Fit ECIs manually #345
Comments
|
If you look at the Alternatively, you can use any method you want to do the fitting and then use the Python function |
|
Hi Brian,
My train file is like this:
# alias_or_name selected
SCEL1_1_1_1_0_0_0/0 1
SCEL1_1_1_1_0_0_0/1 1
SCEL1_1_1_1_0_0_0/2 1
SCEL1_1_1_1_0_0_0/3 1
SCEL1_1_1_1_0_0_0/4 1
SCEL1_1_1_1_0_0_0/5 1
SCEL1_1_1_1_0_0_0/6 1
SCEL1_1_1_1_0_0_0/7 1
......
Do you mean adding a third column to this "train" file with the column name
"formation_free_energy", and setting the y value as "formation_free_energy"
in the fit.json file?
Do I need to change "filetype" to csv or json and query the "corr" and
"formation_free_energy" to an output csv or json file and let casm-learn
read that output file?
If I am correct, casm-learn will somehow identify the "train" selection
file, and read the X and y data from somewhere (where?) if X and y is not
present in the "train" file?
Thanks!
{
"problem_specs": {
"data": {
"filename": "train",
"filetype": "selection",
"X": "corr",
"y": "formation_energy",
"kwargs": null
},
…On Tue, Jan 9, 2024 at 11:47 AM Brian Puchala ***@***.***> wrote:
If you look at the casm learn --settings-format help for the
"problem_specs"/"data" parameters, there are some options for specifying
the input data. You could for instance add a new column to a "train"
selection file and set "X": "new_column_name".
Alternatively, you can use any method you want to do the fitting and then
use the Python function casm.project.write_eci
<https://github.com/prisms-center/CASMpython/blob/7cb55a10cfecf66e3fae1895aa9d1d2f9feb794c/casm/casm/project/io.py#L11C5-L11C14>
to write the eci.json file.
—
Reply to this email directly, view it on GitHub
<#345 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AEDNA5JTXRH5YWXZR6NIRKTYNVYBVAVCNFSM6AAAAABBPBPH4KVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQOBTGQYDSOJWGY>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
--
Lei Zhang
Postdoc Scholar
Mechanical Engineering
Carnegie Mellon University
Google Scholar
<https://scholar.google.com/citations?hl=en&view_op=list_works&gmla=AJsN-F51HKnfWWVxbsFk4uAWX43G9CzcwBr_848Z2BxrFiXa0q6bPGZqhzAADPpA0vojEKOVP8bkTFTV3wDI8XvJfy5V2hvIuRNadTccjO0J3niAX63sVSA&user=139-zKwAAAAJ>
Linkedin <https://www.linkedin.com/in/lei-zhang-0b8aa126/>
|
|
Hi Brian, I did what you said, but I got an error. Could you please help me on this?
Loading fit_1_ga.json Hall of Fame size: 10Get problem data...Traceback (most recent call last): The above exception was the direct cause of the following exception: Traceback (most recent call last): |
Dear CASM developers,
In the case that I want to modify the total energy at 0 K from DFT by, say, adding vibrational contribution, or any other type of corrections, and then fit the ECIs, what would be the easiest way to do that?
Should I do it completely separately from casm-learn functionalities, or try to modify some of the key input files that casm-learn uses (the energy part specifically)?
Thanks!
The text was updated successfully, but these errors were encountered: