形状不匹配的Python

Question

So I'm trying to create a 3D array with some standard values defined by myself (actually other people, but that doesn't matter) This is my code:

Tt = 120 # Total duration (sec)
delta = 0.001 # Time bin (sec)
T = Tt/delta # number of time bins
Ncells = 8;
Cmap = np.zeros([Ncells,Ncells,int(T)])
ExcInhID = np.expand_dims([1, -1, 1, -1, 1, -1, 1], axis =1)
a = list(range(0,int(T)))

for t in a:
    Cmap[:,:,t] = (-1)*np.identity(Ncells)
    Cmap[1:,0,t] = list(ExcInhID*np.ones([Ncells-1,1])*(1 - min(max(t-T/3,0)*(1/(T/3)),0)))
    CC5 = list(np.arange(0,Ncells)) ; del CC5[4];
    Cmap[CC5,4,t] = list(ExcInhID*np.ones([Ncells-1,1])*(min(max(t-T/3,0)*(1/(T/3)),0)))
    Cmap[6,2,t] = 1; Cmap[5,7,t] = -1; Cmap[1,5,t] = 1;

This does work until

Cmap[CC5,4,t] = list(ExcInhID*np.ones([Ncells-1,1])*(min(max(t-T/3,0)*(1/(T/3)),0)))  Cmap[CC5,4,t] = list(ExcInhID*np.ones([Ncells-1,1])*(min(max(t-T/3,0)*(1/(T/3)),0)))

Where I get the error:

ValueError: shape mismatch: value array of shape (7,1) could not be broadcast to indexing result of shape (7,)

Which is weird since it did work on the other line over here:

 Cmap[1:,0,t] = list(ExcInhID*np.ones([Ncells-1,1])*(1 - min(max(t-T/3,0)*(1/(T/3)),0)))

I've tried everything and don't know what to do anymore.. Does any of you have suggestions?

Answer 1

I am not exactly sure why you did not get error on the first instance and got only in the second. But if you change your lines as below, things would work.

From

Cmap[1:,0,t] = list(ExcInhID*np.ones([Ncells-1,1])*(1 - min(max(t-T/3,0)*(1/(T/3)),0)))
....
Cmap[CC5,4,t] = list(ExcInhID*np.ones([Ncells-1,1])*(min(max(t-T/3,0)*(1/(T/3)),0)))

to

Cmap[1:,0,t] = np.stack(ExcInhID*np.ones([Ncells-1,1])*(1 - min(max(t-T/3,0)*(1/(T/3)),0)), axis=1)
....
Cmap[CC5,4,t] = np.stack(ExcInhID*np.ones([Ncells-1,1])*(min(max(t-T/3,0)*(1/(T/3)),0)), axis=1)