Promotion / feat(config): Vereinheitlichung aller Export- und Theme-Konfigurationen
- Refaktorierung aller Konfigurationsdateien (config_korrelation.py, config_netzwerk.py, config_deskriptive_literaturauswahl.py) zur einheitlichen Steuerung von Visualisierungs- und PNG-Export über zentrale Schalter:
- export_fig_visual
- export_fig_png
- Einzel-Flags pro Plot als Ableitung des zentralen Schalters
- Integration der CI-konformen Templatefunktionen (get_standard_layout, get_colors, get_plot_styles) in alle Analyse-Skripte
- Standardisierte Exportlogik mittels `export_figure(...)` aus plotly_template, inkl. PNG-Support
- Titelergänzung mit Quelle (BibTeX-Dateiname), dynamischem Datum und Fallzahl
- Konsequent responsive Darstellung aller Visualisierungen
📁 Neue Datei:
- deskriptive-literaturauswahl.py: Ergänzung einer eigenständigen Analyse für deskriptive SC-Zeitreihenvisualisierung, basierend auf winsorisiertem Median, Quartilen und Bias-Schwellenwerten.
🔧 Hinweis:
- Die Exportdateien (HTML/PNG) sind CI-konform benannt und versionierbar
This commit is contained in:
@ -2,6 +2,7 @@
|
||||
import os
|
||||
|
||||
# Import theme and all export_fig_... variables from central config before any use
|
||||
from ci_template.plotly_template import export_figure
|
||||
from config_korrelation import (
|
||||
theme,
|
||||
export_fig_visual,
|
||||
@ -58,26 +59,8 @@ def prepare_figure_export(fig, name):
|
||||
safe_filename = slugify(f"{name}_{bib_filename.replace('.bib', '')}")
|
||||
return f"{safe_filename}.html"
|
||||
|
||||
# Zentrale Exportfunktion
|
||||
def export_and_transfer_figure(fig, function_name, export_flag):
|
||||
# Plot immer anzeigen, unabhängig vom Exportflag
|
||||
fig.show(config={"responsive": True})
|
||||
print(f"🔄 Exportversuch: {function_name} | Aktiv: {export_flag}")
|
||||
|
||||
if export_flag:
|
||||
export_path = prepare_figure_export(fig, function_name)
|
||||
try:
|
||||
fig.write_html(export_path, full_html=True, include_plotlyjs="cdn")
|
||||
print(f"HTML gespeichert unter: {export_path}")
|
||||
|
||||
remote_path = "jochen-hanisch@sternenflottenakademie.local:/mnt/deep-space-nine/public/plot/promotion/"
|
||||
result = subprocess.run(["scp", export_path, remote_path], check=True, capture_output=True, text=True)
|
||||
print("✅ Übertragung erfolgreich")
|
||||
# Nach erfolgreichem scp Transfer lokale Datei löschen
|
||||
os.remove(export_path)
|
||||
print(f"🗑️ Lokale Datei '{export_path}' wurde gelöscht.")
|
||||
except Exception as e:
|
||||
print(f"❌ Fehler: {str(e)}")
|
||||
export_figure(fig, function_name, export_flag, export_fig_png)
|
||||
|
||||
# BibTeX-Datei laden
|
||||
bib_path = os.path.join("Research", "Charité - Universitätsmedizin Berlin", "Systematische Literaturrecherche", "Bibliothek", bib_filename)
|
||||
|
||||
@ -1,6 +1,8 @@
|
||||
|
||||
from config_netzwerk import theme, export_fig_visual, bib_filename
|
||||
|
||||
from ci_template.plotly_template import export_figure
|
||||
|
||||
import os
|
||||
|
||||
# Clear the terminal
|
||||
@ -61,20 +63,8 @@ from config_netzwerk import (
|
||||
export_fig_visualize_languages,
|
||||
)
|
||||
|
||||
# Zentrale Exportfunktion für Visualisierungen
|
||||
def export_figure(fig, name, flag, bib_filename=None):
|
||||
if flag:
|
||||
export_path = prepare_figure_export(fig, name)
|
||||
fig.write_html(export_path, full_html=True, include_plotlyjs="cdn")
|
||||
remote_path = "jochen-hanisch@sternenflottenakademie.local:/mnt/deep-space-nine/public/plot/promotion/"
|
||||
try:
|
||||
subprocess.run(["scp", export_path, remote_path], check=True, capture_output=True, text=True)
|
||||
print(f"✅ Datei '{export_path}' erfolgreich übertragen.")
