From 55af464bbd1231d858b6fce3634eefd7746503cd Mon Sep 17 00:00:00 2001
From: Michael Reuscher <michael.reuscher@desy.de>
Date: Mon, 24 Jul 2023 15:28:17 +0200
Subject: [PATCH] new plot for long
---
simrunner/data_config.json | 4 +-
visualizer/visualizer/DataFrameCreator.py | 74 +++++++++++++++--------
visualizer/visualizer/SeabornPlotter.py | 36 +++++++----
visualizer/visualizer/main.py | 18 ++++--
4 files changed, 88 insertions(+), 44 deletions(-)
diff --git a/simrunner/data_config.json b/simrunner/data_config.json
index e8bedd8..2ceb794 100644
--- a/simrunner/data_config.json
+++ b/simrunner/data_config.json
@@ -4,8 +4,8 @@
"-i": "0.1s",
"-N": "4"},
"data_config_long":{
- "-s": "4096",
- "-n": "200",
+ "-s": "32",
+ "-n": "5000",
"-i": "0.1s",
"-N": "4"},
"data_config_help":{
diff --git a/visualizer/visualizer/DataFrameCreator.py b/visualizer/visualizer/DataFrameCreator.py
index 38d09a1..ce372d2 100644
--- a/visualizer/visualizer/DataFrameCreator.py
+++ b/visualizer/visualizer/DataFrameCreator.py
@@ -5,6 +5,8 @@ import pandas as pd
class DataFrameCreator:
def __init__(self, f_path):
self.folder_path = f_path
+ self.v_dfs = {}
+ self.l_dfs = {}
@staticmethod
def _merge_dataframes(dfs):
@@ -15,6 +17,48 @@ class DataFrameCreator:
merged_dfs[outer_key] = merged_dataframe_sorted
return merged_dfs
+ def read_various_data(self, root, folder):
+ for filename in folder:
+ if filename.endswith('.dat'):
+ file_path = os.path.join(root, filename)
+ dataframe_name = os.path.splitext(filename)[0]
+ dataframe_type = dataframe_name.split('-')[1]
+ dataframe = pd.read_csv(file_path, delimiter=" ", header=0)
+
+ if dataframe_type in self.v_dfs:
+ inner_dict = self.v_dfs[dataframe_type]
+ if dataframe_name in inner_dict:
+ # DataFrame hinzufügen
+ inner_dict[dataframe_name] = pd.concat([inner_dict[dataframe_name], dataframe],
+ ignore_index=True)
+ else:
+ inner_dict[dataframe_name] = dataframe
+ else:
+ self.v_dfs[dataframe_type] = {
+ dataframe_name: dataframe
+ }
+
+ def read_long_data(self, root, folder):
+ for filename in folder:
+ if filename.endswith('.dat'):
+ file_path = os.path.join(root, filename)
+ dataframe_name = os.path.splitext(filename)[0]
+ dataframe_type = dataframe_name.split('-')[1]
+ dataframe = pd.read_csv(file_path, delimiter=" ", header=0)
+
+ if dataframe_type in self.l_dfs:
+ inner_dict = self.l_dfs[dataframe_type]
+ if dataframe_name in inner_dict:
+ # DataFrame hinzufügen
+ inner_dict[dataframe_name] = pd.concat([inner_dict[dataframe_name], dataframe],
+ ignore_index=True)
+ else:
+ inner_dict[dataframe_name] = dataframe
+ else:
+ self.l_dfs[dataframe_type] = {
+ dataframe_name: dataframe
+ }
+
def create(self):
dfs = {}
for root, _, folder in os.walk(self.folder_path):
@@ -22,31 +66,13 @@ class DataFrameCreator:
print("this was statistic")
continue
elif os.path.basename(root) == "Data_Set_Long":
- print("this was long")
- continue
- for filename in folder:
- if filename.endswith('.dat'):
- file_path = os.path.join(root, filename)
- dataframe_name = os.path.splitext(filename)[0]
- dataframe_type = dataframe_name.split('-')[1]
- dataframe = pd.read_csv(file_path, delimiter=" ", header=0)
-
- if dataframe_type in dfs:
- inner_dict = dfs[dataframe_type]
- if dataframe_name in inner_dict:
- # DataFrame hinzufügen
- inner_dict[dataframe_name] = pd.concat([inner_dict[dataframe_name], dataframe],
