_key: key of output data to store in data store
:param str assign_to_key: key of the input data to assign function applied-output to. (optional)
:param str hist_col: column/key in input df/dict that contains the histogram. default is 'histogram'
:param str suffix: column/key of rolling histogram. default is 'roll' -> column = 'histogram_roll'
:param args: (tuple, optional): residual args passed on to func_mean and func_std
:param kwargs: (dict, optional): residual kwargs passed on to func_mean and func_std
"""
super().__init__(
store_key,
right_read_key=read_key,
hist_col=hist_col,
assign_to_key=assign_to_key,
func_right=func_hist_collector,
right_kwargs=kwargs,
suffix2=suffix,
prefix=suffix,
)


class RollingHistComparer(HistComparer):
Expand Down Expand Up @@ -177,12 +245,11 @@ def __init__(
rolling_hist,
read_key,
store_key,
read_key,
hist_col,
suffix,
assign_to_key=read_key,
hist_col=hist_col,
suffix=suffix,
window=window,
shift=shift,
hist_name=hist_col,
)
self.read_key = read_key
self.window = window
Expand Down Expand Up @@ -244,11 +311,10 @@ def __init__(
expanding_hist,
read_key,
store_key,
read_key,
hist_col,
suffix,
assign_to_key=read_key,
hist_col=hist_col,
suffix=suffix,
shift=shift,
hist_name=hist_col,
)
self.read_key = read_key

Expand Down Expand Up @@ -282,9 +348,9 @@ def __init__(
hist_sum,
reference_key,
store_key,
assign_to_key,
hist_col,
suffix,
assign_to_key=assign_to_key,
hist_col=hist_col,
suffix=suffix,
metrics=[hist_col],
)
self.reference_key = reference_key
Expand All @@ -297,6 +363,29 @@ def transform(self, datastore):
return super().transform(datastore)


class RollingInputFixedReference(GenericHistComparer):
def __init__(
self,
read_key,
reference_key,
store_key,
assign_to_key=None,
window=1,
shift=1,
hist_col="histogram",
suffix1="roll",
suffix2="ref",
prefix="rollref",
):
super().__init__(
read_key, # left read key
store_key,
# right_read_key = reference_key,
hist_col=hist_col,
assign_to_key=assign_to_key,
)


class NormHistComparer(Pipeline):
"""Base pipeline to compare histogram to normalized histograms"""

Expand Down
8 changes: 7 additions & 1 deletion popmon/pipeline/metrics_pipelines.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -102,7 +102,9 @@ def get_splitting_modules(
features=features,
feature_begins_with=f"{time_axis}:",
),
PreviousHistComparer(read_key="split_hists", store_key="comparisons"),
PreviousHistComparer(
read_key="split_hists", store_key="comparisons", suffix="prev1"
),
HistProfiler(read_key="split_hists", store_key="profiles"),
]
return modules
Expand Down Expand Up @@ -226,6 +228,7 @@ def __init__(
reference_key="split_hists",
assign_to_key="split_hists",
store_key="comparisons",
suffix=reference_prefix,
),
RefMedianMadPullCalculator(
reference_key="comparisons",
Expand Down Expand Up @@ -288,6 +291,7 @@ def __init__(
reference_key="split_ref_hists",
assign_to_key="split_hists",
store_key="comparisons",
suffix=reference_prefix,
),
HistProfiler(read_key="split_ref_hists", store_key="ref_profiles"),
RefMedianMadPullCalculator(
Expand Down Expand Up @@ -343,6 +347,7 @@ def __init__(
window=settings.comparison.window,
shift=settings.comparison.shift,
store_key="comparisons",
suffix=reference_prefix,
),
RefMedianMadPullCalculator(
reference_key="comparisons",
Expand Down Expand Up @@ -399,6 +404,7 @@ def __init__(
read_key="split_hists",
shift=settings.comparison.shift,
store_key="comparisons",
suffix=reference_prefix,
),
# 4. profiling of histograms, then pull calculation compared with reference mean and std,
# to obtain normalized residuals of profiles
Expand Down