AdaDelta - #447
Open
Xin999 wants to merge 1162 commits into
Open
Conversation
…ult is not good enough: #363
enable rebin capability in stats step
…xception: Noticed that container memory is 2048, while java process consumes 1800+128, I guess there may be some off-heap memory used. Use 1700 m to add 100 m buffer
Conflicts: src/main/java/ml/shifu/shifu/core/processor/ExportModelProcessor.java
…st to avoid iteration in another thread
# Conflicts: # CHANGES.txt # pom.xml
…ong categorical value limitcation(16k)
fix the iv bug in stats and flatten categorical values in when generating GBT tree
trim categorical value whose length is larger than 16k
Fix two bugs in NS supporting and PMML exporting
change categorical group variable delimiter from ^ to @^
merge latest shifu develop code
Split number type into integer and number
add candidate variable flage and support one-hot encoding for categorical variables
…er worker size in training
Conflicts: src/main/java/ml/shifu/shifu/util/CommonUtils.java
Author
|
AdaDelta & AdaGrad |
| /** | ||
| * | ||
| */ | ||
| package ml.shifu.shifu.core.dtrain; |
Contributor
There was a problem hiding this comment.
please use https://github.com/ShifuML/shifu/blob/master/shifu-codetemplates.xml and https://github.com/ShifuML/shifu/blob/master/shifu-formatter.xml
here is for license.
| import org.slf4j.LoggerFactory; | ||
|
|
||
| /** | ||
| * @author xchen7 |
| if (LOGGER.isDebugEnabled()) { | ||
| LOGGER.debug("Initialized with decay " + decay + " and epsilon " + epsilon + " and learning rate " | ||
| + learningRate); | ||
| } |
Contributor
There was a problem hiding this comment.
here is SLF4j, no need wrappered by LOGGER.isDebugEnabled()
|
|
||
| /* Update gradients after computation. */ | ||
| prevGradients = currGradients; | ||
|
|
Contributor
There was a problem hiding this comment.
the logic looks good, while not sure if master is done, prevGradients will not be recovered, then what will happen? Just image Master is done in iteration 10, then it is recovered to iteration 10, but prevGradients is empty. Other implementations may have the same issue.
MiniZhuwei
force-pushed
the
develop
branch
3 times, most recently
from
July 17, 2018 05:04
1a24e23 to
8d3885b
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.