Refactor benchmark scripts: update naming conventions to explicit key=value format for consistency across files

Author: tae898

bindings/python/examples/benchmark-vector/benchmark_vector_params.py

@@ -479,19 +479,17 @@ def run_benchmark():
     global arcadedb
     import arcadedb_embedded as arcadedb
 
-    # Create config suffix for naming
-    config_suffix = f"xmx{args.xmx}_loc{args.location_cache_size}_graph{args.graph_build_cache_size}_mut{args.mutations_before_rebuild}"
-    if args.quantization != "NONE":
-        config_suffix += f"_quant{args.quantization}"
-    if args.store_vectors_in_graph:
-        config_suffix += "_storeVectors"
-
-    db_base_path = f"./jvector_{args.dataset}_size_{args.dataset_size}_{config_suffix}"
+    # Create config suffix for naming with explicit key=value format
+    config_suffix = f"xmx={args.xmx}_loccache={args.location_cache_size}_graphcache={args.graph_build_cache_size}_mutations={args.mutations_before_rebuild}_quant={args.quantization}_store={'ON' if args.store_vectors_in_graph else 'OFF'}"
+
+    db_base_path = (
+        f"./jvector_dataset={args.dataset}_size={args.dataset_size}_{config_suffix}"
+    )
     k_values = [10]
 
-    # Output file
+    # Output file with explicit key=value naming
     md_file = (
-        f"benchmark_jvector_{args.dataset}_size_{args.dataset_size}_{config_suffix}.md"
+        f"benchmark_dataset={args.dataset}_size={args.dataset_size}_{config_suffix}.md"
     )
 
     all_dataset_sizes = {
@@ -554,11 +552,7 @@ def run_benchmark():
                 # quantization passed via args now
 
                 # Unique DB path for this build config
-                db_path = f"{db_base_path}_{max_connections}_{beam_width}"
-                if args.quantization != "NONE":
-                    db_path += f"_{args.quantization}"
-                if args.store_vectors_in_graph:
-                    db_path += "_graphstore"
+                db_path = f"{db_base_path}_maxconn={max_connections}_beam={beam_width}"
 
                 print(
                     f"\n  [Build Config] max_connections={max_connections}, beam={beam_width}, quant={args.quantization}, graph_store={args.store_vectors_in_graph}"

bindings/python/examples/benchmark-vector/plot_benchmark_results.py

@@ -110,12 +110,21 @@ def plot_dataset(dataset_key, dataset_name):
         if "jvector" in filename:
             algo_name = "JVector"
             is_jvector = True
-            log_pattern = f"jvector-*-full_*_memory.log"  # Simplified pattern matching
-            # Need to be more specific to match dataset
-            if "euclidean" in dataset_key:
-                log_pattern = "jvector-euclidean-full_*_memory.log"
-            else:
-                log_pattern = "jvector-angular-full_*_memory.log"
+
+            # Try matching new explicit key=value format log files
+            # Pattern: jvector-dataset={dataset_key}_size=full_..._memory.log
+            # We use glob with wildcards for other params
+            # Note: Now uses xmx= instead of heap= and includes mutations=
+            log_pattern = f"jvector-dataset=*{dataset_key}*_size=full_*_memory.log"
+
+            # Fallback for older formats if not found later
+            if not glob.glob(os.path.join(LOGS_DIR, log_pattern)):
+                log_pattern = f"jvector-*-full_*_memory.log"
+                # Need to be more specific to match dataset
+                if "euclidean" in dataset_key:
+                    log_pattern = "jvector-euclidean-full_*_memory.log"
+                else:
+                    log_pattern = "jvector-angular-full_*_memory.log"
 
         elif "faiss" in filename:
             dataset_type = "euclidean" if "euclidean" in dataset_key else "angular"
@@ -231,11 +240,9 @@ def plot_dataset(dataset_key, dataset_name):
     plt.xlim(0.0, 1.025)
 
     # Save plot
-    output_png = os.path.join(FIGURES_DIR, f"plot_{dataset_key}.png")
-    output_pdf = os.path.join(FIGURES_DIR, f"plot_{dataset_key}.pdf")
-    plt.savefig(output_png)
+    output_pdf = os.path.join(FIGURES_DIR, f"plot_dataset={dataset_key}.pdf")
     plt.savefig(output_pdf)
-    print(f"Saved plots to {output_png} and {output_pdf}")
+    print(f"Saved plot to {output_pdf}")
 
 
 def main():