search for: analyse

...at google.com> > wrote: > >> >> >> On Tue, Jul 12, 2016 at 10:57 PM, Chandler Carruth <chandlerc at gmail.com> >> wrote: >> >>> Yea, this is a nasty problem. >>> >>> One important thing to understand is that this is specific to analyses >>> which hold references to other analyses. While this isn't unheard of, it >>> isn't as common as it could be. Still, definitely something we need to >>> address. >>> >> >> We can call this type of dependencies (holding references) >>...

...gt; >>>> On Tue, Jul 12, 2016 at 10:57 PM, Chandler Carruth <chandlerc at gmail.com >>>> > wrote: >>>> >>>>> Yea, this is a nasty problem. >>>>> >>>>> One important thing to understand is that this is specific to analyses >>>>> which hold references to other analyses. While this isn't unheard of, it >>>>> isn't as common as it could be. Still, definitely something we need to >>>>> address. >>>>> >>>> >>>> We can call this ty...

...gt; > chandlerc at gmail.com > wrote: > > > > > > > > > > > Yea, this is a nasty problem. > > > > > > > > > > > > > > > One important thing to understand is that this is specific to > > > > > analyses which hold references to other analyses. While this > > > > > isn't > > > > > unheard of, it isn't as common as it could be. Still, > > > > > definitely > > > > > something we need to address. > > > > > > >...

Dear all, I'm using R 2.8.1 under Vista. I programmed a Simulation with the code enclosed at the end of the eMail. After the simulation I want to analyse the columns of the single simulation-runs, i.e. e.g. Simulation[[1]][,1] sth. like that but I cannot address these columns... Can anybody please help? Best, Thomas ############################ CODE ############################ analyse<-read.csv2("C:\\Users\\Thomas\\Desktop\\PCA_Kohle_2...

...th < >>>>>> chandlerc at gmail.com> wrote: >>>>>> >>>>>>> Yea, this is a nasty problem. >>>>>>> >>>>>>> One important thing to understand is that this is specific to >>>>>>> analyses which hold references to other analyses. While this isn't unheard >>>>>>> of, it isn't as common as it could be. Still, definitely something we need >>>>>>> to address. >>>>>>> >>>>>> >>>>>>...

Hello, I've been attempting to make a generic method that dispatches on the first argument, which can be either an S3 or an S4 class. This is as far as I've gotten. Any suggestions about what to try next ? library(aroma.affymetrix) library(GenomicRanges) setGeneric("analyse", function(x, y, ...) standardGeneric("analyse")) setMethodS3("analyse", "AffymetrixCelSet", function(x, y, z, ...) { x; UseMethod("analyse") } ) setGeneric("analyse") setMethod("analyse", "GRangesList", function(x, y,...

On Tue, Jul 12, 2016 at 10:57 PM, Chandler Carruth <chandlerc at gmail.com> wrote: > Yea, this is a nasty problem. > > One important thing to understand is that this is specific to analyses > which hold references to other analyses. While this isn't unheard of, it > isn't as common as it could be. Still, definitely something we need to > address. > We can call this type of dependencies (holding references) hard-dependency. The soft dependency refers to the case w...

On Fri, Jul 15, 2016 at 8:39 PM, Sean Silva <chisophugis at gmail.com> wrote: > It looks like there is really no sane fix within the current > infrastructure. I've had to essentially trigger invalidation (except in the > PreservedAnalyses::all() case) in the function pass manager and function to > loop adapters. > invalidation of *everything* I mean. -- Sean Silva > > So we basically need to get the analysis manager dependency tracking fixed. > > Davide and I will get measurements on the resident set impact of...

...; > > > > > > > > > > > > > > > > > > > One important thing to understand is that this is > > > > > > > > > specific > > > > > > > > > to > > > > > > > > > analyses which hold references to other analyses. > > > > > > > > > While > > > > > > > > > this > > > > > > > > > isn't > > > > > > > > > unheard of, it isn't as common as it could be. Still...

...similar in the new PM and old PM (the main difference is that the notion of "adapters" is split out in the new PM). The analysis handling seems to be what makes the old PM so difficult to understand (e.g. it is the cause of the multiple inheritance in the implementation). Trying to unify analyses and transformations (and some questionable (in hindsight) implementation decisions) seems to be the main "problem" with the design of the old PM AFAICT (there are other issues, but they are more "nice to have"). IMO it is an anti-pattern to think of analyses as "passes&qu...

