Find the line index of the first line in a 2D array thjat is non full of NaN-CodePudding

I have 2D array with plenty of lines fully fed with NaN values. I would like to find where the first real number is present, i.e. the line number where it appears. My simplified array looks like this:

m = np.array([[np.nan, np.nan, np.nan], [np.nan, np.nan, np.nan], [np.nan, 17, np.nan],[np.nan, 18, 25]])

The expected result is 2, the line number where 17 appears! Thank you

CodePudding user response：

You can grab the first row where at least one element is not nan using sum and isnan:

np.argmax((~np.isnan(m)).sum(axis=1) > 0)

Edit

Based on I'mahdi's answer, I realize you can also just use any:

np.argmax((~np.isnan(m)).any(axis=1))

CodePudding user response：

You can use numpy.isnan() and numpy.argmax().

m = np.array([[np.nan, np.nan, np.nan], [np.nan, np.nan, np.nan], [np.nan, 17, np.nan],[np.nan, 18, 25]])
r = np.argmax((~np.isnan(m)).any(1))
print(r)
# 2

If you want to find any indexes you can use numpy.argwhere() in this example if you want to find first and second:

np.argwhere((~np.isnan(m)).any(1))
# [[2] first is exist in index 2
#  [3] second is exist in index 3
# ]