diff options
| author | Yuhao Yang <hhbyyh@gmail.com> | 2016-01-11 14:48:35 -0800 |
|---|---|---|
| committer | Joseph K. Bradley <joseph@databricks.com> | 2016-01-11 14:48:35 -0800 |
| commit | 4f8eefa36bb90812aac61ac7a762c9452de666bf (patch) | |
| tree | 289fb7093b3c83d92ead9906837b9e471e10aa9d /network | |
| parent | ee4ee02b86be8756a6d895a2e23e80862134a6d3 (diff) | |
| download | spark-4f8eefa36bb90812aac61ac7a762c9452de666bf.tar.gz spark-4f8eefa36bb90812aac61ac7a762c9452de666bf.tar.bz2 spark-4f8eefa36bb90812aac61ac7a762c9452de666bf.zip | |
[SPARK-12685][MLLIB] word2vec trainWordsCount gets overflow
jira: https://issues.apache.org/jira/browse/SPARK-12685
the log of `word2vec` reports
trainWordsCount = -785727483
during computation over a large dataset.
Update the priority as it will affect the computation process.
`alpha = learningRate * (1 - numPartitions * wordCount.toDouble / (trainWordsCount + 1))`
Author: Yuhao Yang <hhbyyh@gmail.com>
Closes #10627 from hhbyyh/w2voverflow.
Diffstat (limited to 'network')
0 files changed, 0 insertions, 0 deletions