This is what I'm trying to do:
Divide a numerically integrated array (over the whole data range) by a "quad" integrated function over a specific integration range:
maxw = lambda t,en: exp(-en/(kB*t)) # where the following is inside a loop through t values, and B is a data array with the same dimension as en
aa=[]
cc=[]
A=B*maxw(t,en)
aa.append(trapz(A,en)) # integrate numerically
cc.append(quad(lambda energ: exp(-energ/kB*t),e0,inf)) # integrate fn over range e0, inf
#end of loop in t
dd=array(aa)/array(cc)
which gives the error: "operands could not be broadcast together with shapes (n) (n,2)" and n is the length of the t-loop
Or to put it another way, for each value of "t" in the loop: (1) integrate a set of numerical data using trapz, and turn it into an array; and (2) integrate a function using quad between certain limits, also turned into an array; then (3) divide one array by the other to get a final array.
There's obviously something I'm missing. How should I do this?