Thank you for uploading such video tutorials which might help several people in fitting their data. However, I am just wondering about the "goodness of fit" parameter which is missing in this method. So, basically we don't have idea whether we could tune some parameter to make the fit better.
Thank you for the video, it has been helpful. I keep running into this problem with some datasets where it gives an error that says I have exceeded the maximum number of iterations and it can't converge. Any idea how I can change the allowed number of iterations?
I guess you have soled your problem by now. Sorry for late reply. I was down(lost access) for several months due to some technical fault in YT. Back now.
Hey, thanks a lot for the video. But I get a message saying that - " The fit does not converge-mutual dependency exists between parameters. You may have over parameterized the fitting function. Fixing one of them may eliminate this problem." Any suggestions ?? Thanks in advance!
It depends on the nature of data. If the baseline of the data is similar to what I showed in the video, then yes, you can do without baseline correction.
Hi ... I have a doubt . Can we fit two curves one is standard spectra curve and other is data of unknown curve and then we need to find out the values of unknown curve in comparison with the standard spectra . Thank you.
That would be very useful, I am sure. I think there are some specific software to do that. However, I have not used any. Following article may help. inscilab.com/tips-tricks-tutorials/multiple-peaks-fitting-of-xps-raman-pl-spectroscopy-curve-fitting-using-origin-explanation/ This is how I do it.
Guys! Sorry for late reply. I was down(lost access) for several months due to some technical fault in YT. Back now. I guess you have soled your problem by now. If not, let me know.
You have to understand the shape from the expt. data. Identify the shape and fit with the function of that shape. However, I am not expert. So, I just try with closest possible shapes and choose the best fitted one.
Hello, I cannot pass to the selection of the peaks because it shows an error: "the input data are not monotonic increase". I already tried changing to ascending/descending the columns/worksheet and nothing worked. Do you know by any chance why this is occurring?
Check data points manually. one or more points have unexpected values or negative values or values are missing. Check and let me know if don't find. i will try to check the file, then.
Guys! Sorry for late reply. I was down for several months due to some technical fault in YT. Back now. If you still looking for help, please elaborate.
BY clicking on exact point only. And that's disappointing, I know. I guess you have soled your problem by now. Sorry for late reply. I was down(lost access) for several months due to some technical fault in YT. Back now.
Sorry for late reply. I was down(lost access) for several months due to some technical fault in YT. Back now.I guess you have soled your problem by now.
When i am following the same procedure suggested by you i am getting a message that the fit didn't converge after many iteration. So what to do. can i sen u my data file?
Yes I got it. I have tried to fit. But, could not do it properly. The problem lies in data. How many peaks should be in the range 100 - 10000? I am surprised to see that at least 32 peaks are there. Below 200 at least 13 peaks are there. I guess these are causing problem. This simple method will not be sufficient for your data. Or I can suggest you to check your samples properties and phase.
Yes! Sorry for late reply. I was down(lost access) for several months due to some technical fault in YT. Back now. I guess you have soled your problem by now.
Guys! Sorry for late reply. I was down(lost access) for several months due to some technical fault in YT. Back now. I guess you have soled your problem by now.
Please, can you help me to do the curve fitting for Raman spectra using Origin to obtain four lorentzian shapes (G, D, D2, D4 ) and one gaussian shape (D3) as it is reported in the article below: Raman microspectroscopy of soot and related carbonaceous material: Spectral analysis and structural information. Carbon, 43, (2005) 1731-1742.
I never tried this kind of mixed shape fitting. I guess following way can be easier. I am not sure. A) Cut the data into 2 set. 1 will have the only D3. then separately do the fitting for each and finally do convolution of those. Or B) Choose the psedo function which is a complicated mix of Lorentzian and Gaussian. I guess if the fit comes okay then you can understand that the software can do it automatically. Or C) Write function manually........ I am totally stupid in this.
