Cytoscape Docker image, profiling

cytoscape_docker/Dockerfile

@@ -0,0 +1,37 @@
+# based on official Cytoscape image: https://github.com/cytoscape/docker-cytoscape-desktop
+# version more compact due to:
+# - removed unnecessary packages and instructions --> remote display not used
+# - included all environment variables within this Dockerfile, no need to use Docker compose for the build process
+FROM ubuntu:22.04
+
+# environment variables
+ENV CYTOSCAPE_VERSION=3.10.2
+ENV VIRTUAL_SCREEN_WIDTH=1920
+ENV VIRTUAL_SCREEN_HEIGHT=1080
+ENV CYREST_PORT=1234
+ENV SUPERVISOR_CFG="/etc/supervisor/conf.d/supervisord.conf"
+
+# labels
+LABEL cytoscape_version="${CYTOSCAPE_VERSION}"
+LABEL context="independent CyREST instance for graph rendering"
+
+# change user
+USER root
+# install necessary packages
+RUN apt update && apt full-upgrade -y && apt -y install \
+    libxcursor1 \
+    openjdk-17-jre \
+    supervisor \
+    wget \
+    xvfb
+# set Java home directory as permanently set ENV var
+RUN echo "export JAVA_HOME=/usr/lib/jvm/java-17-openjdk-amd64" >> /root/.bashrc
+# download and extract Cytoscape
+RUN wget https://github.com/cytoscape/cytoscape/releases/download/${CYTOSCAPE_VERSION}/cytoscape-unix-${CYTOSCAPE_VERSION}.tar.gz
+RUN tar -xf cytoscape-unix-${CYTOSCAPE_VERSION}.tar.gz && rm cytoscape-unix-${CYTOSCAPE_VERSION}.tar.gz
+
+# supervisord config
+# supervisor used to start the virtual display via Xvfb and the Cytoscape process with web server (CyREST)
+COPY supervisord.conf ${SUPERVISOR_CFG}
+EXPOSE ${CYREST_PORT}
+CMD ["/usr/bin/supervisord", "-c", "/etc/supervisor/conf.d/supervisord.conf"]

cytoscape_docker/supervisord.conf

@@ -0,0 +1,22 @@
+# based on official Cytoscape image: https://github.com/cytoscape/docker-cytoscape-desktop
+[supervisord]
+nodaemon=true
+
+[program:xvfb]
+priority=1
+directory=/
+command=/usr/bin/Xvfb :1 -screen 0 "%(ENV_VIRTUAL_SCREEN_WIDTH)s"x"%(ENV_VIRTUAL_SCREEN_HEIGHT)s"x24 +extension RANDR
+user=root
+autostart=true
+autorestart=true
+stopsignal=QUIT
+stdout_logfile=/var/log/xvfb.log
+stderr_logfile=/var/log/xvfb.err
+
+[program:cytoscape]
+priority=30
+command=/bin/bash -c 'rm -rf /root/CytoscapeConfiguration && /cytoscape-unix-%(ENV_CYTOSCAPE_VERSION)s/cytoscape.sh --rest %(ENV_CYREST_PORT)s'
+user=root
+autostart=true
+stopsignal=QUIT
+environment=DISPLAY=":1",HOME="/root"

scripts/analyse_dataset.py

@@ -1,5 +1,8 @@
+import cProfile
+import pstats
 import typing
-from typing import cast
+from pathlib import Path
+from typing import Final, cast
 
 from pandas import DataFrame
 
@@ -38,6 +41,11 @@ from lang_main.types import (
     TimelineCandidates,
 )
 
+# ** profiling
+ONLY_PROFILING_REPORT: Final[bool] = True
+USE_PROFILING: Final[bool] = True
+PROFILE_REPORT_NAME: Final[str] = 'prof_report.profdata'
+
 # ** build pipelines
 pipe_target_feat = build_base_target_feature_pipe()
 pipe_merge = build_merge_duplicates_pipe()
@@ -168,4 +176,18 @@ def main() -> None:
 
 
 if __name__ == '__main__':
-    main()
+    report_path = Path.cwd() / 'profiling'
+    if not report_path.exists():
+        report_path.mkdir(parents=True, exist_ok=True)
+    report_file = report_path / PROFILE_REPORT_NAME
+    if ONLY_PROFILING_REPORT:
+        p_stats = pstats.Stats(str(report_file))
+        p_stats.sort_stats(pstats.SortKey.CUMULATIVE).print_stats(60)
+        p_stats.sort_stats('tottime').print_stats(60)
+    elif USE_PROFILING:
+        cProfile.run('main()', str(report_file))
+        p_stats = pstats.Stats(str(report_file))
+        p_stats.sort_stats(pstats.SortKey.CUMULATIVE).print_stats(30)
+        p_stats.sort_stats('tottime').print_stats(30)
+    else:
+        main()

scripts/lang_main_config.toml

@@ -2,10 +2,10 @@
 
 [paths]
 inputs = './inputs/'
-results = './results/dummy_N_1000/'
-dataset = '../data/Dummy_Dataset_N_1000.csv'
-# results = './results/test_20240807/'
-# dataset = '../data/02_202307/Export4.csv'
+# results = './results/dummy_N_1000/'
+# dataset = '../data/Dummy_Dataset_N_1000.csv'
+results = './results/test_20240807/'
+dataset = '../data/02_202307/Export4.csv'
 #results = './results/Export7/'
 #dataset = './01_03_Rohdaten_202403/Export7_59499_Zeilen.csv'
 #results = './results/Export7_trunc/'
@@ -36,7 +36,7 @@ threshold_amount_characters = 5
 threshold_similarity = 0.8
 
 [graph_postprocessing]
-threshold_edge_weight = 1
+threshold_edge_weight = 150
 
 [time_analysis.uniqueness]
 threshold_unique_texts = 4