|
||||
os.remove(export_path)
|
||||
print(f"🗑️ Lokale Datei '{export_path}' wurde gelöscht.")
|
||||
except subprocess.CalledProcessError as e:
|
||||
print("❌ Fehler beim Übertragen:")
|
||||
print(e.stderr)
|
||||
export_figure(fig, name, flag, export_fig_png)
|
||||
|
||||
from ci_template.plotly_template import get_colors, get_plot_styles, get_standard_layout
|
||||
|
||||
|
||||
15
config_deskriptive_literaturauswahl.py
Normal file
15
config_deskriptive_literaturauswahl.py
Normal file
@ -0,0 +1,15 @@
|
||||
# Konfiguration für deskriptive-literaturauswahl.py
|
||||
|
||||
# Theme: "light" oder "dark"
|
||||
theme = "light"
|
||||
|
||||
# Exportsteuerung
|
||||
export_fig_visual = False # Steuerung, ob Visualisierungen exportiert werden
|
||||
export_fig_png = False # Separater Schalter für PNG-Export
|
||||
|
||||
# Einzelne Exportflags (optional überschreibbar)
|
||||
export_fig_silhouette_plot = export_fig_visual
|
||||
export_fig_n_jahresverteilung = export_fig_visual
|
||||
export_fig_quartile_trendlinien = export_fig_visual
|
||||
export_fig_balkenverteilung_n = export_fig_visual
|
||||
export_fig_winsor_median = export_fig_visual
|
||||
104
deskriptive-literaturauswahl.py
Normal file
104
deskriptive-literaturauswahl.py
Normal file
@ -0,0 +1,104 @@
|
||||
# ===============================
|
||||
# Deskriptive Analyse: Silhouette-Scores & Fallzahlen
|
||||
# ===============================
|
||||
|
||||
# --- System- und Modul-Imports ---
|
||||
import os
|
||||
import numpy as np
|
||||
import matplotlib.pyplot as plt
|
||||
import plotly.graph_objects as go
|
||||
from datetime import date
|
||||
from scipy.stats.mstats import winsorize
|
||||
|
||||
# --- CI-Template & Konfiguration ---
|
||||
from ci_template.plotly_template import (
|
||||
get_standard_layout,
|
||||
get_colors,
|
||||
set_theme
|
||||
)
|
||||
from ci_template.plotly_template import export_figure
|
||||
from config_deskriptive_literaturauswahl import theme, export_fig_visual
|
||||
|
||||
# --- Initialisierung ---
|
||||
os.system('cls' if os.name == 'nt' else 'clear')
|
||||
set_theme(theme)
|
||||
colors = get_colors()
|
||||
current_date = date.today().isoformat()
|
||||
|
||||
# --- Datenbasis ---
|
||||
years = np.array([
|
||||
2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017,
|
||||
2018, 2019, 2020, 2021, 2022, 2023, 2024, 2025
|
||||
])
|
||||
sc_values = np.array([
|
||||
1.0000, 0.9655, 0.8571, 1.0000, 0.9583, 1.0000, 1.0000, 1.0000,
|
||||
0.9895, 1.0000, 0.9968, 0.9854, 0.9916, 0.9702, 0.9208, 0.9696
|
||||
])
|
||||
n_values = np.array([
|
||||
7, 29, 7, 28, 24, 28, 25, 98,
|
||||
95, 202, 303, 377, 430, 899, 780, 192
|
||||
])
|
||||
|
||||
# --- Berechnungen ---
|
||||
q1 = np.percentile(sc_values, 25)
|
||||
q2 = np.percentile(sc_values, 50)
