Simulating position eigenkets

[VanDammeHouten]

Hi I am currently trying to simulate the probabilities of measuring single photons with wavefunctions   that are defined as:

 

I'm doing this in the discrete domain by taking finite position eigenkets 

Thus the single photon kets become 

Currently I am representing my single photons as column vectors  in numpy 

```

x = np.arange(-l/2,l/2+dx,dx)

psis = GuassianPsf(x,sigma)

```

however whenever I try to convert them to kets

```

ketPsis = qutip.ket(psis)

```

I get "OverflowError: value too large"

At the moment (psi.shape == 2001).

I'm relatively new to Qutip and want to leverage it for my simulations but I feel like I'm possibly going about it the wrong way.

Any help is appreciated!

Cheers

[Simon Cross]

Hi Zach,

qutip.ket is not the right function for your use case. It creates a single ket that is the tensor product of basis states. So qutip.ket("01") -> |01>.

You should probably call qutip.Qobj(psis, dims=[[n], [1]]) or similar to directly create the ket you want.

You can leave out the dims in this case because Qobj will default to the right value. I included it only because it's a good idea and often forgotten in more complex cases (including by me and the QuTiP development team :).

Regards,

Simon

[VanDammeHouten]

Fantastic that works! Thanks for your help

[Simon Cross]

Yay! Pleasure.