File size: 10,732 Bytes
bf620c6 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 |
import java.io.BufferedReader;
import java.io.BufferedWriter;
import java.io.IOException;
import java.nio.charset.StandardCharsets;
import java.nio.file.Files;
import java.nio.file.Path;
import java.nio.file.Paths;
import java.util.*;
import java.util.function.Consumer;
/**
* LRMCseedsTUD_streamsafe.java
*
* Streaming LRMC seeder for graph-classification datasets (e.g., TUDataset:
* ENZYMES, PROTEINS, COLLAB, D&D). Accepts either a single canonical
* 0-indexed undirected edgelist file (u v with u < v), or a directory of such
* files, and produces seeds JSON using the streaming reconstruction entry point
* in clique2_ablations_parallel2.
*
* Usage:
* Single file:
* java -Xmx4g LRMCseedsTUD_streamsafe input_graph.txt output_seeds.json [DIAM|INV_SQRT_LAMBDA2] [epsilon]
*
* Directory of files:
* java -Xmx4g LRMCseedsTUD_streamsafe input_dir output_dir [DIAM|INV_SQRT_LAMBDA2] [epsilon]
*
* Input edgelists are expected 0-indexed, undirected, no self-loops, one line
* per edge with u < v. Output JSON mirrors LRMCseedsReddit_streamsafe format.
*/
public class LRMCseedsTUD_streamsafe {
public static void main(String[] args) throws Exception {
if (args.length < 2) {
System.err.println("Usage: java LRMCseedsTUD_streamsafe <input_file_or_dir> <output_file_or_dir> [alpha_kind] [epsilon]");
return;
}
final Path inPath = Paths.get(args[0]);
final Path outPath = Paths.get(args[1]);
final AlphaKind alphaKind = (args.length >= 3 ? parseAlpha(args[2]) : AlphaKind.DIAM);
final double eps = (args.length >= 4 ? Double.parseDouble(args[3]) : 1e-6);
if (Files.isDirectory(inPath)) {
Files.createDirectories(outPath);
try (java.util.stream.Stream<Path> stream = Files.list(inPath)) {
stream.filter(Files::isRegularFile)
.filter(p -> {
String s = p.getFileName().toString().toLowerCase(Locale.ROOT);
return s.endsWith(".txt") || s.endsWith(".csv");
})
.forEach(p -> {
try {
GraphData Gi = loadEdgeList0Based(p);
String base = p.getFileName().toString().replaceFirst("\\.(txt|csv)$", "");
Path outFile = outPath.resolve(base + ".json");
runOnce(Gi, outFile, alphaKind, eps);
} catch (Exception e) {
throw new RuntimeException("Failed on " + p, e);
}
});
}
} else {
GraphData G = loadEdgeList0Based(inPath);
System.out.printf(Locale.US, "# Loaded edge list: n=%d, m=%d%n", G.n, G.m);
runOnce(G, outPath, alphaKind, eps);
}
}
static void runOnce(GraphData G, Path outSeeds, AlphaKind alphaKind, double eps) throws Exception {
PeakTracker tracker = new PeakTracker(G, eps, alphaKind);
// Streaming entry point (required by current pipeline)
clique2_ablations_parallel2.runLaplacianRMCStreaming(G.adj1Based, tracker);
tracker.writeJson(outSeeds);
System.out.println("# Done. wrote " + outSeeds.toAbsolutePath());
}
// Streaming peak tracker (same output schema as Reddit seeder)
static final class PeakTracker implements Consumer<clique2_ablations_parallel2.SnapshotDTO> {
final GraphData G;
final double epsilon;
final AlphaKind alphaKind;
final boolean[] inC;
final Map<Integer, Integer> bestIdxByComp = new LinkedHashMap<>();
final Map<Integer, Double> bestScoreByComp = new HashMap<>();
final List<Rec> arrivals = new ArrayList<>();
int idx = 0;
static final class Rec {
final int compId;
final int sid;
final double score;
final int[] nodes;
Rec(int compId, int sid, double score, int[] nodes) {
this.compId = compId; this.sid = sid; this.score = score; this.nodes = nodes;
}
}
PeakTracker(GraphData G, double epsilon, AlphaKind alphaKind) {
this.G = G; this.epsilon = epsilon; this.alphaKind = alphaKind;
this.inC = new boolean[G.n];
}
@Override
public void accept(clique2_ablations_parallel2.SnapshotDTO s) {
final int[] nodes = s.nodes;
final int k = nodes.length;
if (k == 0) return;
for (int u : nodes) inC[u] = true;
final double Q = s.Q;
final double sc = k / (Q + epsilon);
final int compId = s.componentId;
final int sid = idx++;
if (!bestIdxByComp.containsKey(compId) || sc > bestScoreByComp.get(compId)) {
