MrMarc

3123 Reputation

16 Badges

12 years, 315 days

MaplePrimes Activity


These are replies submitted by MrMarc

@marc005 It turns out that finding a cloude storage provider that does not use https is quite a challenge. Most of the cloud providers also lacks the option to create direct download links and not to mentions sky drives expiring urls which is a big joke! You post a link on your website and a few days later the link has expired ha ha

@Markiyan Hirnyk he he I could not replicate that unfortunatly. I still got timed out :-(

@Markiyan Hirnyk he he I could not replicate that unfortunatly. I still got timed out :-(

@acer me and Mr Burns (The Simpsons) thinks the datatable component
was an exellent idea :-) thanx

@Christopher2222 can you please check if the limitation exist for Maple 17?
If it dose not exist it might be a reason to upgrade. I havnt upgraded because
Maple 17 does not have any new things (compared to Maple 16) that I value
except for potential removal of this 10001 max matrix size limitation,

@acer thanx for explaning that. Then I dont have to bang my head against the wall.
If you type:

Matrix(1 .. 10001, 1 .. 4, 5);  # everything is ok
Matrix(1 .. 10002, 1 .. 4, 5);  # error message

However, I must say that the error message

"Unable to display structured data".

is a bit obscure. It only tells me things are not working which I all ready know.
It does not tell me why. It would have been easy to put:

"The largest matrix Maple can display is 10001*4"

also 10001 seems kind of small?!  Yeahhh I have never been that impressed by the
Matrix browser. Last time a checked it displayed a column vector if you had a column
row and vise versa. I think that has been fixed now but still.


The error seems to come from this section of the code:

sscanf(data, cat("%{", nrows, ",", ncols, "}lvm"));

Error, (in sscanf) input and format strings expected

The error message comes from row 10002

z := GetCSV("http://www.quandl.com/api/v1/datasets/USER_18Q/18R.csv?&auth_token=BXckVSG3twWkV272c9Q4&trim_start=1968-04-01&trim_end=2013-04-05&collapse=daily&sort_order=desc");
z[1 .. 10001, 1 .. 4];   # everything is ok
z[1 .. 10002, 1 .. 4];   # error message

however I could not find anything special with that row. It appears Maple can handle
the missing observations however this "ghost bug" is apperently to much for
Maple to handle.....hummm strange.

If you run the below in Maple:

GetCSV("http://www.quandl.com/api/v1/datasets/USER_18Q/18R.csv?&auth_token=BXckVSG3twWkV272c9Q4&trim_start=1968-04-01&trim_end=2013-04-05&collapse=daily&sort_order=desc")

you get a matrix with the right dimensions (11746*4) however when you double click on it
you get the same error message as previously "Unable to display structured data".

The original csv file have some missing observations ie 23.4,78,,7
Anything with ,, indicate missing data and I am not sure Maple handles that correctly :-(

When I get on my pc I will play around with it. It looks promising :-)

@Axel Vogt well it is a cvs (comma separated value) file but it is somewhat standard.

Maple should be able to load it pretty straight forward with string-tools the trick is just

to figure out how and to separate numerical values from column titles ie string.

@herclau 

thanx for your input. Yes you could indeed store the file locally and use Maples

exel tools to import the data. However, some of the elegance is lost because cvs file the 

url points to contain the constantly updated version. Each day you manually have to download the

file and store it locally. It would be more convenient to get the data directly from the cloud Into Maple.

 

I just thought it was strange that:

restart:
with(Statistics):

P := Vector([1, 1, 2, 3, 3]):
X := RandomVariable(EmpiricalDistribution(P)):

DensityPlot(X);


worked but:


restart:
with(Statistics):

P := Sample(RandomVariable(Normal(10, 10)), 500):
X := RandomVariable(EmpiricalDistribution(P)):

DensityPlot(X);

did not work. This is also the reason why it is frustrating to work with
Maple sometime because it is not consistent ie sometime very ad hoc.

Another thing I reacted to was that you can use a legend option with the plot
command but not with KernelDensityPlot. Again a bit ad hoc.

Maple are making great progress but I am still missing some things:

- ability to embed pdf-files in maple documents as with video

- return procedureds (first difference, log, percentage) for the finance package
that takes a price matrix as an imput.

- A simple linear regression procedure that takes a data matrix as input and return
the standard regression output as in excel ie R2,t-value,p-values etc

Thank you for pointing that out to me :-) Maybe I will. I guess you dont have
to be Einstein to figure out who is annoying me...

4 5 6 7 8 9 10 Last Page 6 of 33