search

glm-tools / pyglmnet

Python implementation of elastic-net regularized generalized linear models
http://glm-tools.github.io/pyglmnet/
MIT License
279 stars 83 forks source link

MRG: Pyglmnet paper #313

Closed jasmainak closed 4 years ago

jasmainak commented 4 years ago

closes #297

Hi everyone,

@pavanramkumar and I created a first draft of our paper for submission to JOSS. If you want to be listed as an author, please read the draft and leave comments. More importantly:

The guidelines for JOSS submission are here.

Compiling the paper

If you want to compile the paper

Compile!

Tagging all contributors (check mark if we got a response)

codecov-io commented 4 years ago

Codecov Report

Merging #313 into master will not change coverage. The diff coverage is n/a.

Impacted file tree graph

@@           Coverage Diff           @@
##           master     #313   +/-   ##
=======================================
  Coverage   75.66%   75.66%           
=======================================
  Files           4        4           
  Lines         678      678           
  Branches      149      149           
=======================================
  Hits          513      513           
  Misses        128      128           
  Partials       37       37

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 243959c...8bc250f. Read the comment docs.

pavanramkumar commented 4 years ago

Thanks @jasmainak ! Looks good to me

daniel-acuna commented 4 years ago

wow this is great! Syracuse University is good.. but how do I compile the paper? #noob

jasmainak commented 4 years ago

That's a fantastic question @daniel-acuna . Here's how:

Compile!

I'll update this in the first comment so others can see it too.

koerding commented 4 years ago

I am getting "Your paper failed to compile with the following errors:

Can't find any papers to compile. Make sure there's a file named paper.md in your repository."

On Mon, Nov 4, 2019 at 12:09 AM Mainak Jas notifications@github.com wrote:

That's a fantastic question @daniel-acuna https://github.com/daniel-acuna . Here's how:

Compile!

I'll update this in the first comment so others can see it too.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/glm-tools/pyglmnet/pull/313?email_source=notifications&email_token=AA2XUNMUQTN7PHM67AWIMDDQR6OBDA5CNFSM4IYD6K42YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEC6INQY#issuecomment-549226179, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA2XUNKHJE4MGX275463NW3QR6OBDANCNFSM4IYD6K4Q .

-- I love those that desire the impossible - Goethe.

Konrad Kording @kordinglab, UPenn, Richards 404b cell:773 - liberat (5423728) office:(215) 898-5826 kordinglab.com

daniel-acuna commented 4 years ago

@koerding You have to select the paper branch. Works like a charm

koerding commented 4 years ago

ah yes thanks. it does work great. Should the paper contain more information on potential usage scenarios or is it good like this?

On Mon, Nov 4, 2019 at 8:43 AM Daniel E. Acuna notifications@github.com wrote:

@koerding https://github.com/koerding You have to select the paper branch. Works like a charm

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/glm-tools/pyglmnet/pull/313?email_source=notifications&email_token=AA2XUNIZJPLC2AA2VJU7EI3QSARGXA5CNFSM4IYD6K42YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEC7IWOQ#issuecomment-549358394, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA2XUNKCLZVBUWBSY6KUKILQSARGXANCNFSM4IYD6K4Q .

-- I love those that desire the impossible - Goethe.

Konrad Kording @kordinglab, UPenn, Richards 404b cell:773 - liberat (5423728) office:(215) 898-5826 kordinglab.com

cxrodgers commented 4 years ago

Thanks everyone especially jasmainak and pavanramkumar for spearheading this great project. I confirm my name and affiliation. I added my orcid as a code suggestion (0000-0003-1762-3450). Reviewing the manuscript now.

titipata commented 4 years ago

I totally agree with @koerding. I think it would be great to add usage or example usage section similar to examples in README and documentation. Additionally, we can provide examples in the summary section on use cases where distribution and regularization are not provided in statsmodels or scikit-learn.

cxrodgers commented 4 years ago

I made a few stylistic suggestions in this PR: https://github.com/jasmainak/pyglmnet/pull/1

