question about data generation

[lin01109]

Hi, there! Thanks for your great work. Recently I'm using your code to generate dataset. But I noticed that the code of saving dataset involved truncation of the parameter string. It seems that the first five digits of the string are kept and the rest discarded. I wonder why this is done, is it for precision? Because when I try to generate a dataset with parameter 0.00101, the dataset name is only retained to 0.001, does this mean that the rest is out of precision? I hope you can understand my question and appreciate any reply.

Here's a snippet of code I've taken, and you can see that the dataset is named using only the first five digits of the parameter string

    jnp.save(cwd+cfg.multi.save+'1D_Advection_Sols_beta'+str(beta)[:5], uu)
    jnp.save(cwd + cfg.multi.save + '/x_coordinate', xc)
    jnp.save(cwd + cfg.multi.save + '/t_coordinate', tc)

jnp.save(cwd + cfg.multi.save+'/2D_ReacDiff_Multi_beta'+str(beta)[:5]+'_key'+str(cfg.multi.init_key), uu)
jnp.save(cwd + cfg.multi.save+'/x_coordinate', xc)
jnp.save(cwd + cfg.multi.save+'/y_coordinate', yc)
jnp.save(cwd + cfg.multi.save+'/t_coordinate', tc)
jnp.save(cwd + cfg.multi.save+'/nu_diff_coef_beta'+str(beta)[:5]+'_key'+str(cfg.multi.init_key), nu[:,2:-2,2:-2])

    jnp.save(cwd + cfg.multi.save+'/ReacDiff_Nu'+str(nu)[:5]+'_Rho'+str(rho)[:5], uu)
    jnp.save(cwd + cfg.multi.save+'/x_coordinate', xc)
    jnp.save(cwd + cfg.multi.save+'/t_coordinate', tc)