About tracking the value of fingerprint

[Seo, Seung-Woo]

Dear Dr. Artrith,

Could you let me know how to track the calculated values of fingerprint?

I want to know how wide structure field is covered by dataset in fingerprint space. 

I think generate.x debug mode might give this information but I don't know how to use it.

Thank you

Best Regards,

Eric

--

Seo, Seung-Woo

Ph. D

Research Staff Member

Computer Aided Engineering group (CAE)

Platform Technology Lab.

Samsung Advanced Institute of Technology (SAIT)

Samsung Electronics. Inc. 

+82-10-2487-1683

[Nongnuch Artrith]

Dear Eric,

You can use the tool 'trnset2ASCII.x' to convert the binary training set file generated by 'generate.x' to a human-readable text file.  The aenet tools are compiled with 'make tools -f makefiles/Makefile...' from the 'src' directory, and the executables are placed in the 'tools' directory (not in 'bin').

The tool is used as follows:

  trnset2ASCII.x --raw TiO.train OUT.txt

where the '--raw' option prevents the rescaling of the fingerprint functions.

There 'TiO.train' should be replaced with the name of your training set file.  The format of the output file is unfortunately a bit cryptic.  Please let us know if you face any problems with this.

Here also the GitHub link: https://github.com/atomisticnet/aenet/blob/master/src/tools/trnset2ASCII.f90

Best,

Nong

[인원여]

Dear Dr. Artrith

Conversely, is it possible or is there a tool to convert the human-readable text file to the binary training set file?

Thank you

Best regards,

In Won Yeu

========================================================

Ph.D. Course

Dielectric Thin Film Laboratory

Department of Materials Science & Engineering

Seoul National University

2019년 2월 22일 금요일 오후 12시 55분 11초 UTC+9, Nong Artrith 님의 말:

[Nongnuch Artrith]

Dear In Won Yeu,

The tool ’trnset2ASCII.x’ can also do the reverse conversion, i.e., read in text-format training set files and write out files in binary format.  

This is activated with the command line flag ‘--to-binary’.

Best regards,

Nong

[인원여]

Dear Dr. Artrith

Thank you for early response.

I tried the conversion and reverse conversion between the binary training set file and human-readable text file.

The conversion from binary to ASCII file was successful by the tool ’trnset2ASCII.x’.

However, the reverse conversion from the generated ASCII to binary file did not work leaving the following message:

 Converting ASCII format to binary format.

forrtl: severe (256): unformatted I/O to unit open for formatted transfers, unit 20, file /home/~

Image              PC                              Routine                  Line           Source             

libintlc.so.5      00002AE4175A3A1E  Unknown               Unknown  Unknown

libintlc.so.5      00002AE4175A24B6  Unknown               Unknown  Unknown

libifcore.so.5     00002AE41696219E  Unknown               Unknown  Unknown

libifcore.so.5     00002AE4168D1C4E  Unknown               Unknown  Unknown

libifcore.so.5     00002AE4168D114D  Unknown               Unknown  Unknown

libifcore.so.5     00002AE416906B04  Unknown               Unknown  Unknown

trnset2ASCII.x     00000000004B04AC  Unknown               Unknown  Unknown

trnset2ASCII.x     0000000000409D19  Unknown               Unknown  Unknown

trnset2ASCII.x     0000000000409C2C  Unknown               Unknown  Unknown

libc.so.6          0000003A39A1ED1D  Unknown               Unknown  Unknown

trnset2ASCII.x     0000000000409B29  Unknown               Unknown  Unknown

I am not sure what this error message means and cannot fix the problem. 

Thank you

Best regards,

In Won Yeu

2019년 7월 22일 월요일 오후 11시 28분 16초 UTC+9, Nong Artrith 님의 말:

[Nongnuch Artrith]

Dear In Won Yeu,

There was indeed a typo in the source file ’trainset.f90’.

On line 661 in that file, the following line

   ts1 = load_TrnSet_info(unit=u)

should have been

   ts1 = load_TrnSet_info_ASCII(unit=u)

I changed the file in the GitHub repository (https://github.com/atomisticnet/aenet), and I confirmed that with this change the following does not crash

$ trnset2ASCII.x test.train test.ascii
$ trnset2ASCII.x --to-binary test.ascii test2.bin

Please use either the attached version of ’trainset.f90’ or download the aenet source again from GitHub (https://github.com/atomisticnet/aenet).

Thank you very much for spotting this bug!

Best regards,
Nong