JetBrains-Research / pubtrends

Scientific literature explorer. Runs a Pubmed or Semantic Scholar search and allows user to explore high-level structure of result papers
Apache License 2.0
36 stars 2 forks source link

Node2vec probabilities computation step took 18mins for graph with 1.6k nodes and 330k edges #274

Closed olegs closed 3 years ago

olegs commented 3 years ago

Analysis for the paper in Pubmed "The hallmarks of aging"

[2021-07-20 23:38:46,268: DEBUG/ForkPoolWorker-2] Built similarity graph - 1656 nodes and 333056 edges
[2021-07-20 23:38:46,269: DEBUG/ForkPoolWorker-2] Compute aggregated similarity
[2021-07-20 23:38:47,787: INFO/ForkPoolWorker-2] Extracting topics from paper similarity graph
[2021-07-20 23:38:47,788: DEBUG/ForkPoolWorker-2] Extracting topics from paper similarity graph with node2vec
[2021-07-20 23:38:47,788: DEBUG/ForkPoolWorker-2] Creating weighted graph
[2021-07-20 23:38:50,905: DEBUG/ForkPoolWorker-2] Precomputing random walk probabilities
[2021-07-20 23:38:52,547: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 1 nodes
[2021-07-20 23:40:10,632: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 101 nodes
[2021-07-20 23:41:15,087: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 201 nodes
[2021-07-20 23:42:23,169: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 301 nodes
[2021-07-20 23:43:28,289: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 401 nodes
[2021-07-20 23:44:34,813: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 501 nodes
[2021-07-20 23:45:47,960: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 601 nodes
[2021-07-20 23:46:54,773: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 701 nodes
[2021-07-20 23:48:01,496: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 801 nodes
[2021-07-20 23:49:09,341: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 901 nodes
[2021-07-20 23:50:15,713: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 1001 nodes
[2021-07-20 23:51:08,287: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 1101 nodes
[2021-07-20 23:52:22,966: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 1201 nodes
[2021-07-20 23:53:40,755: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 1301 nodes
[2021-07-20 23:54:43,404: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 1401 nodes
[2021-07-20 23:55:40,184: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 1501 nodes
[2021-07-20 23:56:37,405: DEBUG/ForkPoolWorker-2] Analyzed probabilities for 1601 nodes
olegs commented 3 years ago

Launch node2vec only on sparse graph, see commit: ad7ba075637d73766079ae8bb20a7b12c7f4b146