search

Philipp-Neubauer / FirstAssessment

Analysing the time to first assessment of fish stocks
0 stars 0 forks source link

Compiling knitr draft doc #2

Closed Philipp-Neubauer closed 7 years ago

Philipp-Neubauer commented 7 years ago

Just in case there's more problems along these lines:

Rstudio is giving me the error:

Error in match.arg(options$results, c("verbatim", "tex", "hide")) : 
  'arg' should be one of "verbatim", "tex", "hide"

I've gone through and changed 'hide' to "hide", and 'asis' to "verbatim". However, it still gives me that error.

Philipp-Neubauer commented 7 years ago

see if this works:

  1. Go to Global Options in Rstudio
  2. Select 'Sweave' on the lefthand menu
  3. First option on the right - weave .rnw with: select "knitr". It might complain about not having knitr installed, so install the package if you haven't already.

For the things you changed, you probably need to change "verbatim" back to 'asis'; the former lets you print R code without latex evaluation (e.g., if you just want to print code without having it evaluated in R - e.g., if you're printing a BUGS model). Asis means it will get used "as is", so evaluated by R, then weaved with knitr and type-set by LaTex.

James-Thorson commented 7 years ago

OK, a new error: I don't see "First year of assessment/model.outTue Oct 25 15:07:50 2016.rda" in our dropbox folder?

On Tue, Oct 25, 2016 at 12:36 PM, Philipp Neubauer <notifications@github.com

wrote:

see if this works:

  1. Go to Global Options in Rstudio
  2. Select 'Sweave' on the lefthand menu
  3. First option on the right - weave .rnw with: select "knitr". It might complain about not having knitr installed, so install the package if you haven't already.

For the things you changed, you probably need to change "verbatim" back to 'asis'; the former lets you print R code without latex evaluation (e.g., if you just want to print code without having it evaluated in R - e.g., if you're printing a BUGS model). Asis means it will get used "as is", so evaluated by R, then weaved with knitr and type-set by LaTex.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#issuecomment-256152291, or mute the thread https://github.com/notifications/unsubscribe-auth/AHnqTeshsv31OKjVFVwHFdSkXdgQCvQbks5q3loigaJpZM4KgZNA .

Philipp-Neubauer commented 7 years ago

Suggests the file is synced at my end...strange..can others see the file?

James-Thorson commented 7 years ago

OK, I think it won't download to a windows machine because ":" isn't allowed in windows paths. When I manually downloaded the name auto-changed to "model.outTue Oct 25 15-07-50 2016.rda". So perhaps change that naming convention on your end and I think it should sync from now on?

On Tue, Oct 25, 2016 at 3:12 PM, James Thorson jamest.esq@gmail.com wrote:

Yeah, that's a mystery to me. I can:

  1. see it in my dropbox account
  2. confirm that I have free space
  3. confirm that DB is updating other files today (most recent, 45 min ago)

For now, I'll just pull it manually from dropbox.

jim

On Tue, Oct 25, 2016 at 2:59 PM, Philipp Neubauer < notifications@github.com> wrote:

Suggests the file is synced at my end...strange..can others see the file?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#issuecomment-256189610, or mute the thread https://github.com/notifications/unsubscribe-auth/AHnqTcqBO8z_jGMLzRO53O9dZZ1DNAiKks5q3nvfgaJpZM4KgZNA .

James-Thorson commented 7 years ago

OK, its now working and I'm uploading the converted PDF, so I'm closing this issue

James-Thorson commented 7 years ago

Yeah, that's a mystery to me. I can:

  1. see it in my dropbox account
  2. confirm that I have free space
  3. confirm that DB is updating other files today (most recent, 45 min ago)

For now, I'll just pull it manually from dropbox.

jim

On Tue, Oct 25, 2016 at 2:59 PM, Philipp Neubauer notifications@github.com wrote:

Suggests the file is synced at my end...strange..can others see the file?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#issuecomment-256189610, or mute the thread https://github.com/notifications/unsubscribe-auth/AHnqTcqBO8z_jGMLzRO53O9dZZ1DNAiKks5q3nvfgaJpZM4KgZNA .

Philipp-Neubauer commented 7 years ago

Ok, done - changed name to Weibull_model_output.rda

On Wed, Oct 26, 2016 at 11:17 AM, Jim Thorson notifications@github.com wrote:

OK, I think it won't download to a windows machine because ":" isn't allowed in windows paths. When I manually downloaded the name auto-changed to "model.outTue Oct 25 15-07-50 2016.rda". So perhaps change that naming convention on your end and I think it should sync from now on?

On Tue, Oct 25, 2016 at 3:12 PM, James Thorson jamest.esq@gmail.com wrote:

Yeah, that's a mystery to me. I can:

  1. see it in my dropbox account
  2. confirm that I have free space
  3. confirm that DB is updating other files today (most recent, 45 min ago)

For now, I'll just pull it manually from dropbox.

jim

On Tue, Oct 25, 2016 at 2:59 PM, Philipp Neubauer < notifications@github.com> wrote:

Suggests the file is synced at my end...strange..can others see the file?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/ issues/2#issuecomment-256189610, or mute the thread https://github.com/notifications/unsubscribe-auth/AHnqTcqBO8z_ jGMLzRO53O9dZZ1DNAiKks5q3nvfgaJpZM4KgZNA

.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#issuecomment-256193855, or mute the thread https://github.com/notifications/unsubscribe-auth/ACJDCxIwCSCu6rhC7AI_56QmB87NUbLJks5q3n_qgaJpZM4KgZNA .

