Eddy identification on multiple files

[Mrvy29]

Hi there:)

Recently found your product on Aviso + any differences from the identification here?
I have multiple files (for each day ) to run the eddy identification on and cannot seem to understand how to properly fit the date parameter. when running individual files this is date = datetime( 2016, 11, 02) for example.
I am trying a loop such as (parameter in question between "**"
for f in files:
g.eddy_identification("adt", "u", "v", date, 0.002, shape_error=55)
in my loop the list of identifications has the same date...
I tried to find a solution on my own but maybe you can share your solution in the list of examples provided?

That would be extremely helpful !
Thank you !

[Mrvy29]

Hi , I have also asked the question on stack overflow: Question

[AntSimi]

In the identification loop did you take a look at

print(a.time)
# Where a is anticyclonic given by eddy_identification

[Mrvy29]

Yes of course the problem is all identification has the same date coming from the last file of the loop.. can you maybe share a pipeline for identification on multiple files? I tried a multidate .nc file but the function doesn’t allow for a list only a single date time object. Thanks so much for the support.

On Mon, 28 Nov 2022 at 18:51 Antoine Delepoulle ***@***.***> wrote: In the identification loop did you take a look at print(a.time)# Where a is anticyclonic given by eddy_identification — Reply to this email directly, view it on GitHub <#178 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A2ZPD6UBKW6MVJO22ELEU6TWKTPKXANCNFSM6AAAAAASHTN3DQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>

-- [image: Seafloor Scene Icon]

[AntSimi]

Could you share the full script that you use?

[Mrvy29]

Thanks I understand, would you be able to share an example of what I did wrong and how to do it right?

[AntSimi]

Your script is ok, just this two lines must be in your loop:

a, c = g.eddy_identification("adt", "u", "v", date , 0.002, shape_error=55)
print(a.time)

[Mrvy29]

Oh but of course they are, it just didn't ident properly while pasting the code.

[AntSimi]

But in your display only one ouput are displaying.
Could you return output of this for each step of your loop:

print(f, a.time)

[Mrvy29]

OK it seems to have solved it 👍 thanks a lot , I appreciate it

for f in files:
    g = RegularGridDataset(f,"longitude","latitude")
    g.add_uv("adt")
    g.bessel_high_filter("adt", 500)
    date = datetime(int(os.path.splitext(f)[0][24:-13]),int(os.path.splitext(f)[0][28:-11]),
                    int(os.path.splitext(f)[0][30:-9]))
    a, c = g.eddy_identification("adt", "u", "v", date , 0.002, shape_error=55)
    a_name = 'a'+date.strftime('%Y_%m_%d')+'.nc'
    c_name = 'c'+date.strftime('%Y_%m_%d')+'.nc'
    a.write_file(filename = a_name)
    c.write_file(filename = c_name)