Jupyter has completely changed my workflow in Python and Julia. I've got my
notebooks numbered 01_Planning_Invasion, 02_Moving_the_Troops, etc. I write
most of my code inside the notebook interactively, then if it's good
enough, it gets promoted to a .jl file. It takes about a week to complete
SageMathCloud doesn't seem to work that well on an iPad, unfortunately
For lots of small, possibly unrelated and possibly hard to classify
functions, I'll typically put them in a file called utils.jl (utilities),
misc.jl (miscellaneous) or common.jl (methods shared across modules). It's
all roughly the same to me.
On Saturday, May 14, 2016, Sisyphuss
But what does "Utilities" mean?
I plan to put all my small functions frequently used by all other pieces of
code into a single file.
Is it the right file name for it?
On Friday, May 13, 2016 at 7:11:45 PM UTC+2, Kristoffer Carlsson wrote:
>
> Utilities.
The error you ran into is because of this PR:
https://github.com/JuliaLang/julia/pull/16235
Gallium was updated for this, but those updates haven't made their way into
a tagged release yet, so if you want to run on a Julia more recent than the
above commit, as Tim pointed out, you'll need to get
Julia & Gallium are changing so rapidly, and work together so closely, that if
you're wanting to use Gallium, you probably need to update both daily (or more
frequently).
Best,
--Tim
On Saturday, May 14, 2016 10:45:28 AM David van Leeuwen wrote:
> Hello,
>
> On Saturday, May 14, 2016 at
Terrific. Thanks.
On Friday, May 13, 2016 at 7:48:54 PM UTC-4, Tom Breloff wrote:
>
> Check out the Formatting package, as well as the spin-off
> https://github.com/ScottPJones/StringUtils.jl
>
> On Friday, May 13, 2016, Salman Haider
> wrote:
>
>> Hey guys,
>>
>> I have
Hello,
On Saturday, May 14, 2016 at 10:57:18 AM UTC+2, Tim Holy wrote:
>
> On Saturday, May 14, 2016 01:05:39 AM David van Leeuwen wrote:
> > Which version of Julia should I use? My 0.4.5 complains on install
> "Gallium
> > can't be installed because it has no versions that support 0.4.5 of
I'm actually one who is trying to increase overall things like integer
parsing. Currently in the CSV.jl package, we have some beta-mode fast
parsing functions. I generated a file with:
open("test_integers","w") do f
for i = 1:7_068_650
println(f, rand(Int16))
end
end
And reading
On version 0.5, after warm-up. (Notice that I use collect() instead of [],
since the latter doesn't work like that anymore)
function test_broad(N)
@time for _ in 1:N broadcast( (x...) -> +(x...), collect(1:1000),
collect(1001:2000) ) end
@time for _ in 1:N broadcast( (x...) -> +(x...),
On Sat, May 14, 2016 at 8:31 AM, Ford Ox wrote:
> type Tokenizer
> tokens::Array{ASCIIString, 1}
> index::Int
> Tokenizer(s::ASCIIString) = new(split(strip(s)), 0)
> end
>
> Julia still runs 11 seconds...
The main cost is not coming from the dynamic dispatch,but
I don't see any "Please help me speed up my code"s or even any question marks
at all. I just see complaints about Julia even after receiving initial
suggestions from busy people (without any "Thank you"s). That may be one reason
you're not receiving more helpful responses. Just a thought...
type Tokenizer
tokens::Array{ASCIIString, 1}
index::Int
Tokenizer(s::ASCIIString) = new(split(strip(s)), 0)
end
Julia still runs 11 seconds...
Dne sobota 14. května 2016 14:08:48 UTC+2 Milan Bouchet-Valat napsal(a):
>
> Le samedi 14 mai 2016 à 05:01 -0700, Ford Ox a écrit :
> >
Le samedi 14 mai 2016 à 05:01 -0700, Ford Ox a écrit :
> Fixed. Julia now takes 11 seconds to finish
> type Tokenizer
> tokens::Array{AbstractString, 1}
> index::Int
> Tokenizer(s::AbstractString) = new(split(strip(s)), 0)
> end
>
> type Buffer
> stream::IOStream
>
Fixed. Julia now takes 11 seconds to finish
type Tokenizer
tokens::Array{AbstractString, 1}
index::Int
Tokenizer(s::AbstractString) = new(split(strip(s)), 0)
end
type Buffer
stream::IOStream
tokenizer::Tokenizer
Buffer(stream) = new(stream, Tokenizer(""))
end
Dne sobota
There's lots of things that could be done in terms of clever optimizations,
but currently this is how it is. If you want C speed, concretely type your
fields. For flexibility, this can sometimes entail having type parameters.
On Sat, May 14, 2016 at 1:52 PM, Ford Ox wrote:
Interesting.
The compiler can't use the same hack it uses for multiple dispatch to
generate multiple versions of types and then use only some of them?
Dne sobota 14. května 2016 13:39:24 UTC+2 Stefan Karpinski napsal(a):
>
> Your types have totally untyped fields – the compiler has to emit very
I have fixed the Buffer constructor
Buffer(stream) = new(stream, Tokenizer(""))
Result is still the same.
