Merge pull request #13 from dPys/automate-machine-gen

feat: make machine.json generation automated

nxbench/cli.py

@@ -1,6 +1,7 @@
 import json
 import logging
 import os
+import platform
 import shutil
 import subprocess
 import sys
@@ -11,8 +12,10 @@
 
 import click
 import pandas as pd
+import psutil
 import requests
 
+from nxbench._version import __version__
 from nxbench.benchmarks.config import DatasetConfig
 from nxbench.benchmarks.utils import get_benchmark_config
 from nxbench.data.loader import BenchmarkDataManager
@@ -33,6 +36,33 @@ def validate_executable(path: str | Path) -> Path:
     return executable
 
 
+def generate_machine_info(results_dir: Path) -> None:
+    """Generate machine info JSON file if it doesn't already exist."""
+    machine = platform.node()
+    results_dir = Path(f"{results_dir}/{machine}")
+    results_dir.mkdir(parents=True, exist_ok=True)
+    machine_info_path = results_dir / "machine.json"
+
+    if machine_info_path.exists():
+        logger.debug(f"Machine info already exists: {machine_info_path}")
+        return
+
+    machine_info = {
+        "arch": platform.machine(),
+        "cpu": platform.processor(),
+        "machine": machine,
+        "num_cpu": str(psutil.cpu_count(logical=True)),
+        "os": f"{platform.system()} {platform.release()}",
+        "ram": str(psutil.virtual_memory().total),
+        "version": __version__,
+    }
+
+    with machine_info_path.open("w") as f:
+        json.dump(machine_info, f, indent=4)
+
+    logger.info(f"Generated machine info at: {machine_info_path}")
+
+
 def get_latest_commit_hash(github_url: str) -> str:
     """
     Fetch the latest commit hash from a GitHub repository.

