+- Python Forum (https://python-forum.io)
+-- Forum: Python Coding (https://python-forum.io/forum-7.html)
+--- Forum: General Coding Help (https://python-forum.io/forum-8.html)
+--- Thread: Removing rows from array (/thread-29731.html)
Removing rows from array - claw91 - Sep-17-2020
Hello, I want to replicate this Matlab code into Python
FILE(1:985,:) = [ ];
FILE is an nx2 array. n si quite large but it's not important. If I understand correctly, the code should remove from the FILE array all rows from first to 985th.
This is what I've come up with in Python
FILE = np.delete(FILE, FILE[0:985,...].astype('int'), axis = 0)However it doesn't seem to work as I want, it removes only 19 rows from the array. Why is that?
Here you can find the FILE array: https://pastebin.com/ it's a 40.001 x 2
The resulting FILE at the end should be 39.016 x 2 but I get 39.982 x 2
RE: Removing rows from array - scidam - Sep-17-2020
If you want to drop first 985 rows, just do:
FILE = FILE[985:, ...]If you want to drop first 985 rows and convert the array to integer type do:
FILE = FILE[985:, ...].astype(int) # maybe astype(np.int64), astype(np.int16) choose appropriate one.If you still want to use delete, you can do the same as follows:
FILE = np.delete(FILE, range(985), axis=0)Note, that the second argument is an array of row numbers to delete.
Finally, it is not pep8-compliant to use capitalized letters for variable names.
RE: Removing rows from array - claw91 - Sep-18-2020
Thank you.
Can you please explain why my second argument
FILE[0:985,...]is not the same as yours
range(985)?
RE: Removing rows from array - scidam - Sep-18-2020
(Sep-18-2020, 01:50 PM)claw91 Wrote: Can you please explain why my second argument
Python Code: (Double-click to select all)
1
FILE[0:985,...]
is not the same as yours
Python Code: (Double-click to select all)
1
range(985)
?
You passed
FILE[0:985,...].astype('int'), that are values of the first column converted to ints, but you need to pass row numbers which you want to be deleted, range(985) stands (approx.) for [0, 1, ..., 984] - these are row numbers.