|
||||
q3 = np.percentile(sc_values, 75)
|
||||
max_value = np.max(sc_values)
|
||||
min_value = np.min(sc_values)
|
||||
|
||||
sc_winsorized = winsorize(sc_values, limits=[0.1, 0.1])
|
||||
median_winsorized = np.median(sc_winsorized)
|
||||
|
||||
fatigue_threshold = 0.96
|
||||
circadian_optimum = 0.99
|
||||
|
||||
# --- Visualisierung ---
|
||||
fig = go.Figure()
|
||||
|
||||
# SC-Linie
|
||||
fig.add_trace(go.Scatter(x=years, y=sc_values, name='Silhouette-Scores',
|
||||
mode='lines+markers', yaxis='y1'))
|
||||
|
||||
# n-Balken
|
||||
fig.add_trace(go.Bar(x=years, y=n_values, name='n-Werte (Fallzahlen)',
|
||||
yaxis='y2', opacity=0.3, marker_color=colors["depthArea"]))
|
||||
|
||||
# Quartile & Bezugslinien
|
||||
fig.add_trace(go.Scatter(x=years, y=[q1]*len(years), mode='lines', name=f'SC Q1 = {q1:.4f}',
|
||||
line=dict(dash='dot', color='green'), yaxis='y1'))
|
||||
fig.add_trace(go.Scatter(x=years, y=[q2]*len(years), mode='lines', name=f'SC Median (Q2) = {q2:.4f}',
|
||||
line=dict(dash='dot', color='blue'), yaxis='y1'))
|
||||
fig.add_trace(go.Scatter(x=years, y=[q3]*len(years), mode='lines', name=f'SC Q3 = {q3:.4f}',
|
||||
line=dict(dash='dot', color='purple'), yaxis='y1'))
|
||||
fig.add_trace(go.Scatter(x=years, y=[min_value]*len(years), mode='lines', name=f'SC Min = {min_value:.4f}',
|
||||
line=dict(dash='dash', color='red'), yaxis='y1'))
|
||||
fig.add_trace(go.Scatter(x=years, y=[max_value]*len(years), mode='lines', name=f'SC Max = {max_value:.4f}',
|
||||
line=dict(dash='dash', color='orange'), yaxis='y1'))
|
||||
fig.add_trace(go.Scatter(x=years, y=[fatigue_threshold]*len(years), mode='lines',
|
||||
name=f'Schwelle: Erschöpfungs-Bias = {fatigue_threshold:.4f}',
|
||||
line=dict(dash='dashdot', color='firebrick'), yaxis='y1'))
|
||||
fig.add_trace(go.Scatter(x=years, y=[circadian_optimum]*len(years), mode='lines',
|
||||
name=f'Idealer Wert (circadian) = {circadian_optimum:.4f}',
|
||||
line=dict(dash='dashdot', color='darkcyan'), yaxis='y1'))
|
||||
fig.add_trace(go.Scatter(x=years, y=[median_winsorized]*len(years), mode='lines',
|
||||
name=f'Winsorisierter Median = {median_winsorized:.4f}',
|
||||
line=dict(dash='dot', color='black'), yaxis='y1'))
|
||||
|
||||
# Layout
|
||||
fig.update_layout(
|
||||
**get_standard_layout(
|
||||
title=f"Silhouette-Scores und Fallzahlen (n={sum(n_values)}, Stand: {current_date})",
|
||||
x_title="Jahr",
|
||||
y_title="Silhouette-Score",
|
||||
yaxis2=dict(
|
||||
title="Fallzahlen (n)",
|
||||
showgrid=False
|
||||
)
|
||||
)
|
||||
)
|
||||
|
||||
# Anzeige
|
||||
fig.show(config={"responsive": True})
|
||||
|
||||
# --- Export ---
|
||||
export_figure(fig, "silhouette_scores_und_fallzahlen", export_fig_silhouette_plot, export_fig_png)
|
||||
Reference in New Issue
Block a user