Thanks! I wait better explanation for the best understanding especially on how to select peaks, selected peaks exactness confirmation criteria, calculating deconvoluted peaks area and getting the best deconvoluted peaks. If you gave me your email, I can send my protein secondary structure data to benefit more. Thank you so much!! My email: wondimagegneams@yahoo.com
Thank you for uploading such video tutorials which might help several people in fitting their data. However, I am just wondering about the "goodness of fit" parameter which is missing in this method. So, basically we don't have idea whether we could tune some parameter to make the fit better.
Thanks for your video. I have one question: how can we set the baseline? In my case, the auto-created baseline is over the plot. Thanks
thank you so much for this video tutorial, this helped me a lot.
Very useful video, thank you so much.
How did you put baseline in the raman spectra after fitting
Thank you for the explanation. So how i can identify my peak if I have amourphous phase with other peak in DRX
Best regards
thank you very much!
Many thanks
thanks, this video was really helpful. can you please explain regarding the amide 1 curve fitting for protein secondary structure determination?
watch this explanation if it helps th-cam.com/video/rscWyufhFko/w-d-xo.html
Thank you for the video, it has been helpful. I keep running into this problem with some datasets where it gives an error that says I have exceeded the maximum number of iterations and it can't converge. Any idea how I can change the allowed number of iterations?
I guess you have soled your problem by now.
Sorry for late reply.
I was down(lost access) for several months due to some technical fault in YT.
Back now.
Hey, thanks a lot for the video. But I get a message saying that - " The fit does not converge-mutual dependency exists between parameters. You may have over parameterized the fitting function. Fixing one of them may eliminate this problem." Any suggestions ??
Thanks in advance!
Try changing the estimated number or positions of the peaks A LITTLE.
after we got peak position, width etc. from origin , what about next software can use auto-matching instead reading literature?
Thanks Sir
How do you know when to choose Gaussian or Lorenztia or mixed?
1. By looking at the shape.
2. Trial and error method to get the best fit.
I think this done by stimulating annealing methods
But I wish to know it's algorithms
HOW TO IDENTIFY WHICH ELEMENT OR COMPOUND IS PRESENT BY LOOKING THE PEAK? ALSO KINDLY LET ME KNOW HOW QUANTITATIVE ANALYSIS CAN BE DONE.
what is the sound title played ?
If I have a raman data is it ok to do multiple peak fit analysis without doing the baseline correction pleasereply
It depends on the nature of data.
If the baseline of the data is similar to what I showed in the video, then yes, you can do without baseline correction.
Hi ... I have a doubt . Can we fit two curves one is standard spectra curve and other is data of unknown curve and then we need to find out the values of unknown curve in comparison with the standard spectra . Thank you.
That would be very useful, I am sure.
I think there are some specific software to do that. However, I have not used any.
Following article may help. inscilab.com/tips-tricks-tutorials/multiple-peaks-fitting-of-xps-raman-pl-spectroscopy-curve-fitting-using-origin-explanation/
This is how I do it.
Sir ,please make a video on fit2D software, which is used to analysis 2D graph on GiD
thank you for video! please I have question how we plot differentes curve in same graphe ( X1, Y1 ), (X2,Y2)....in origin
Guys! Sorry for late reply.
I was down(lost access) for several months due to some technical fault in YT.
Back now.
I guess you have soled your problem by now. If not, let me know.
why only Lorentzian method, what is the main diff bwt G-L-V, when to choose?
You have to understand the shape from the expt. data. Identify the shape and fit with the function of that shape. However, I am not expert. So, I just try with closest possible shapes and choose the best fitted one.
I got so many Gaussian peaks from di/dv spectroscopy of stm.....how can I fix it?
watch this explanation if it helps th-cam.com/video/rscWyufhFko/w-d-xo.html
Hello, I cannot pass to the selection of the peaks because it shows an error: "the input data are not monotonic increase". I already tried changing to ascending/descending the columns/worksheet and nothing worked. Do you know by any chance why this is occurring?
