Download EELS data

37 views
Skip to first unread message

xiangchen hu

unread,
Jul 23, 2019, 8:29:43 AM7/23/19
to hyperspy-users
Hi everyone,

When I use PCA algorithm to denoise EELS data, I can't find a approach to download the data after denoise. Clicking on the Download icon, a blank web page will appear. I am not familiar with Hyperspy, I hope someone can give me some suggestions.

Thanks a lot!

Xiangchen Hu

Josh Taillon

unread,
Jul 23, 2019, 11:10:11 AM7/23/19
to hypersp...@googlegroups.com
Hi Xiangchen,

Clicking that "save" icon is only for the figure itself. If you want to get access to the PCA-denoised signal, you need to use `s.get_decomposition_model()` as described in the user guide: http://hyperspy.org/hyperspy-doc/current/user_guide/mva.html#denoising

Hope that helps,
Josh Taillon

--
You received this message because you are subscribed to the Google Groups "hyperspy-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hyperspy-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hyperspy-users/b64bef9d-f168-402a-baa2-b0ad01de7cae%40googlegroups.com.


--
Joshua Taillon
Message has been deleted

Josh Taillon

unread,
Jul 23, 2019, 11:20:49 PM7/23/19
to hypersp...@googlegroups.com
I'm glad you were able to denoise the data. If you must export the data into another program, there are a couple of options. Digital Micrograph can read the .rpl/.raw format for spectrum images using the following script: https://github.com/hyperspy/ImportRPL  That would require you to save the denoised data as a .rpl file first.

I've also written some code (although it's been some time since I used it) that can load a HyperSpy signal into Matlab: https://github.com/jat255/readHyperSpyH5  

I would highly recommend that you try to continue your data processing in HyperSpy though, since once you take the data into another program, you lose all the calibration and associated metadata.

Also, I would recommend closely reading the User Guide (http://hyperspy.org/hyperspy-doc/current/user_guide/io.html#saving-data-to-files talks about saving files), since both of the questions you've had are covered in that document.

All the best,
Josh

On Tue, Jul 23, 2019 at 8:58 PM xiangchen hu <hux...@gmail.com> wrote:
Hi Josh,

denoise.PNG

Thank you very much for reply. I have tried the method that you provided, but still have some problems.

When I get the denoise SI data, I can't find a suitable format to save. Because I want to check the SI data on Digtal Micrograph or Matlab, and I know '.dm3' format cannot be save in Hyperspy. So how can I save the SI data, rather than a image. Thanks for your assistance!

Best wishes!
Xiangchen Hu

在 2019年7月23日星期二 UTC+8下午11:10:11，Josh Taillon写道：
Hi Xiangchen,

Clicking that "save" icon is only for the figure itself. If you want to get access to the PCA-denoised signal, you need to use `s.get_decomposition_model()` as described in the user guide: http://hyperspy.org/hyperspy-doc/current/user_guide/mva.html#denoising

Hope that helps,
Josh Taillon

On Tue, Jul 23, 2019 at 6:29 AM xiangchen hu <hux...@gmail.com> wrote:
Hi everyone,

When I use PCA algorithm to denoise EELS data, I can't find a approach to download the data after denoise. Clicking on the Download icon, a blank web page will appear. I am not familiar with Hyperspy, I hope someone can give me some suggestions.

Thanks a lot!

Xiangchen Hu

--
You received this message because you are subscribed to the Google Groups "hyperspy-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hypersp...@googlegroups.com.


--
Joshua Taillon

--
You received this message because you are subscribed to the Google Groups "hyperspy-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hyperspy-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hyperspy-users/a6ae127d-0932-4d69-a266-8fd1e15cf409%40googlegroups.com.


--
Joshua Taillon

xiangchen hu

unread,
Jul 23, 2019, 11:47:25 PM7/23/19
to hyperspy-users
Thank you for your help! As a beginner, I have many questions about Hyperspy. Your advise is useful for me,  I will learn the fundamental operation as fast as possible.

Best wishes!
Xiangchen Hu

在 2019年7月24日星期三 UTC+8上午11:20:49，Josh Taillon写道：


--
Joshua Taillon
Reply all
Reply to author
Forward
0 new messages