-
hi @rbgramacy and @sentian,
I've updated this repo (note that it has moved from mataddy to taddylab, which makes it easier for me to manage collaboration).
The new reorg has the old material in code…
-
After creating a model multiplexer for regr.glmnet, any attempt to call predict with that learner will generate an error. For example...
```
library(mlr)
base.learners = list(
makeLearner("re…
-
When I am uploading files (HD->ACD), I find that when there is some additional activity on the drive will lower my upload rate considerably.
I have found the root issue to be that the read size by ea…
-
See http://www.codewars.com/kata/reviews/552253df667a8dbf76000447/groups/55365b9986b7406e5e000071#553c8adc60ed0b9bbc000037. I've tried that "solution" on a random other kata, and it worked. That's qui…
-
I expect my FLS (flexible least squares) implementation will be similar to your LinReg. Do you have a paper or documentation explaining the method you're using? I don't quite follow what the sweep m…
-
As you see from the recent updates, I'm finally getting a chance to get back into `pcox`. I'm working on changing functionality to accommodate methods now (in particular `predict.pcox()`, but it will …
-
Why do `regr.penalized.lasso` and `regr.penalized.lasso` have property "missings"? I don't think that the `penlized` function of package `penalized` can deal with missing data.
-
First of all let me thank you for very needed ensemble package which you wrote
I have come out to an issue when attempting to train a large number of models all for regression pre-selecting the mode…
-
Reported at [SO](http://stackoverflow.com/q/16402236/166749):
```
classifier = OneVsRestClassifier(SVC(class_weight='auto'))
classifier.fit(X1, y1)
y2 = classifier.predict(X2)
```
raises an exceptio…
-
The log likelihood for WLS should. Be the same as for GLS when when the weights are chosen 1/sqrt(sigma) when sigma is an nobs by 1 vector for use in GLS.
This is not the case and the log likeliho…