Dne sobota 14. května 2016 13:36:30 UTC+2 Ford Ox napsal(a):
>
> I have written exact same code in java and julia for reading integers from
> file.
> Julia code was A LOT slower. (12
Your types have totally untyped fields – the compiler has to emit very
pessimistic code about this. Rule of thumb: locations (fields, collections)
should be as concretely typed as possible; parameters don't need to be.
On Sat, May 14, 2016 at 1:36 PM, Ford Ox wrote:
> I
I have written exact same code in java and julia for reading integers from
file.
Julia code was A LOT slower. (12 seconds vs 1.16 seconds)
import Base.isempty, Base.close
##Tokenizer ##
type Tokenizer
tokens
index
Tokenizer(s::AbstractString) = new(split(strip(s)), 0)
end
Thanks,
Here's my first try restricted to Float64:
function replaceNaN(df)
nrows, ncols = size(df)
for row in 1:nrows
for col in 1:ncols
v = df[row,col]
if typeof(v) == Float64 && isnan(v)
df[row,col] = NA
end
end
end
df
end
Am looking for an ideomatic and
On Friday, May 13, 2016 at 7:48:54 PM UTC-4, Tom Breloff wrote:
>
> Check out the Formatting package, as well as the spin-off
> https://github.com/ScottPJones/StringUtils.jl
Thanks for the tip of the hat, and right back at you!
People should be aware, that some of the best parts of the
Also, in the meantime, you can use something like
function nan2na(df)
for c in eachcol(df)
nans = find(x->isa(x,AbstractFloat) && isnan(x),c[2])
df[nans,c[1]] = NA
end
df
end
nan2na(mlmf)
to convert NaNs to NAs.
Best,
Tamas
On Sat, May 14 2016, Milan Bouchet-Valat
On Saturday, May 14, 2016 01:05:39 AM David van Leeuwen wrote:
> Which version of Julia should I use? My 0.4.5 complains on install "Gallium
> can't be installed because it has no versions that support 0.4.5 of julia",
> in my 0.5.0 there is no package Gallium in Metadata.
You probably need to
Le samedi 14 mai 2016 à 17:49 +1000, Андрей Логунов a écrit :
> using RDatasets, DataFrames
> mlmf = dataset("mlmRev","Gcsemv");
>
> Produces
>
> │ 1 │ "20920" │ "16" │ "M" │ 23.0 │ NaN │
> │ 2 │ "20920" │ "25" │ "F" │ NaN │ 71.2 │
> │ 3 │ "20920" │ "27" │ "F"
Yes, I total forgot the fact that I need to use fftn in matlab my bad.
After changing that and using an array of 2^12-1 I have the same time
between Julia and Matlab.
Also is it the right way to use plan_fft with the flag FFTW.PATIENT?
pfft=plan_fft(Mb;flags=FFTW.PATIENT);
and for one
Unfortunately, backporting Gallium to 0.4.x is unlikely to be possible, so
you have to be on a more recent Julia. It might still be worth stepping
through the code on a 0.5 Julia for this purpose if the code works on both
versions.
On Sat, May 14, 2016 at 10:05 AM, David van Leeuwen <
Thanks,
On Wednesday, May 11, 2016 at 11:21:48 PM UTC+2, Kristoffer Carlsson wrote:
>
> If you get Gallium https://github.com/Keno/Gallium.jl working you can
> simply step through the functions.
Which version of Julia should I use? My 0.4.5 complains on install
"Gallium can't be installed
(1) I use Debug.jl to debug the code, and it works.
https://github.com/toivoh/Debug.jl
(2) I use JuliaDT for code editor, which is similar to PyDev. The current
version is 0.2. The future version would be a very good julia IDE. I
currently use it for my daily work.
using RDatasets, DataFrames
mlmf = dataset("mlmRev","Gcsemv");
Produces
│ 1│ "20920" │ "16"│ "M"│ 23.0│ NaN│
│ 2│ "20920" │ "25"│ "F"│ NaN │ 71.2 │
│ 3│ "20920" │ "27"│ "F"│ 39.0│ 76.8 │
│ 4│ "20920" │ "31"│ "F"│ 36.0│ 87.9
Again: please provide a self-contained example.
On Sat, May 14 2016, Андрей Логунов wrote:
> The misuse of the word is all mine.
> But the problem persists. RDatasets in Win10 produce NaN-values for
> unvailable values (NAs) as compared to Unices.
> So the funcs dropna() and complete_cases()
The misuse of the word is all mine.
But the problem persists. RDatasets in Win10 produce NaN-values for
unvailable values (NAs) as compared to Unices.
So the funcs dropna() and complete_cases() 'do not work' as needed? no
filtering done. As I understand Complete_cases() uses a bitarray. But is
Shouldn't macro use the same principle as functions? That is:
If your macro changes the value of variable you passed as argument (say
@increment x), thus it works similar to a function accepting pointer, it
should be marked with "!" at the end? (@increment! x)
If your macro changes variables
On Sat, May 14 2016, Андрей Логунов wrote:
> To add, fiddling with array comprehensions as per the problem with NaNs
> found a buggy thing.
> The following code does not work:
>
> [x for x in filter(!isnan, convert(Array,dataframe[:fld]))]
This is not a bug. ! does not operate on functions,
34 matches
Mail list logo