kaivalnp commented on code in PR #932:
URL: https://github.com/apache/lucene/pull/932#discussion_r911849144
##########
lucene/core/src/test/org/apache/lucene/util/hnsw/KnnGraphTester.java:
##########
@@ -480,11 +537,15 @@ private int[][] getNN(Path docPath, Path queryPath)
throws IOException {
String hash = Integer.toString(Objects.hash(docPath, queryPath, numDocs,
numIters, topK), 36);
String nnFileName = "nn-" + hash + ".bin";
Path nnPath = Paths.get(nnFileName);
- if (Files.exists(nnPath) && isNewer(nnPath, docPath, queryPath)) {
+ if (Files.exists(nnPath)
Review Comment:
`readNN` reads the cached topK results (if we ran the test with the same
parameters previously). Since our filter is random, we only write and read this
cache if it is deterministic (selectivity == 1)
This won't affect computing recall (which will be done in both cases), but
we recalculate topK every time for the random filter
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
