joel/largeRCRF
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

Update README file

Browse source
This commit is contained in:
Joel Therrien 2019-06-05 14:18:51 -07:00
parent f3559c56d2
commit a3551694bd
1 changed files with 10 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
README.md
View file

@ -54,3 +54,13 @@ R> model <- train(..., data=data.env, ...)
when in practice only the trees being trained on need to be loaded. when in practice only the trees being trained on need to be loaded.
You can specify `savePath` to give a directory for `largeRCRF` to save trees in during training, You can specify `savePath` to give a directory for `largeRCRF` to save trees in during training,
which will allow to `largeRCRF` to conserve memory for only those trees being currently trained. which will allow to `largeRCRF` to conserve memory for only those trees being currently trained.
### Training stalls immediately at 0 trees and the CPU is idle
This issue has been observed before on one particular system (and only on that system) but it's not clear what causes it.
It would be appreciated if you could report this bug to [joelt@sfu.ca](mailto:joelt@sfu.ca) and give your operating system
and the version of Java installed (the entire output of `java --version`).
As a workaround, this issue seems to occur randomly; so try restarting your code to see if it runs.