nxbench/configs/all.yaml

@@ -0,0 +1,312 @@
+algorithms:
+  - name: "pagerank"
+    func: "networkx.pagerank"
+    params:
+      alpha: 0.9
+      tol: 1.0e-6
+    requires_directed: false
+    groups: ["centrality", "random_walk"]
+    min_rounds: 10
+    warmup: true
+    warmup_iterations: 50
+
+  - name: "eigenvector_centrality"
+    func: "networkx.eigenvector_centrality"
+    requires_directed: false
+    groups: ["centrality", "path_based"]
+    min_rounds: 5
+    warmup: true
+    warmup_iterations: 20
+    validate_result: "nxbench.validation.validate_node_scores"
+
+  - name: "betweenness_centrality"
+    func: "networkx.betweenness_centrality"
+    params:
+      normalized: true
+      endpoints: false
+    requires_directed: false
+    groups: ["centrality", "path_based"]
+    min_rounds: 5
+    warmup: true
+    warmup_iterations: 20
+    validate_result: "nxbench.validation.validate_node_scores"
+
+  # - name: "edge_betweenness_centrality"
+  #   func: "networkx.edge_betweenness_centrality"
+  #   params:
+  #     normalized: true
+  #   requires_directed: false
+  #   groups: ["centrality", "path_based"]
+  #   min_rounds: 5
+  #   warmup: true
+  #   warmup_iterations: 20
+  #   validate_result: "nxbench.validation.validate_edge_scores"
+
+  # - name: "approximate_all_pairs_node_connectivity"
+  #   func: "networkx.algorithms.approximation.connectivity.all_pairs_node_connectivity"
+  #   params: {}
+  #   requires_directed: false
+  #   groups: ["connectivity", "approximation"]
+  #   min_rounds: 3
+
+  - name: "average_clustering"
+    func: "networkx.average_clustering"
+    params: {}
+    requires_directed: false
+    groups: ["clustering", "graph_structure"]
+    min_rounds: 3
+    validate_result: "nxbench.validation.validate_scalar_result"
+
+  - name: "square_clustering"
+    func: "networkx.square_clustering"
+    params: {}
+    requires_directed: false
+    groups: ["clustering", "graph_structure"]
+    min_rounds: 3
+    validate_result: "nxbench.validation.validate_node_scores"
+
+  - name: "transitivity"
+    func: "networkx.transitivity"
+    params: {}
+    requires_directed: false
+    groups: ["clustering", "graph_structure"]
+    min_rounds: 3
+    validate_result: "nxbench.validation.validate_scalar_result"
+
+  # - name: "all_pairs_node_connectivity"
+  #   func: "networkx.algorithms.connectivity.connectivity.all_pairs_node_connectivity"
+  #   params: {}
+  #   requires_directed: false
+  #   groups: ["connectivity", "paths"]
+  #   min_rounds: 3
+
+  # - name: "local_efficiency"
+  #   func: "networkx.local_efficiency"
+  #   params: {}
+  #   requires_directed: false
+  #   groups: ["efficiency", "graph_measure"]
+  #   min_rounds: 3
+  #   validate_result: "nxbench.validation.validate_node_scores"
+
+  - name: "number_of_isolates"
+    func: "networkx.number_of_isolates"
+    params: {}
+    requires_directed: false
+    groups: ["isolates", "graph_measure"]
+    min_rounds: 3
+
+  # - name: "all_pairs_all_shortest_paths"
+  #   func: "networkx.all_pairs_all_shortest_paths"
+  #   params:
+  #     weight: "weight"
+  #   requires_directed: false
+  #   groups: ["paths", "all_pairs"]
+  #   min_rounds: 3
+
+  # - name: "all_pairs_shortest_path_length"
+  #   func: "networkx.all_pairs_shortest_path_length"
+  #   params: {}
+  #   requires_directed: false
+  #   groups: ["paths", "distance"]
+  #   min_rounds: 3
+  #   validate_result: "nxbench.validation.validate_scalar_result"
+
+  # - name: "all_pairs_shortest_path"
+  #   func: "networkx.all_pairs_shortest_path"
+  #   params: {}
+  #   requires_directed: false
+  #   groups: ["paths", "distance"]
+  #   min_rounds: 3
+
+  # - name: "all_pairs_dijkstra"
+  #   func: "networkx.all_pairs_dijkstra"
+  #   params:
+  #     weight: "weight"
+  #   requires_directed: false
+  #   groups: ["paths", "weighted"]
+  #   min_rounds: 3
+
+  - name: "all_pairs_dijkstra_path_length"
+    func: "networkx.all_pairs_dijkstra_path_length"
+    params:
+      weight: "weight"
+    requires_directed: false
+    groups: ["paths", "weighted"]
+    min_rounds: 3
+    validate_result: "nxbench.validation.validate_scalar_result"
+
+  - name: "all_pairs_bellman_ford_path_length"
+    func: "networkx.all_pairs_bellman_ford_path_length"
+    params:
+      weight: "weight"
+    requires_directed: false
+    groups: ["paths", "weighted"]
+    min_rounds: 3
+    validate_result: "nxbench.validation.validate_scalar_result"
+
+  # - name: "johnson"
+  #   func: "networkx.johnson"
+  #   params:
+  #     weight: "weight"
+  #   requires_directed: false
+  #   groups: ["paths", "weighted"]
+  #   min_rounds: 3
+
+  # - name: "closeness_vitality"
+  #   func: "networkx.closeness_vitality"
+  #   params: {}
+  #   requires_directed: false
+  #   groups: ["vitality", "centrality"]
+  #   min_rounds: 3
+
+datasets:
+  - name: "twitter"
+    source: "networkrepository"
+    params: {}
+
+  - name: "08blocks"
+    source: "networkrepository"
+    params: {}
+
+  - name: "amazon"
+    source: "networkrepository"
+    params: {}
+
+  - name: "google"
+    source: "networkrepository"
+    params: {}
+
+  - name: "enron"
+    source: "networkrepository"
+    params: {}
+
+  - name: "citationCiteseer"
+    source: "networkrepository"
+    params: {}
+
+  - name: "karate"
+    source: "networkrepository"
+    params: {}
+
+  - name: "netscience"
+    source: "networkrepository"
+    params: {}
+
+  - name: "email-Eu-core"
+    source: "networkrepository"
+    params: {}
+
+  - name: "jazz"
+    source: "networkrepository"
+    params: {}
+
+  - name: "erdos_renyi_small"
+    source: "generator"
+    params:
+      generator: "networkx.erdos_renyi_graph"
+      n: 1000
+      p: 0.01
+    metadata:
+      directed: false
+      weighted: false
+
+  - name: "watts_strogatz_small"
+    source: "generator"
+    params:
+      generator: "networkx.watts_strogatz_graph"
+      n: 10000
+      k: 6
+      p: 0.1
+    metadata:
+      directed: false
+      weighted: false
+
+  - name: "barabasi_albert_small"
+    source: "generator"
+    params:
+      generator: "networkx.barabasi_albert_graph"
+      n: 1000
+      m: 3
+    metadata:
+      directed: false
+      weighted: false
+
+  - name: "powerlaw_cluster_small"
+    source: "generator"
+    params:
+      generator: "networkx.powerlaw_cluster_graph"
+      n: 1000
+      m: 2
+      p: 0.1
+    metadata:
+      directed: false
+      weighted: false
+
+  - name: "erdos_renyi_small"
+    source: "generator"
+    params:
+      generator: "networkx.erdos_renyi_graph"
+      n: 1000
+      p: 0.01
+    metadata:
+      directed: false
+      weighted: false
+
+  # - name: "watts_strogatz_large"
+  #   source: "generator"
+  #   params:
+  #     generator: "networkx.watts_strogatz_graph"
+  #     n: 10000
+  #     k: 6
+  #     p: 0.1
+  #   metadata:
+  #     directed: false
+  #     weighted: false
+
+  # - name: "barabasi_albert_large"
+  #   source: "generator"
+  #   params:
+  #     generator: "networkx.barabasi_albert_graph"
+  #     n: 10000
+  #     m: 3
+  #   metadata:
+  #     directed: false
+  #     weighted: false
+
+  # - name: "powerlaw_cluster_large"
+  #   source: "generator"
+  #   params:
+  #     generator: "networkx.powerlaw_cluster_graph"
+  #     n: 10000
+  #     m: 2
+  #     p: 0.1
+  #   metadata:
+  #     directed: false
+  #     weighted: false
+
+validation:
+  skip_slow: false
+  validate_all: true
+  error_on_fail: true
+  report_memory: true
+
+matrix:
+  backend:
+    - "networkx"
+    - "parallel"
+    - "graphblas"
+    - "cugraph"
+  num_threads:
+    - "1"
+    - "4"
+    - "8"
+
+env_config:
+  req:
+    - "networkx==3.4.2"
+    - "nx-parallel-0.3rc0.dev0"
+    - "graphblas_algorithms==2023.10.0"
+    - "nx-cugraph_cu11==24.8.0"
+  pythons:
+    - "3.10"
+    - "3.11"

nxbench/configs/example.yaml

@@ -144,14 +144,6 @@ algorithms:
     min_rounds: 3
     validate_result: "nxbench.validation.validate_scalar_result"
 
-  # - name: "all_pairs_bellman_ford_path"
-  #   func: "networkx.all_pairs_bellman_ford_path"
-  #   params:
-  #     weight: "weight"
-  #   requires_directed: false
-  #   groups: ["paths", "weighted"]
-  #   min_rounds: 3
-
   # - name: "johnson"
   #   func: "networkx.johnson"
   #   params: