I have an array of numbers with shape (1220,) called x
I'm looking at numbers greater than 1.0,
mask1 = [i for i in x if i>1.0 ]
returning
[1.2958354, 1.0839227, 1.1919032]
My question now is then how am able to determine the index location of these values in my initial array x?
I've tried each individually, but an error occurs
list(x).index(1.2958354)
ValueError: 1.2958354 is not in list
You've tagged this as numpy, and describe a shape (not len). Which leads me to think you have a numpy array.
In [665]: x=np.random.rand(10)
In [666]: x
Out[666]:
array([ 0.6708692 , 0.2856505 , 0.19186508, 0.59411697, 0.1188686 ,
0.54921919, 0.77402055, 0.12569494, 0.08807101, 0.11623299])
In [667]: x>.5
Out[667]: array([ True, False, False, True, False, True, True, False, False, False], dtype=bool)
In [668]: list(x).index(.6708692)
ValueError: 0.6708692 is not in list
The reason for the ValueError is that you are looking for a float, and those often don't exactly match. If the array was of ints, then such an index would work.
In [669]: list(np.arange(10)).index(5)
Out[669]: 5
This reasoning applies whether x was an array or a list.
numpy has a where that returns the index of boolean true values in an array
In [670]: np.where(x>.5)
Out[670]: (array([0, 3, 5, 6], dtype=int32),)
x>.5 is the boolean array as shown above, and [0,3,5,6] the index values where this is true.
In [671]: x[np.where(x>.5)]
Out[671]: array([ 0.6708692 , 0.59411697, 0.54921919, 0.77402055])
The equality test doesn't work any better
In [672]: x[np.where(x==0.6708692)]
Out[672]: array([], dtype=float64)
For floats there is the concept of close - the difference is within some tolerance (np.allclose is particularly useful):
In [679]: np.where(np.isclose(x,0.59411697))
Out[679]: (array([3], dtype=int32),)
For lists, one of the enumerate solutions is great, and also works with 1d arrays. But it is already an array, use the where.
You can use enumerate function, for example:
mask1 = [(i, value) for i, value in enumerate(x) if value>1.0 ]
print mask1
Try:
mask1 = [i for i in range(len(x)) if x[i]>1.0]
You can use the index function on x, which will return the first index of each value. To get a list of all the indices from mask1 try:
map(x.index, mask1)
mask_1 = [index for index, value in enumerate(x) if value > 1.0]
Use enumerate to create tuple pairs of the index and the filtered value, then use zip with the * operator to unpack the variables into separate lists.
a = np.array([0, 1, 2, 3])
idx, vals = zip(*[(i, v) for i, v in enumerate(a) if v > 1])
>>> idx
(2, 3)
>>> vals
(2, 3)
Try using enumerate() to get the index and value together:
mask1 = [(i,v) for i,v in enumerate(x) if v > 1.0]
