General features

.gitignore

@@ -166,6 +166,24 @@ cython_debug/
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
 #.idea/
+.idea/
 
 # PyPI configuration file
 .pypirc
+
+# GEDLIB build artifacts
+gedlib/build/
+gedlib/cmake-build-debug/
+gedlib/lib/
+gedlib/bin/
+gedlib/docs/
+gedlib/median/bin/
+gedlib/median/output/
+gedlib/tests/*/bin/
+gedlib/tests/*/output/
+gedlib/ext/.INSTALLED
+gedlib/ext/**/build/
+gedlib/ext/**/lib/*.dylib
+gedlib/ext/**/lib/*.so
+gedlib/ext/**/bin/
+gedlib/ext/libsvm.3.22/*.o

Makefile

@@ -0,0 +1,100 @@
+SHELL := /bin/bash
+.SHELLFLAGS := -eu -o pipefail -c
+
+POETRY ?= poetry
+PYTHON ?= python
+POETRY_RUN ?= $(POETRY) run
+RUN_PY ?= $(POETRY_RUN) $(PYTHON)
+CMAKE ?= cmake
+GEDLIB_BUILD_DIR ?= gedlib/build
+GEDLIB_LIB ?= gxl
+GXL_DATASET ?= aids
+GXL_OPTIONS := aids imdb proteins
+DATA_ROOT ?= data
+TUD_ROOT ?= $(DATA_ROOT)/_tud
+DATASETS ?= AIDS IMDB-BINARY PROTEINS
+
+.PHONY: help gedlib-install gedlib-configure gedlib-build gedlib-all gedlib-test \
+	convert-gxl convert-json convert-txt lower-bound lower-bound-validate \
+	exact-ged gedlib-run gedlib-edit-path simgnn-edit-path apply-edit-path \
+	gedlib-validate-path simgnn-validate-path generate-gxl-collection \
+	generate-json-pairs simgnn-train simgnn-eval test-python install-datasets
+
+help: ## Show available make targets
+	@printf "Available targets:\\n"
+	@grep -E '^[a-zA-Z0-9_.-]+:.*?##' $(MAKEFILE_LIST) | sort | awk 'BEGIN {FS = ":.*?##"} {printf "  %-24s %s\n", $$1, $$2}'
+
+gedlib-install: ## Run GEDLIB install script with cleanup and docs (set GEDLIB_LIB=gxl|... as needed)
+	cd gedlib && $(PYTHON) install.py --clean --doc --lib $(GEDLIB_LIB)
+
+gedlib-configure: ## Configure GEDLIB CMake build directory
+	$(CMAKE) -S gedlib -B $(GEDLIB_BUILD_DIR)
+
+gedlib-build: gedlib-configure ## Build GEDLIB via CMake
+	$(CMAKE) --build $(GEDLIB_BUILD_DIR)
+
+gedlib-all: ## Install and build GEDLIB
+	$(MAKE) gedlib-install
+	$(MAKE) gedlib-build
+
+gedlib-test: ## Run GEDLIB ctest suite after building
+	cd $(GEDLIB_BUILD_DIR) && ctest
+
+convert-gxl: ## Convert a dataset to GXL/XML for GEDLIB (set GXL_DATASET=$(GXL_OPTIONS))
+	cd src/converters/gxl_xml && \
+	if [ ! -f preprocess_$(GXL_DATASET).py ]; then \
+		echo "Unknown dataset '$(GXL_DATASET)'. Choose from: $(GXL_OPTIONS)"; \
+		exit 1; \
+	fi; \
+	$(RUN_PY) preprocess_$(GXL_DATASET).py
+
+convert-json: ## Convert datasets to JSON pairs for SimGNN
+	cd src/converters/json && $(RUN_PY) preprocess_all.py
+
+convert-txt: ## Convert datasets to TXT graph pairs
+	cd src/converters/txt && $(RUN_PY) preprocess_all.py
+
+lower-bound: ## Estimate lower bounds for graph pairs
+	cd heuristics && $(RUN_PY) estimate_lower_bound.py
+
+lower-bound-validate: ## Validate lower bound estimations
+	cd heuristics && $(RUN_PY) validate_lower_bounds.py
+
+exact-ged: ## Compute exact GED using AStar-BMao parser
+	cd src/c++_parsers && $(RUN_PY) astar_exact_ged.py
+
+gedlib-run: ## Run GEDLIB parser for approximate GED
+	cd src/c++_parsers && $(RUN_PY) gedlib_parser.py
+
+gedlib-edit-path: ## Extract GEDLIB edit paths
+	cd src/c++_parsers && $(RUN_PY) gedlib_edit_path.py
+
+simgnn-edit-path: ## Extract SimGNN edit paths
+	cd SimGNN/src && $(RUN_PY) simgnn_extract_edit_path.py
+
+apply-edit-path: ## Apply edit paths to simulate edits
+	cd src/edit_path_test && $(RUN_PY) apply_edit_path.py
+
+gedlib-validate-path: ## Validate GEDLIB edit paths
+	cd src/edit_path_test/test && $(RUN_PY) gedlib_validate_edit_path.py
+
+simgnn-validate-path: ## Validate SimGNN edit paths
+	cd SimGNN/src && $(RUN_PY) simgnn_validate_edit_path.py
+
+generate-gxl-collection: ## Generate synthetic GXL collection for edit-path testing
+	cd src/edit_path_test/generate_synthetic_graphs && $(RUN_PY) generate_gxl_collection.py
+
+generate-json-pairs: ## Generate synthetic JSON pairs for edit-path testing
+	cd src/edit_path_test/generate_synthetic_graphs && $(RUN_PY) generate_json_pairs.py
+
+simgnn-train: ## Train SimGNN model
+	cd SimGNN/src && $(RUN_PY) main.py
+
+simgnn-eval: ## Evaluate SimGNN model
+	cd SimGNN/src && $(RUN_PY) simgnn_evaluate.py
+
+test-python: ## Run pytest suite
+	$(RUN_PY) -m pytest
+
+install-datasets: ## Download TU datasets (any torch_geometric TUDataset name) into data/<dataset>
+	$(RUN_PY) scripts/install_datasets.py --datasets $(DATASETS) --download-root $(TUD_ROOT) --target-root $(DATA_ROOT)