ValueError: operands could not be broadcast together with shapes (41696,3,3) (41684,1,1) , Error While converting cp2k data to input data for DeepMD using dpdata

[369pankaj]

While Converting CP2K data to DeepMD input using dpdata I faced the following error:
"CP2KDATA| Virial Parsing using cp2kdata as plug in for dpdata was not multiplied by volume before cp2kdata v0.6.4 please check the cp2kdata version and the virial.npy
Traceback (most recent call last):
File "/home/pankaj/ML_DPMD/aimd/64W/NVT/conf2/pdn/0.5fs_ts/create_data.py", line 8, in
dp = dpdata.LabeledSystem(cp2kmd_dir, cp2k_output_name=cp2kmd_output_name, fmt="cp2kdata/md")
File "/home/pankaj/anaconda3/envs/deepmd/lib/python3.10/site-packages/dpdata/system.py", line 202, in init
self.from_fmt(
File "/home/pankaj/anaconda3/envs/deepmd/lib/python3.10/site-packages/dpdata/system.py", line 239, in from_fmt
return self.from_fmt_obj(load_format(fmt), file_name, **kwargs)
File "/home/pankaj/anaconda3/envs/deepmd/lib/python3.10/site-packages/dpdata/system.py", line 1226, in from_fmt_obj
data = fmtobj.from_labeled_system(file_name, **kwargs)
File "/home/pankaj/anaconda3/envs/deepmd/lib/python3.10/site-packages/cp2kdata/dpdata_plugin.py", line 144, in from_labeled_system
data['virials'] = cp2kmd.stress_tensor_list*volumes/EV_ANG_m3_TO_GPa
ValueError: operands could not be broadcast together with shapes (41696,3,3) (41684,1,1)"

I have used cp2k version 2024.3. Please Help me to resolve this issue. Thanks in advance.

[robinzyb]

Hello, I need your cp2k md outputs for debugging.

[369pankaj]

Please find the link attached for the files, https://drive.google.com/drive/folders/1dmJJpOw97qy-AWTat6Cgm3fgnDPT3uFX?usp=drive_link. And let me know. Thanks in advance.

[369pankaj]

https://drive.google.com/drive/folders/116BQDIo9h8K1_jFLzM5WPm_wzHK4UbB6?role=writer
Please look into these files also, it contains very small trajectory, with same error while converting. What I found is it can happen with any system(how small it is) if the cp2k data generation has multiple restart.

[369pankaj]

https://drive.google.com/drive/folders/1L2_sCi8rj80nJShy4cfjA-aWPOhTykVp?usp=drive_link Please look into these files also, it contains smaller trajectory, with same error while converting. *Thanks & Regards* Pankaj Adhikary

…

On Fri, Jan 24, 2025 at 10:14 PM Yong-Bin Zhuang ***@***.***> wrote: Hello, I need your cp2k md outputs for debugging. — Reply to this email directly, view it on GitHub <#789 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BC7L2YCRVOKR6J7SVDNGQJD2MJUVLAVCNFSM6AAAAABVZLX4Z6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTCOJUGUZTSNY> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[369pankaj]

https://drive.google.com/drive/folders/116BQDIo9h8K1_jFLzM5WPm_wzHK4UbB6?role=writer Very small trajectory with multiple restart step in the data (cp2k aimd) generation process. Having the same error. So what I understand, perhaps it's happening because of multiple restarts in the trajectory. On Mon, Jan 27, 2025 at 3:10 PM CY22D506 Pankaj Adhikary < ***@***.***> wrote:

…

https://drive.google.com/drive/folders/1L2_sCi8rj80nJShy4cfjA-aWPOhTykVp?usp=drive_link Please look into these files also, it contains smaller trajectory, with same error while converting. *Thanks & Regards* Pankaj Adhikary On Fri, Jan 24, 2025 at 10:14 PM Yong-Bin Zhuang ***@***.***> wrote: > Hello, I need your cp2k md outputs for debugging. > > — > Reply to this email directly, view it on GitHub > <#789 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/BC7L2YCRVOKR6J7SVDNGQJD2MJUVLAVCNFSM6AAAAABVZLX4Z6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTCOJUGUZTSNY> > . > You are receiving this because you authored the thread.Message ID: > ***@***.***> >

[robinzyb]

Hello, I am busy these days. cp2kdata cannot deal with an output containing multiple restart MD.

[369pankaj]

Thanks for your reply. However I've talked with some user of the older version of the dpdata they have successfully converted multiple restarted cp2k data with dpdata. With limited resources that we have, running cp2k jobs continuously is very memory extensive and it automatically got killed after some time so restart is inevitable in my case. Please help me to resolve this issue. Thanks in advance.

…

On Thu, 30 Jan, 2025, 2:44 am Yong-Bin Zhuang, ***@***.***> wrote: Hello, I am busy these days. cp2kdata cannot deal with an output containing multiple restart MD. — Reply to this email directly, view it on GitHub <#789 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BC7L2YD7RHNTIUX3U276LOL2NFACZAVCNFSM6AAAAABVZLX4Z6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTEMBQGAZDKMQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>