bestIdxByComp.put(compId, sid);
bestScoreByComp.put(compId, sc);
}
arrivals.add(new Rec(compId, sid, sc, Arrays.copyOf(nodes, nodes.length)));
for (int u : nodes) inC[u] = false;
}
void writeJson(Path outJson) throws IOException {
final int n = G.n;
boolean[] covered = new boolean[n];
try (BufferedWriter w = Files.newBufferedWriter(outJson, StandardCharsets.UTF_8)) {
w.write("{\n");
w.write("\"meta\":{");
w.write("\"epsilon\":" + epsilon);
w.write(",\"alpha_kind\":\"" + (alphaKind == AlphaKind.DIAM ? "DIAM" : "INV_SQRT_LAMBDA2") + "\"");
w.write(",\"n\":" + G.n);
w.write(",\"m\":" + G.m);
w.write(",\"mode\":\"peaks_per_component+singletons(stream)\"");
w.write("},\n");
w.write("\"clusters\":[\n");
boolean first = true;
int nextClusterId = 0;
for (Rec r : arrivals) {
Integer best = bestIdxByComp.get(r.compId);
if (best != null && best == r.sid) {
if (!first) w.write(",\n");
first = false;
w.write(" {\"cluster_id\":" + (nextClusterId++));
w.write(",\"component_id\":" + r.compId);
w.write(",\"snapshot_id\":" + r.sid);
w.write(",\"score\":" + r.score);
w.write(",\"k_seed\":" + r.nodes.length);
w.write(",\"members\":" + intArrayToJson(r.nodes));
w.write(",\"seed_nodes\":" + intArrayToJson(r.nodes));
w.write("}");
for (int u : r.nodes) covered[u] = true;
}
}
for (int u = 0; u < n; u++) {
if (!covered[u]) {
if (!first) w.write(",\n");
first = false;
int[] singleton = new int[]{u};
w.write(" {\"cluster_id\":" + (nextClusterId++));
w.write(",\"component_id\":-1");
w.write(",\"snapshot_id\":-1");
w.write(",\"score\":0.0");
w.write(",\"k_seed\":1");
w.write(",\"members\":" + intArrayToJson(singleton));
w.write(",\"seed_nodes\":" + intArrayToJson(singleton));
w.write(",\"is_singleton\":true");
w.write("}");
}
}
w.write("\n]}");
}
}
}
// Loader for canonical 0-based undirected edgelists (u < v). Builds 1-based adjacency.
static GraphData loadEdgeList0Based(Path edgesFile) throws IOException {
int[] deg = new int[1 << 12];
int maxNode = -1;
long mUndir = 0;
try (BufferedReader br = Files.newBufferedReader(edgesFile, StandardCharsets.UTF_8)) {
String s;
while ((s = br.readLine()) != null) {
s = s.trim();
if (s.isEmpty() || s.startsWith("#")) continue;
String[] tok = s.split("\\s+|,");
if (tok.length < 2) continue;
int u = Integer.parseInt(tok[0]);
int v = Integer.parseInt(tok[1]);
if (u == v) continue;
int needed = Math.max(u, v) + 1;
if (needed > deg.length) {
int newLen = deg.length;
while (newLen < needed) newLen <<= 1;
deg = Arrays.copyOf(deg, newLen);
}
deg[u]++; deg[v]++;
if (u < v) mUndir++;
if (u > maxNode) maxNode = u;
if (v > maxNode) maxNode = v;
}
}
final int n = maxNode + 1;
@SuppressWarnings("unchecked")
List<Integer>[] adj1 = (List<Integer>[]) new List<?>[n + 1];
for (int i = 1; i <= n; i++) adj1[i] = new ArrayList<>(deg[i - 1]);
try (BufferedReader br = Files.newBufferedReader(edgesFile, StandardCharsets.UTF_8)) {
String s;
while ((s = br.readLine()) != null) {
s = s.trim();
if (s.isEmpty() || s.startsWith("#")) continue;
String[] tok = s.split("\\s+|,");
if (tok.length < 2) continue;
int u = Integer.parseInt(tok[0]);
int v = Integer.parseInt(tok[1]);
if (u == v) continue;
adj1[u + 1].add(v + 1);
adj1[v + 1].add(u + 1);
}
}
GraphData G = new GraphData();
G.n = n; G.m = mUndir; G.adj1Based = adj1;
G.labels = new int[n]; Arrays.fill(G.labels, -1);
G.labelNames = new String[0];
return G;
}
static AlphaKind parseAlpha(String s) {
String t = s.trim().toUpperCase(Locale.ROOT);
if (t.startsWith("DIAM")) return AlphaKind.DIAM;
if (t.contains("LAMBDA")) return AlphaKind.INV_SQRT_LAMBDA2;
return AlphaKind.DIAM;
}
static String intArrayToJson(int[] arr) {
StringBuilder sb = new StringBuilder();
sb.append('[');
for (int i = 0; i < arr.length; i++) {
if (i > 0) sb.append(',');
sb.append(arr[i]);
}
sb.append(']');
return sb.toString();
}
enum AlphaKind {DIAM, INV_SQRT_LAMBDA2}
static final class GraphData {
int n;
long m;
List<Integer>[] adj1Based;
int[] labels;
String[] labelNames;
}
}
|