aboutsummaryrefslogtreecommitdiffstats
path: root/csit.infra.dash/app/cdash/report/graphs.py
diff options
context:
space:
mode:
Diffstat (limited to 'csit.infra.dash/app/cdash/report/graphs.py')
-rw-r--r--csit.infra.dash/app/cdash/report/graphs.py42
1 files changed, 32 insertions, 10 deletions
diff --git a/csit.infra.dash/app/cdash/report/graphs.py b/csit.infra.dash/app/cdash/report/graphs.py
index 44c57d4183..0627411d0f 100644
--- a/csit.infra.dash/app/cdash/report/graphs.py
+++ b/csit.infra.dash/app/cdash/report/graphs.py
@@ -14,11 +14,11 @@
"""Implementation of graphs for iterative data.
"""
-
import plotly.graph_objects as go
import pandas as pd
from copy import deepcopy
+from numpy import percentile
from ..utils.constants import Constants as C
from ..utils.utils import get_color, get_hdrh_latencies
@@ -74,7 +74,7 @@ def select_iterative_data(data: pd.DataFrame, itm:dict) -> pd.DataFrame:
def graph_iterative(data: pd.DataFrame, sel: list, layout: dict,
- normalize: bool=False) -> tuple:
+ normalize: bool=False, remove_outliers: bool=False) -> tuple:
"""Generate the statistical box graph with iterative data (MRR, NDR and PDR,
for PDR also Latencies).
@@ -83,15 +83,19 @@ def graph_iterative(data: pd.DataFrame, sel: list, layout: dict,
:param layout: Layout of plot.ly graph.
:param normalize: If True, the data is normalized to CPU frequency
Constants.NORM_FREQUENCY.
- :param data: pandas.DataFrame
- :param sel: list
- :param layout: dict
- :param normalize: bool
+ :param remove_outliers: If True the outliers are removed before
+ generating the table.
+ :type data: pandas.DataFrame
+ :type sel: list
+ :type layout: dict
+ :type normalize: bool
+ :type remove_outliers: bool
:returns: Tuple of graphs - throughput and latency.
:rtype: tuple(plotly.graph_objects.Figure, plotly.graph_objects.Figure)
"""
- def get_y_values(data, y_data_max, param, norm_factor, release=str()):
+ def get_y_values(data, y_data_max, param, norm_factor, release=str(),
+ remove_outliers=False):
if param == "result_receive_rate_rate_values":
if release == "rls2402":
y_vals_raw = data["result_receive_rate_rate_avg"].to_list()
@@ -100,6 +104,17 @@ def graph_iterative(data: pd.DataFrame, sel: list, layout: dict,
else:
y_vals_raw = data[param].to_list()
y_data = [(y * norm_factor) for y in y_vals_raw]
+
+ if remove_outliers:
+ try:
+ q1 = percentile(y_data, 25, method=C.COMP_PERCENTILE_METHOD)
+ q3 = percentile(y_data, 75, method=C.COMP_PERCENTILE_METHOD)
+ irq = q3 - q1
+ lif = q1 - C.COMP_OUTLIER_TYPE * irq
+ uif = q3 + C.COMP_OUTLIER_TYPE * irq
+ y_data = [i for i in y_data if i >= lif and i <= uif]
+ except TypeError:
+ pass
try:
y_data_max = max(max(y_data), y_data_max)
except TypeError:
@@ -142,7 +157,12 @@ def graph_iterative(data: pd.DataFrame, sel: list, layout: dict,
y_units.update(itm_data[C.UNIT[ttype]].unique().tolist())
y_data, y_tput_max = get_y_values(
- itm_data, y_tput_max, C.VALUE_ITER[ttype], norm_factor, itm["rls"]
+ itm_data,
+ y_tput_max,
+ C.VALUE_ITER[ttype],
+ norm_factor,
+ itm["rls"],
+ remove_outliers
)
nr_of_samples = len(y_data)
@@ -192,7 +212,8 @@ def graph_iterative(data: pd.DataFrame, sel: list, layout: dict,
itm_data,
y_band_max,
C.VALUE_ITER[f"{ttype}-bandwidth"],
- norm_factor
+ norm_factor,
+ remove_outliers=remove_outliers
)
if not all(pd.isna(y_band)):
y_band_units.update(
@@ -221,7 +242,8 @@ def graph_iterative(data: pd.DataFrame, sel: list, layout: dict,
itm_data,
y_lat_max,
C.VALUE_ITER["latency"],
- 1 / norm_factor
+ 1 / norm_factor,
+ remove_outliers=remove_outliers
)
if not all(pd.isna(y_lat)):
customdata = list()