I like the current usage example with glm.fit and glm.predict. I think it might be nice to add a few lines that demonstrate the different noise models (poisson, binomial, etc) and different regularization terms (L1, L2, etc) since those are the main features that are emphasized in the text.

koerding commented 4 years ago

hey. I can' seem to edit it. I forked pyglmnet but the paper does not seem to be there. Can I provide feedback in a google doc or by email? sorry, my github skills are not-so-elite

On Mon, Nov 4, 2019 at 10:21 AM Titipat Achakulvisut < notifications@github.com> wrote:

I totally agree with @koerding https://github.com/koerding. I think it would be great to add usage or example usage section similar to examples in README and documentation. Additionally, we can provide examples in the summary section on use cases where distribution and regularization are not provided in statsmodels or scikit-learn.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/glm-tools/pyglmnet/pull/313?email_source=notifications&email_token=AA2XUNILOKD5QKJUX63OUEDQSA4YFA5CNFSM4IYD6K42YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEC7TVCI#issuecomment-549403273, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA2XUNIMAP255BIPKQZ6R7TQSA4YFANCNFSM4IYD6K4Q .

-- I love those that desire the impossible - Goethe.

Konrad Kording @kordinglab, UPenn, Richards 404b cell:773 - liberat (5423728) office:(215) 898-5826 kordinglab.com

jasmainak commented 4 years ago

@koerding if you go here, you should be able to edit the paper and submit a pull request. Can you give it a try? If it doesn't work, we can do this over email or google doc

koerding commented 4 years ago

ah that is great. Thanks! Sorry.

On Mon, Nov 4, 2019 at 10:57 AM Mainak Jas notifications@github.com wrote:

@koerding https://github.com/koerding if you go here https://github.com/jasmainak/pyglmnet/edit/paper/paper/paper.md, you should be able to edit the paper and submit a pull request. Can you give it a try? If it doesn't work, we can do this over email or google doc

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/glm-tools/pyglmnet/pull/313?email_source=notifications&email_token=AA2XUNIW36H6WQ4MUBWB5ILQSBBANA5CNFSM4IYD6K42YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEC7XWSQ#issuecomment-549419850, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA2XUNOTJME6XGXFMMFLN73QSBBANANCNFSM4IYD6K4Q .

-- I love those that desire the impossible - Goethe.

Konrad Kording @kordinglab, UPenn, Richards 404b cell:773 - liberat (5423728) office:(215) 898-5826 kordinglab.com

notuntoward commented 4 years ago

When I try to compile the paper on Whedon, I get the error message:

Your paper failed to compile with the following errors:
Author (Peter Foley) is missing affiliation
jasmainak commented 4 years ago

Please try again. Looks like some of the "code suggestions" were in the incorrect format which broke the compilation. Sorry about that.

pavanramkumar commented 4 years ago

Hey @koerding we'll email you a markdown draft to make it easier!

@titipata @koerding thanks for your comments: happy to receive edits / suggestions!

Have a look here at paper content expectations here: https://joss.readthedocs.io/en/latest/submitting.html

Unlike typical journal papers, JOSS is meant to be a very thin wrapper around the documentation / examples of the OSS project: the latter is peer reviewed and expected to describe use cases and provide examples. Our example gallery currently provides one application with matrix reg (Tikhonov) and one with group lasso: both of which expand the scope of status quo glm libraries significantly

themantalope commented 4 years ago

Hi! Sorry just saw this. Is there something you need me to submit?

jasmainak commented 4 years ago

hi @themantalope can you read the first comment and confirm your authorship by checking the affiliation and adding the orcid? Thanks.

koerding commented 4 years ago

I will be completely out until next thursday. Sorry guys. Huge grant deadlin

On Fri, Nov 8, 2019 at 9:28 AM Mainak Jas notifications@github.com wrote:

hi @themantalope https://github.com/themantalope can you read the first comment https://github.com/glm-tools/pyglmnet/pull/313#issue-319008762 and confirm your authorship by checking the affiliation and adding the orcid? Thanks.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/glm-tools/pyglmnet/pull/313?email_source=notifications&email_token=AA2XUNONXBZZ7SYJZZ5U263QSVZQ5A5CNFSM4IYD6K42YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEDSIUGI#issuecomment-551848473, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA2XUNLV7FJQAEYXIQTOU63QSVZQ5ANCNFSM4IYD6K4Q .

