Uncouple GLM from DataFrames #57
Merged
Conversation
This file contains 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
This adds a method for fitting a GLM by explicitly specifying the design matrix and response vectors. The resulting GlmMod object has empty ModelFrame and formula fields, and I've changed the few functions that reference these fields to first check if they are defined. Eventually it is probably a good idea to follow @lindahua's suggestion from JuliaStats/Roadmap.jl#11 and split out functionality that depends on DataFrames into a separate package, but most of these changes will be necessary for that as well. I have also added a method for fitting a GLM on a new response vector using the same design matrix. Closes JuliaStats#54
simonster
changed the title
|
I've updated this PR to be in line with JuliaData/DataFrames.jl#571. Now we don't need to use DataFrames at all for GLM.jl, and the usage changes minimally. JuliaStats/StatsBase.jl#51 and then that PR need to be merged first, though. |
|
I've merged the other two PRs, so this is ready to go. The Travis failures are because the required versions of StatsBase and DataFrames (still required for the tests) have not yet been released. I'll wait a day or two for comments before I merge. |
|
I merged this, but since we needed fixes for the removal of StoredArray and unexporting of Cholesky in Base and I didn't want to release it immediately, I created a |
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Outdated suggestions cannot be applied.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.
This adds a method for fitting a GLM by explicitly specifying the design matrix and response vectors. The resulting GlmMod object has empty ModelFrame and formula fields, and I've changed the few functions that reference these fields to first check if they are defined, but eventually it is probably a good idea to follow @lindahua's suggestion from JuliaStats/Roadmap.jl#11 and split out functionality that depends on DataFrames into a separate package.I have also added a method for fitting a GLM on a new response vector using the same design matrix,
using the old parameter estimates as the starting point.I am happy to wait for further thinking about design before merging this. I am opening the PR because I needed this functionality now for a project I'm working on, and most of these changes are likely to be necessary no matter how we decide to handle this.Closes #54