...> > > > > > > > > > > > > > One important thing to understand is that this is > > > > > > > > > > > specific > > > > > > > > > > > to > > > > > > > > > > > analyses which hold references to other analyses. > > > > > > > > > > > While > > > > > > > > > > > this > > > > > > > > > > > isn't > > > > > > > > > > > unheard of, it i...

...would become burdensome. However, my preference would be to instead handle #2 by making result APIs accept direct handles to the analysis results they rely on. The reason for preferring this approach is because I think it makes the relationship between these things much more clear to users of the analyses. I think the most annoying of these to handle are aggregation-style analyses results like AAResults. There, I think it might make more sense to handle them along the lines of data-structure-dependencies. I don't think we have so many of those that this would be a significant burden IMO. On T...

...> The difficult part is what guarantees we provide about things being > stale or not and how we invalidate when IRUnit's are deleted or > created. > For example, suppose a function pass DCE's a call which causes an SCC > Foo of 3 functions to no longer be an SCC. When/how do analyses > cached on Foo get invalidated? And is it valid to query them? One of > the expected use cases (I'm told) for CGSCC passes is to propagate > function-attribute like things, so these are being potentially > queried by that same function pass. I don't understand this example. Ju...

...m/IR/PassManager.h)). > The main changes to the analysis manager will be: > - type-erasing the IRUnit > - the elements of the AnalysisResultListMapT will need to keep track > of any dependents > - the analysis manager will need to update those dependencies as it > is re-entered by analyses getting results of other analyses > - the analysis manager will need to walk the dependencies to do > transitive invalidation > I think the most robust approach is for analysis dependencies to be > implicitly constructed by the analysis manager via tracking > entry/exit from get{,C...

...at google.com> > wrote: > >> >> >> On Tue, Jul 12, 2016 at 10:57 PM, Chandler Carruth <chandlerc at gmail.com> >> wrote: >> >>> Yea, this is a nasty problem. >>> >>> One important thing to understand is that this is specific to analyses >>> which hold references to other analyses. While this isn't unheard of, it >>> isn't as common as it could be. Still, definitely something we need to >>> address. >>> >> >> We can call this type of dependencies (holding references) >>...

...e(fit.imp)) print(imputed) # now, how to plot a survival curve with the pooled results ? ########## begin code from linked thread above coxph.mids <- function (formula, data, ...) { call <- match.call() if (!is.mids(data)) stop("The data must have class mids") analyses <- as.list(1:data$m) for (i in 1:data$m) { data.i <- complete(data, i) analyses[[i]] <- coxph(formula, data = data.i, ...) } object <- list(call = call, call1 = data$call, nmis = data$nmis, analyses = analyses) return(obje...

...n and make no changes (and hence preserve all). 3. then GVN had to run and query BasicAA (presumably LICM or MergedLoadStoreMotion didn't make a query to BasicAA, or that query ended up not accessing the dangling TargetLibraryInfo). How should we solve this? I see two potential solutions: 1. Analyses must somehow list the analyses they depend on (either by overriding "invalidate" to make sure that they invalidate them, or something "declarative" that would allow the AnalysisManager to walk the transitive dependencies). 2. The AnalysisManager must do a somewhat complicated d...

...ates a module analysis, the module analysis won't be invalidated until after the ModuleToFunctionPassAdapter returns. So we can end up running an arbitrary number of function transformations that query that stale module analysis. This is (I think?) mostly benign given our current set of module analyses (more info in the log; search for "We may be getting by simply because we seem to not have many module analyses"). But if we invalidate more "correctly" (after every transformation on any contained IRUnit), there's a potential quadratic compile time issue lurking here once...

...vel so that they can run a whole-program analysis which cannot be done at the compiler that works for one compilation unit at a time. I'd like to start a discussion as to what we can do with it and how to make it possible. I'm also sharing my idea about how to make it possible. *Dependency analyses* First, let me start with a few examples of analyses I'm heard of or thinking about. Dependencies between object files can be represented as a graph where vertices are input sections and edges are symbols and relocations. Analyses would work on the dependency graph. Examples of analyses includ...

.... However, my preference would > be to instead handle #2 by making result APIs accept direct handles to the > analysis results they rely on. > > The reason for preferring this approach is because I think it makes the > relationship between these things much more clear to users of the analyses. > > +1 to this. Having the query dependencies explicit at the call site would > generally make the code much easier to understand and thus much more likely > to be correct. I recently ran across an issue in LVI under the old pass > manager that looks highly suspicious, but becaus...