Using dpdata package to generate training set from CP2K2023 version

[StealFox97]

Dear developer:
Dpdata-matser only supports the *.log file of cp2k6.1. But I still want to use the calculated results from CP2K2023.
**I modified the source code which under the path of “../dpdata/cp2k/output.py”, in the dpdata package. I made the code in output.py be able to get the key word in the output file of cp2k2023. I uploaded the “output.py” file that i modified.
For "fmt = ‘cp2k/output’", it can work now!
However, for “fmt = ‘cp2k/aimd_output’”, it seems that the code can only extract the first frame of *.log file.

`import dpdata

data = dpdata.LabeledSystem('run', fmt='cp2k/aimd_output')
print(len(data))
1
`

test.zip

I am poor at Python. Which source code should I modifie next?
(Output.py was uploaded in txt format below）
output.txt

[StealFox97]

/

[robinzyb]

Please check out the devel branch of https://github.com/robinzyb/cp2kdata/tree/devel

[anmol-compchem]

@robinzyb can you please provide any script that I can run to get the desired output(i want to convert cp2k 2022.01 data to npy for deepmd kit?

[StealFox97]

Thank to the developer. The cp2kdata pakage is very esay to use. But I have some leisure time recently ,and i have modified the "output.py" script suuccesfuly. Now, it can work with output file which product by cp2k2023. I think it can also work with cp2k2022.01 because the two cp2k version have same surcture of output file.
@anmol-compchem I can share it with you if you want to try. Remamber change the output file name to *.log
output.py.txt.txt

[robinzyb]

Hello, because I just relocated to a new place recently, I didn't reply in time. Recently, we also update support for cp2k 2023. I will rountinely check the issue now. https://github.com/link89/cp2kdata/tree/fix-2023-support

[th-nic]

Does it support cp2k 8.2? I cannot convert my aimd_output to DeePMD system.

[robinzyb]

it is supposed to support v8.2 if you have any problem in using cp2kdata, please open an issue on cp2kdata repository

[anmol-compchem]

Thankyou for your reply when i am running the script
i am getting this error
Traceback (most recent call last):
File "/home/anmol/work/deep_md/00.data/readdata.py", line 5, in
data = dpdata.LabeledSystem('test', fmt = 'cp2k/aimd_output')
File "/home/anmol/.local/lib/python3.10/site-packages/dpdata/system.py", line 281, in init
self.from_fmt(
File "/home/anmol/.local/lib/python3.10/site-packages/dpdata/system.py", line 318, in from_fmt
return self.from_fmt_obj(load_format(fmt), file_name, **kwargs)
File "/home/anmol/.local/lib/python3.10/site-packages/dpdata/system.py", line 1162, in from_fmt_obj
data = fmtobj.from_labeled_system(file_name, **kwargs)
File "/home/anmol/.local/lib/python3.10/site-packages/dpdata/plugins/cp2k.py", line 13, in from_labeled_system
return tuple(Cp2kSystems(log_file, xyz_file, restart))
File "/home/anmol/.local/lib/python3.10/site-packages/dpdata/cp2k/output.py", line 58, in next
log_info_dict = self.handle_single_log_frame(next(self.log_block_generator))
File "/home/anmol/.local/lib/python3.10/site-packages/dpdata/cp2k/output.py", line 311, in handle_single_log_frame
float(line_list[3]) * AU_TO_EV_EVERY_ANG,
ValueError: could not convert string to float: '[a.u.]'
attaching all files
new.zip

[PolyuWeldingSpock]

Thankyou for your reply when i am running the script i am getting this error Traceback (most recent call last): File "/home/anmol/work/deep_md/00.data/readdata.py", line 5, in data = dpdata.LabeledSystem('test', fmt = 'cp2k/aimd_output') File "/home/anmol/.local/lib/python3.10/site-packages/dpdata/system.py", line 281, in init self.from_fmt( File "/home/anmol/.local/lib/python3.10/site-packages/dpdata/system.py", line 318, in from_fmt return self.from_fmt_obj(load_format(fmt), file_name, **kwargs) File "/home/anmol/.local/lib/python3.10/site-packages/dpdata/system.py", line 1162, in from_fmt_obj data = fmtobj.from_labeled_system(file_name, **kwargs) File "/home/anmol/.local/lib/python3.10/site-packages/dpdata/plugins/cp2k.py", line 13, in from_labeled_system return tuple(Cp2kSystems(log_file, xyz_file, restart)) File "/home/anmol/.local/lib/python3.10/site-packages/dpdata/cp2k/output.py", line 58, in next log_info_dict = self.handle_single_log_frame(next(self.log_block_generator)) File "/home/anmol/.local/lib/python3.10/site-packages/dpdata/cp2k/output.py", line 311, in handle_single_log_frame float(line_list[3]) * AU_TO_EV_EVERY_ANG, ValueError: could not convert string to float: '[a.u.]' attaching all files new.zip

Hi,

Do you solve the problem? I met same issue.

[StealFox97]

I am sorry. I am poor at Python. I have bee find long time, and i can not solve this problem. I also got this error when i use your file. But it is work normally with cp2k2023 version in my computer.

[anmol-compchem]

there is some issue with 2022.1 output but its working well for cp2k2023 output

[PolyuWeldingSpock]

there is some issue with 2022.1 output but its working well for cp2k2023 output

Did you replace the output.py in dpdata/cp2k/output.py with the newfile mentioned above?