-- I love those that desire the impossible - Goethe.

Konrad Kording @kordinglab, UPenn, Richards 404b cell:773 - liberat (5423728) office:(215) 898-5826 kordinglab.com

jasmainak commented 4 years ago

no worries, we are expecting you to make at least two pull requests since you are taking extra time now ;)

koerding commented 4 years ago

I have done my orcid ;) so just one?

On Fri, Nov 8, 2019 at 4:08 PM Mainak Jas notifications@github.com wrote:

no worries, we are expecting you to make at least two pull requests since you are taking extra time now ;)

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/glm-tools/pyglmnet/pull/313?email_source=notifications&email_token=AA2XUNIK66ITPVV664WPFTLQSXILHA5CNFSM4IYD6K42YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEDTLG2Q#issuecomment-551990122, or unsubscribe https://github.com/notifications/unsubscribe-auth/AA2XUNKATT3WYWNDI7G5G43QSXILHANCNFSM4IYD6K4Q .

-- I love those that desire the impossible - Goethe.

Konrad Kording @kordinglab, UPenn, Richards 404b cell:773 - liberat (5423728) office:(215) 898-5826 kordinglab.com

jasmainak commented 4 years ago

@AnchorBlues @BeibinLi @DreHar @the872 @themantalope @marquesVF @evadyer @tommyod @ravigarg27 @timshell

I have not heard from you in this thread. We do need you to confirm your name and affiliation if you want to be included as an author.

daniel-acuna commented 4 years ago

@jasmainak I am in a conference with one of the assoc. editors in chief of JOSS. perhaps there is some way of receiving feedback? maybe email me to deacuna@syr.edu

pavanramkumar commented 4 years ago

Thanks @daniel-acuna for offering to do this. The JOSS review process seems super straightforward and transparent: one of the reasons why we chose the venue. @jasmainak will be in touch!

themantalope commented 4 years ago

Yes, my name and affiliation is correct as written. Thanks!

On Sat, Nov 9, 2019 at 2:05 PM Pavan Ramkumar notifications@github.com wrote:

Thanks @daniel-acuna https://github.com/daniel-acuna for offering to do this. The JOSS review process seems super straightforward and transparent: one of the reasons why we chose the venue. @jasmainak https://github.com/jasmainak will be in touch!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/glm-tools/pyglmnet/pull/313?email_source=notifications&email_token=ABZ7OB2OPVBR523FQMC72ZTQS4JYLA5CNFSM4IYD6K42YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEDUOAUI#issuecomment-552132689, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABZ7OB6ROX5OXD54U4NGBQTQS4JYLANCNFSM4IYD6K4Q .

timshell commented 4 years ago

Name and affiliation are good for me. Thanks for including me! Not sure why I'm not showing up in the contributor graph. Perhaps the e-mail I used for that was my undergraduate university e-mail (which I don't have access to anymore)

jasmainak commented 4 years ago

@timshell okay then we have to do this by updating the .mailmap. Are you up to make a PR to fix it? If not, just tell me your new email address and I'll do it.

BeibinLi commented 4 years ago

Thanks! I would like to be part of this. My name and affiliation look all right.

tommyod commented 4 years ago

Verifying that my name and affiliation is correct as written. Thanks!

ravigarg27 commented 4 years ago

Thank you for including me. My name and affiliation are right.

AnchorBlues commented 4 years ago

Thanks for including me for the paper.

I verified that my name and affiliation are correct.

the872 commented 4 years ago

Thanks for the update, will be happy to represent the Loyola University Chicago commits.

pavanramkumar commented 4 years ago

@the872 thanks! any idea how we can reach out to @marquesVF -- pending confirmation needed from him

marquesVF commented 4 years ago

Hi guys!

Thanks for reaching me. That's correct. I was at Loyola.

I would only suggest to use my full name as Vinicius de F. Marques Thanks and great work!

@pavanramkumar