I have several bumpy arrays and I want to concatenate them. I am using np.concatenate((array1,array2),axis=1). My problem now is that I want to make the number of arrays parametrizable, I wrote this function
x1=np.array([1,0,1])
x2=np.array([0,0,1])
x3=np.array([1,1,1])
def conc_func(*args):
xt=[]
for a in args:
xt=np.concatenate(a,axis=1)
print xt
return xt
xt=conc_func(x1,x2,x3)
this function returns ([1,1,1]), I want it to return ([1,0,1,0,0,1,1,1,1]). I tried to add the for loop inside the np.concatenate as such
xt =np.concatenate((for a in args: a),axis=1)
but I am getting a syntax error. I can't used neither append nor extend because I have to deal with numpy arrays and not lists. Can somebody help?
Thanks in advance
concatenate can accept a sequence of array-likes, such as args:
In [11]: args = (x1, x2, x3)
In [12]: xt = np.concatenate(args)
In [13]: xt
Out[13]: array([1, 0, 1, 0, 0, 1, 1, 1, 1])
By the way, although axis=1 works, the inputs are all 1-dimensional arrays (so they only have a 0-axis). So it makes more sense to use axis=0 or omit axis entirely since the default is axis=0.