expression in variables

[Paul]

Can we use expression in variables

such as

N=2

p=N*2

pulse_2_1_[1:p]=10

I want to vary the parameter N as scan parameter

Thanks

Paul

[Rongchun Zhang]

No. You have to use scan_par

Refer to the manual

--
You received this message because you are subscribed to the Google Groups "spinev-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to spinev-discus...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

--

Best regards,

Rongchun

**************************************************
张荣纯 Rongchun Zhang
Room 4002, Chemistry Building, 

Biophysics and Department of Chemistry,

University of Michigan, Ann Arbor, MI, US.

Mobile: (01)734-353-9636

Email: zrc...@gmail.com
           rozh...@umich.edu

**************************************************

[Mikhail Veshtort]

Hi Paul,

You can certainly do something to that end, but not in the way you see it.

You should understand that things like pulse_2_1_[1:3]=10 are actually a shorthand for

pulse_2_1_1=10

pulse_2_1_2=10

pulse_2_1_3=10

So, hoping to vary p in your example would mean that you want to compute different expressions for each value of the scan parameter. This is, of course, impossible.

However, if for each value of p you simply want to assign different values to components of some vector, this can be easily done using expressions like ["1:p"]. So, for example,

pulse_2_1(["1:p"]')=10 is a perfectly legal expression. See the Matrices section in the Manual (page 9).

Mikhail

[Paul]

Thanks for your reply but my pulse program is such that I cannot use scan parameters

I have 80 pulses for which I want to set 4 pulses to be of 5 us and the rest to be zero

In the next I want to change the value of 14 pulses to be 5 us while the rest to be zero

So I have created a matrix of zeros in the pulse.pp file and want to change the pulse lengths systematically

May be I am not clear but it is hard to explain here I guess

Thanks for the reply once again

Paul

[Paul]

I did something like this

p=["1:N"]

pulse_2_1_p=5

and it gave me illegal expression error. I will try out your way

Thanks for the reply. Will update in a minute

Paul

[Paul]

Hi Mikhail,

Nope it gives me the error

Error: vectorized expression format: in line

[1:p]

Here is my script

****** The System *******

spectrometer(MHz)  700

spinning_freq(kHz) 10

channels           C13  H1

nuclei             C13  H1  H1  H1 

atomic_coords      ch2h.cor

cs_isotropic       0.0 -0.3 0.5 0.5 

csa_parameters     ch2h2.csa

j_coupling         *

quadrupole         *

dip_switchboard    ch2h.dip

csa_switchboard    *

exchange_nuclei    *

bond_len_nuclei    *

bond_ang_nuclei    *

tors_ang_nuclei    *

groups_nuclei      *

******* Pulse Sequence ******************************

CHN 1

timing(usec)      (0)256x200

power(kHz)         0

phase(deg)         0

freq_offs(kHz)     0

CHN 2         

timing(usec)      (../spinev_input/tppmpitppm.pp)

power(kHz)         *

phase(deg)         *

freq_offs(kHz)     *

****** Variables ************************************

N=1

p=N*2

***q=["42:41+p"]

***r=["1:2:p-1"]

***s=["2:2:p"]

pulse_2_1(["1:p"]')=5

pulse_2_1_[42:43]=5

power_2_1_[1:81]=100

phase_2_1_[1:41:42]=15

phase_2_1_[2:41:43]=-15

pulse_1_1_1=N*4*5+5

******* Options **************************************

rho0               I1x

observables        F1p

EulerAngles        rep168

n_gamma            16

line_broaden(Hz)   6

zerofill           *

FFT_dimensions     1

options            -v2 -re

On Thursday, December 12, 2013 7:42:07 PM UTC+5:30, Mikhail Veshtort wrote:

[Mikhail Veshtort]

Right... I think that you just need to put ["..."] in parentheses for the matrix transpose to work: pulse_2_1((["1:p"])').

You should try to debug this kind issues yourself, actually. The easiest way here would be to run "spinev -calc" and experiment with various expressions to see what exactly gives an error. Or you could do this within the input file.

Mikhail

[Paul]

Well I have tried all sorts of permutations and combinations with the -calc option and input file. Even if the calc option gives things as you expect, running the scripts gives an error.

Before posting this query I have tried a lot and I have been using spinevolution for a long time and have not faced something like this.

May be I am wrong but it was quite baffling to me.

and the option pulse_2_1((["1:p"])') does not work, though the expression is quite valid in spinev -calc

Thanks and regards

Paul

[Mikhail Veshtort]

Hi Paul,

It turns out that for some reason the program does not accept expressions like A(["1:p"]). I don't have time now to figure out why exactly this happens, but this can be easily circumvented by using an intermediate variable. In your specific case, you can do

X=(["1:p"])'

pulse_2_1(X)=5

Best regards,

Mikhail