- ignore_index=True)
- else:
- inner_dict[dataframe_name] = dataframe
- else:
- dfs[dataframe_type] = {
- dataframe_name: dataframe
- }
+ self.read_long_data(root, folder)
+ else:
+ self.read_various_data(root, folder)
- merged_dfs = self._merge_dataframes(dfs)
-
- return merged_dfs
+ various_dfs = self._merge_dataframes(self.v_dfs)
+ long_dfs = self._merge_dataframes(self.l_dfs)
+ return various_dfs, long_dfs
diff --git a/visualizer/visualizer/SeabornPlotter.py b/visualizer/visualizer/SeabornPlotter.py
index cb9e077..3c370ee 100644
--- a/visualizer/visualizer/SeabornPlotter.py
+++ b/visualizer/visualizer/SeabornPlotter.py
@@ -12,7 +12,6 @@ class SeabornPlotter:
def __init__(self, dataframe, name):
self.dataframe = dataframe
self.name = name
- self.current_plot = None
@staticmethod
def _add_logo(logo_path, logo_size, logo_x, logo_y):
@@ -33,10 +32,15 @@ class SeabornPlotter:
bins_y = np.logspace(-6, 0, num=101)
return bins_x, bins_y
- @staticmethod
- def plot_histogram_various(x, y, title, bins_x, bins_y):
+ def plot_histogram_various(self):
+ df = self.dataframe
+ x = df['size']
+ y = df['duration']
+ title = "Procedure: {}".format(self.name)
+ bins_x, bins_y = self._prepare_bins(df)
f, ax = plt.subplots(figsize=(16, 9))
sns.histplot(x=x, y=y, color="blue", bins=[bins_x, bins_y])
+ self._add_logo("desy_logo.png", logo_size=0.15, logo_x=1.065, logo_y=1.04)
ax.set_xscale('log')
ax.set_yscale('log')
ax.set_xlabel("Data size [Bytes]")
@@ -44,21 +48,24 @@ class SeabornPlotter:
plt.title(title)
plt.grid(True)
plt.xticks(rotation=45)
+ plt.savefig(title + ".svg", format='svg')
+ plt.show()
- def plot(self):
+ def plot_histogram_long(self):
df = self.dataframe
-
- # if self.name
- x = df['size']
+ x = df['start']
y = df['duration']
title = "Procedure: {}".format(self.name)
-
- bins_x, bins_y = self._prepare_bins(df)
- self.plot_histogram_various(x, y, title, bins_x, bins_y)
-
+ f, ax = plt.subplots(figsize=(16, 9))
+ sns.lineplot(x=x, y=y, hue=list(self.dataframe.index.get_level_values(0)), color="blue")
self._add_logo("desy_logo.png", logo_size=0.15, logo_x=1.065, logo_y=1.04)
-
- # Save and View
+ ax.set_xscale('log')
+ ax.set_yscale('log')
+ ax.set_xlabel("Data size [Bytes]")
+ ax.set_ylabel("Time/Event [s]")
+ plt.title(title)
+ plt.grid(True)
+ plt.xticks(rotation=45)
plt.savefig(title + ".svg", format='svg')
plt.show()
@@ -70,3 +77,6 @@ class SeabornPlotter:
+
+
+
diff --git a/visualizer/visualizer/main.py b/visualizer/visualizer/main.py
index 56dde96..2647010 100644
--- a/visualizer/visualizer/main.py
+++ b/visualizer/visualizer/main.py
@@ -16,16 +16,24 @@ def main():
# Import Data
creator = DataFrameCreator(data_path)
- dataframes = creator.create()
+ v_dataframes, l_dataframes = creator.create()
+ for frame in v_dataframes.keys():
+ if not v_dataframes[frame].empty:
+ # Setup Plotter
+ frame_name = frame
+ v_plotter = SeabornPlotter(v_dataframes[frame], frame_name)
+
+ # Plotter run
+ v_plotter.plot_histogram_various()
- for frame in dataframes.keys():
- if not dataframes[frame].empty:
+ for frame in l_dataframes.keys():
+ if not v_dataframes[frame].empty:
# Setup Plotter
frame_name = frame
- plotter = SeabornPlotter(dataframes[frame], frame_name)
+ l_plotter = SeabornPlotter(l_dataframes[frame], frame_name)
# Plotter run
- plotter.plot()
+ l_plotter.plot_histogram_long()
if __name__ == "__main__":
--
GitLab