Questions about some parameters in param.json during the dpgen run process.

[xuyangmark]

I am trying to replicate LiSi's case, and I have some doubts about the settings of some parameters, mainly about model_devi_jobs:
"model_devi_jobs": [
{"sys_idx": [0],"temps": [200, 300, 400, 500],"press": [1.0, 100.0, 1000.0, 10000.0, 50000.0], "trj_freq": 5, "nsteps": 1000, "ensemble": "npt", "_idx": "00" },
{"sys_idx": [0],"temps": [200, 300, 400, 500],"press": [1.0, 100.0, 1000.0, 10000.0, 50000.0], "trj_freq": 5, "nsteps": 1000, "ensemble": "npt", "_idx": "01" },
{"sys_idx": [0],"temps": [200, 300, 400, 500],"press": [1.0, 100.0, 1000.0, 10000.0, 50000.0], "trj_freq": 10, "nsteps": 3000, "ensemble": "npt", "_idx": "02" },
{"sys_idx": [0],"temps": [200, 300, 400, 500],"press": [1.0, 100.0, 1000.0, 10000.0, 50000.0], "trj_freq": 10, "nsteps": 3000, "ensemble": "npt", "_idx": "03" },
{"sys_idx": [0],"temps": [200, 300, 400, 500],"press": [1.0, 100.0, 1000.0, 10000.0, 50000.0], "trj_freq": 20, "nsteps": 5000, "ensemble": "npt", "_idx": "04" },
{"sys_idx": [0],"temps": [200, 300, 400, 500],"press": [1.0, 100.0, 1000.0, 10000.0, 50000.0], "trj_freq": 20, "nsteps": 8000, "ensemble": "npt", "_idx": "05" }
]"
Questions are listed below:

(1)Do I need to set the pressure up to 50000 for my study of the constant-pressure system? Is it okay to just set it to 1?
(2)I noticed that the nsteps varies from 1000 to 8000. Do these numbers have any deeper consideration behind them?
(3)I think it is hard for the pressure to remain constant in small system with few atoms, such as LiSi. Maybe the NVT system is better. Does it make a difference for result of training whether to use nvt or npt system?

Thank you very much for spare time reading this discussion.