Check data points manually. one or more points have unexpected values or negative values or values are missing. Check and let me know if don't find. i will try to check the file, then.
I have rate of change of frequency data and time data . Can I use the peaks to get the average of the rate of change of frequency with 100ms windows?
Guys! Sorry for late reply.
I was down for several months due to some technical fault in YT.
Back now.
If you still looking for help, please elaborate.
this rocks
How to know the Lorentz peak or Gaussion Peak or Mixed peak?
By looking at the shape of the peak/peaks.
Guys! Sorry for late reply.
I was down for several months due to some technical fault in YT.
Back now.
Is it possible to force the position of a peak?
BY clicking on exact point only. And that's disappointing, I know.
I guess you have soled your problem by now.
Sorry for late reply.
I was down(lost access) for several months due to some technical fault in YT.
Back now.
Please, can you help me to do one fitting for Raman spectra using four lorentzian shapes (G, D, D2, D4 ) and one gaussian shape (D3).
Sorry for late reply.
I was down(lost access) for several months due to some technical fault in YT.
Back now.I guess you have soled your problem by now.
It keeps telling me that data is not of monotonic increase, i can't get to the peakfit window after hitting the beaks and baseline option
same problem here :
facing same problem here
Just start your data from downward to higher one than it will works
Still its not working. keeps showing that data is not monotonic increase.
click on the column (with the right mouse button) -> sort column-> ascending
Can I do this on MATLAB or SCILAB ?
I know, people do it on MATLAB too. But I have no experience on that.
When i am following the same procedure suggested by you i am getting a message that the fit didn't converge after many iteration. So what to do. can i sen u my data file?
Sure. Send me your data.
Can u provide me ur mail id?
i have the data on ur mail id contact@easywayhowto.com.
Yes I got it.
I have tried to fit. But, could not do it properly. The problem lies in data. How many peaks should be in the range 100 - 10000? I am surprised to see that at least 32 peaks are there. Below 200 at least 13 peaks are there. I guess these are causing problem. This simple method will not be sufficient for your data. Or I can suggest you to check your samples properties and phase.
The reduced chi-square number seems huge. Should we worry about that?
Yes!
Sorry for late reply.
I was down(lost access) for several months due to some technical fault in YT.
Back now.
I guess you have soled your problem by now.
Hai..it was more informative...can you please explain how the peak shape needs to be identified...
Guys! Sorry for late reply.
I was down(lost access) for several months due to some technical fault in YT.
Back now.
I guess you have soled your problem by now.
Thanks bro
Please, can you help me to do the curve fitting for Raman spectra using Origin to obtain four lorentzian shapes (G, D, D2, D4 ) and one gaussian shape (D3) as it is reported in the article below:
Raman microspectroscopy of soot and related carbonaceous material: Spectral analysis and structural information. Carbon, 43, (2005) 1731-1742.
I never tried this kind of mixed shape fitting. I guess following way can be easier. I am not sure.
A) Cut the data into 2 set. 1 will have the only D3. then separately do the fitting for each and finally do convolution of those.
Or
B) Choose the psedo function which is a complicated mix of Lorentzian and Gaussian. I guess if the fit comes okay then you can understand that the software can do it automatically.
Or
C) Write function manually........ I am totally stupid in this.
Can you try with my data. can you give me your mail,
send your data to contact@easywayhowto.com
Faciing problem: that data is not of monotonic increase
The video was very nice but no explanation. Can you send me with an explanation? Thanks
I will try to upload a video with explanation. Stay tuned. Thanks for asking and watching the video.
Thanks! I wait better explanation for the best understanding especially on how to select peaks, selected peaks exactness confirmation criteria, calculating deconvoluted peaks area and getting the best deconvoluted peaks. If you gave me your email, I can send my protein secondary structure data to benefit more. Thank you so much!! My email: wondimagegneams@yahoo.com
@@wondimagegneassefamolla1289 Watch the video, I have made for you and others who have asked for it. th-cam.com/video/rscWyufhFko/w-d-xo.html