zhengruifeng commented on pull request #31693:
URL: https://github.com/apache/spark/pull/31693#issuecomment-800713459
this PR maybe too big, after a offline discussion with weichen, I will split
it into serveral prs
This is
zhengruifeng commented on pull request #31693:
URL: https://github.com/apache/spark/pull/31693#issuecomment-797195471
@WeichenXu123
avg(abs(diff_raw_prediction)/raw_prediction_from_center_model) =
0.0021191517840809986
max(abs(diff_raw_prediction)/raw_prediction_from_center_model
zhengruifeng commented on pull request #31693:
URL: https://github.com/apache/spark/pull/31693#issuecomment-796504039
I use the scala case in the ticket:
```
// scalastyle:off println
test("BLR") {
import org.apache.spark.ml.feature.VectorAssembler
val centered
zhengruifeng commented on pull request #31693:
URL: https://github.com/apache/spark/pull/31693#issuecomment-794681499
@srowen We need a seperate PR for branches before 3.1, because of the
blockification introduced in 3.1.1.
We should backport it to 3.0.
as to 2.4, If 2.4.8 is the EOL
zhengruifeng commented on pull request #31693:
URL: https://github.com/apache/spark/pull/31693#issuecomment-793230111
retest this please
This is an automated message from the Apache Git Service.
To respond to the message, ple
zhengruifeng commented on pull request #31693:
URL: https://github.com/apache/spark/pull/31693#issuecomment-792677693
glmnet result for the update testsuite:
-
zhengruifeng commented on pull request #31693:
URL: https://github.com/apache/spark/pull/31693#issuecomment-792496074
@srowen I think it is ok.
@dbtsai @mengxr Could I ping you here? Since it seems that existing
behavior of standardization without removing centers originated from th