Phil

Philipp-Neubauer commented 7 years ago

Did you check if you can see Weibull_model_out.rda in your dropbox? I renamed the model output to that, in case the filename is the problem. Is that also not updating?

On Thu, Oct 27, 2016 at 4:39 AM, Jim Thorson notifications@github.com wrote:

Yeah, that's a mystery to me. I can:

  1. see it in my dropbox account
  2. confirm that I have free space
  3. confirm that DB is updating other files today (most recent, 45 min ago)

For now, I'll just pull it manually from dropbox.

jim

On Tue, Oct 25, 2016 at 2:59 PM, Philipp Neubauer < notifications@github.com> wrote:

Suggests the file is synced at my end...strange..can others see the file?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/ issues/2#issuecomment-256189610, or mute the thread https://github.com/notifications/unsubscribe-auth/AHnqTcqBO8z_ jGMLzRO53O9dZZ1DNAiKks5q3nvfgaJpZM4KgZNA .

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#issuecomment-256386175, or mute the thread https://github.com/notifications/unsubscribe-auth/ACJDC4ZWUd_saAU2XugbisVUh-JCPk1Sks5q33QvgaJpZM4KgZNA .

Phil

James-Thorson commented 7 years ago

Its there! I'm sorry I didn't update this thread, problem fixed

James-Thorson commented 7 years ago

closing

James-Thorson commented 7 years ago

I'm getting the same error when compiling as referenced above

> #load(file.path(DB,"First year of assessment/Weibull_model_output.rda"),v=T)
> load('model.outSun Nov 13 22:25:59 2016.rda',v=T)
Error in readChar(con, 5L, useBytes = TRUE) : cannot open the connection
In addition: Warning message:
In readChar(con, 5L, useBytes = TRUE) :
  cannot open compressed file 'model.outSun Nov 13 22:25:59 2016.rda', probable reason 'Invalid argument'

One obvious problem is that windows doesn't allow colons in file paths, so the name model.outSun Nov 13 22:25:59 2016.rda won't sync from dropbox on a windows machine. However, for some reason, I also don't see a file with that name on dropbox through the online GUI, so maybe the problem is deeper. Anyway, could you change the file name used for loading results?

Philipp-Neubauer commented 7 years ago

Sorry, had changed the path temporarily to the local model output before overwriting the dropbox file. If you pull now, it should work.

On Wed, Nov 16, 2016 at 12:26 PM, Jim Thorson notifications@github.com wrote:

Reopened #2 https://github.com/Philipp-Neubauer/FirstAssessment/issues/2 .

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#event-860254098, or mute the thread https://github.com/notifications/unsubscribe-auth/ACJDC1U1Djx48nuFl_Uzt__hdIc04rjRks5q-j-bgaJpZM4KgZNA .

Phil

James-Thorson commented 7 years ago

OK, thanks!

And as always sorry, but I'm now getting an error at line

> habitats <- get_coef_chains(model.out = a.out, coef.names = 'habitat\\[[0-9]*\\]', var.names = levels(as.factor(year.table$habitat_FB.SLB)))
Error in `colnames<-`(`*tmp*`, value = c("bathydemersal", "benthic", "benthopelagic",  : 
  length of 'dimnames' [2] not equal to array extent

and it appears to arise because it has the wrong number of levels relative to model output, such that this works

> habitats <- get_coef_chains(model.out = a.out, coef.names = 'habitat\\[[0-9]*\\]', var.names = levels(as.factor(year.table$habitat_FB.SLB))[1:6])

any quick guesses at what might cause this?

Philipp-Neubauer commented 7 years ago

Yes, you need to set cache=F for the first run with new model output - the old data in the chunk cache does not have the updated habitat definitions. After the first run, you can set cache=T to compile faster...

On Wed, Nov 16, 2016 at 1:21 PM, Jim Thorson notifications@github.com wrote:

OK, thanks!

And as always sorry, but I'm now getting an error at line

habitats <- get_coefchains(model.out = a.out, coef.names = 'habitat[[0-9]]', var.names = levels(as.factor(year.table$habitat_FB.SLB)))Error in colnames<-(_tmp*, value = c("bathydemersal", "benthic", "benthopelagic", : length of 'dimnames' [2] not equal to array extent

and it appears to arise because it has the wrong number of levels relative to model output, such that this works

habitats <- get_coef_chains(model.out = a.out, coef.names = 'habitat[[0-9]*]', var.names = levels(as.factor(year.table$habitat_FB.SLB))[1:6])

any quick guesses at what might cause this?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#issuecomment-260814134, or mute the thread https://github.com/notifications/unsubscribe-auth/ACJDCxO2wDY_vK8IhBPqxp9_X63MNZNEks5q-kySgaJpZM4KgZNA .

Phil

James-Thorson commented 7 years ago

Hmm, I tried two things:

  1. setting cache=FALSE in line 38
  2. Deleting the repo (to try and delete any locally saved stuff from previous runs), recloning, and doing 1 again

and neither worked. Any other suggestions?

I'll take a swing at the bibliography stuff tomorrow

Philipp-Neubauer commented 7 years ago

hmm - normally deleting the /cache directory should be enough. I had the same issue but that did the trick.

Did your dropbox sync the Weibull_model_out.rda file?

On Wed, Nov 16, 2016 at 1:43 PM, Jim Thorson notifications@github.com wrote:

Hmm, I tried two things:

  1. setting cache=FALSE in line 38
  2. Deleting the repo (to try and delete any locally saved stuff from previous runs), recloning, and doing 1 again

and neither worked. Any other suggestions?

I'll take a swing at the bibliography stuff tomorrow

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#issuecomment-260817872, or mute the thread https://github.com/notifications/unsubscribe-auth/ACJDC23eV-gh8joJ9mDtZSbTNGn_UGH8ks5q-lGngaJpZM4KgZNA .

Phil

James-Thorson commented 7 years ago

Hmm, I just checked the Dropbox online GUI and it's says the latest change to that file was 2 weeks ago, what date should the latest version show?

Philipp-Neubauer commented 7 years ago

oh, maybe mine hasn't synced!

On Wed, Nov 16, 2016 at 2:11 PM, Jim Thorson notifications@github.com wrote:

Hmm, I just checked the Dropbox online GUI and it's says the latest change to that file was 2 weeks ago, what date should the latest version show?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#issuecomment-260822858, or mute the thread https://github.com/notifications/unsubscribe-auth/ACJDC_qbK1db9f0G7BFy2-RP5gylY_2nks5q-lg2gaJpZM4KgZNA .

Phil

James-Thorson commented 7 years ago

Err actually my Dropbox only shows Weibell_model_output.rda and not Weibell_model_out.rda, could you confirm the file name and check that it's uploading on the Dropbox online GUI on your side?

Philipp-Neubauer commented 7 years ago

Ok, it should be the _output. IT's acutaly syncing now. Sorry about that, for some reason the syncing at my end had stalled. Should be there shortly.

On Wed, Nov 16, 2016 at 2:13 PM, Jim Thorson notifications@github.com wrote:

Err actually my Dropbox only shows Weibell_model_output.rda and not Weibell_model_out.rda, could you confirm the file name and check that it's uploading on the Dropbox online GUI on your side?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#issuecomment-260823217, or mute the thread https://github.com/notifications/unsubscribe-auth/ACJDC-d-WSUISdePaN6t21sBPckDZuv3ks5q-lijgaJpZM4KgZNA .

Phil

Philipp-Neubauer commented 7 years ago

1 min ... slow upload here in middle of nowhere NZ...

On Wed, Nov 16, 2016 at 2:18 PM, Philipp Neubauer neubauer.phil@gmail.com wrote:

Ok, it should be the _output. IT's acutaly syncing now. Sorry about that, for some reason the syncing at my end had stalled. Should be there shortly.

On Wed, Nov 16, 2016 at 2:13 PM, Jim Thorson notifications@github.com wrote:

Err actually my Dropbox only shows Weibell_model_output.rda and not Weibell_model_out.rda, could you confirm the file name and check that it's uploading on the Dropbox online GUI on your side?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#issuecomment-260823217, or mute the thread https://github.com/notifications/unsubscribe-auth/ACJDC-d-WSUISdePaN6t21sBPckDZuv3ks5q-lijgaJpZM4KgZNA .

Phil

Phil

mcmelnychuk commented 7 years ago

1 min isn't bad for an earthquake hotspot.

Mike

On 2016-11-15 5:24 PM, Philipp Neubauer wrote:

1 min ... slow upload here in middle of nowhere NZ...

On Wed, Nov 16, 2016 at 2:18 PM, Philipp Neubauer neubauer.phil@gmail.com wrote:

Ok, it should be the _output. IT's acutaly syncing now. Sorry about that, for some reason the syncing at my end had stalled. Should be there shortly.

On Wed, Nov 16, 2016 at 2:13 PM, Jim Thorson notifications@github.com wrote:

Err actually my Dropbox only shows Weibell_model_output.rda and not Weibell_model_out.rda, could you confirm the file name and check that it's uploading on the Dropbox online GUI on your side?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub

https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#issuecomment-260823217, or mute the thread

https://github.com/notifications/unsubscribe-auth/ACJDC-d-WSUISdePaN6t21sBPckDZuv3ks5q-lijgaJpZM4KgZNA .

Phil

Phil

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Philipp-Neubauer/FirstAssessment/issues/2#issuecomment-260825151, or mute the thread https://github.com/notifications/unsubscribe-auth/AV_oVepljYdKzsbVvnVV1osUPVTRt55sks5q-ls7gaJpZM4KgZNA.

James-Thorson commented 7 years ago

OK, working great! And please send my regards to your friends and family -- seems like a hard time to be in Christchurch