rouyang2017 / SISSO

A data-driven method combining symbolic regression and compressed sensing for accurate & interpretable models.
Apache License 2.0
246 stars 81 forks source link

最新版本的SISSO_predict无法读取SISSO.out #71

Open siyu-g opened 4 days ago

siyu-g commented 4 days ago

你好,我之前一直可以正常使用SISSO,但是最近一次用最新版本的SISSO编译之后,运行SISSO是没有问题,但是我发现output的SISSO.out文件里的模型文本发生了改变,并且用SISSO_predict这个utility的时候会无法读取SISSO.out,请问是不是最近编译的改变导致的?有没有可能同步更新一下SISSO_predict的安装文件? 我的SISSO.out:

**************************************************
Dimension:   1
-------------------
Feature Construction (FC) starts ...
Total number of features in the space Phi00:             13

Generating Phi01 ...
mpirank =   10  #generated =             13  #selected =             26  progress = 100.0%
mpirank =   11  #generated =             10  #selected =             23  progress = 100.0%
mpirank =   12  #generated =              6  #selected =             17  progress = 100.0%
mpirank =    8  #generated =             18  #selected =             27  progress = 100.0%
mpirank =    9  #generated =             15  #selected =             25  progress = 100.0%
mpirank =    2  #generated =             52  #selected =             26  progress = 100.0%
mpirank =    3  #generated =             46  #selected =             23  progress = 100.0%
mpirank =    4  #generated =             40  #selected =             26  progress = 100.0%
mpirank =    5  #generated =             34  #selected =             24  progress = 100.0%
mpirank =    6  #generated =             28  #selected =             20  progress = 100.0%
mpirank =    0  #generated =             64  #selected =             36  progress = 100.0%
mpirank =    7  #generated =             20  #selected =             33  progress = 100.0%
mpirank =    1  #generated =             58  #selected =             36  progress = 100.0%
Total number of newly generated features:             404
Total number of features in the space Phi01:            417

Redundant check on selected features ...

Sure Independence Screening on selected features ...
Size of the SIS-selected subspace from Phi01:         20

Descriptor Identification (DI) starts ...
dimension =   1   mpirank =    1   progress = 100.0%
dimension =   1   mpirank =    2   progress = 100.0%
dimension =   1   mpirank =    6   progress = 100.0%
dimension =   1   mpirank =    3   progress = 100.0%
dimension =   1   mpirank =    4   progress = 100.0%
dimension =   1   mpirank =    5   progress = 100.0%
dimension =   1   mpirank =    7   progress = 100.0%
dimension =   1   mpirank =    8   progress = 100.0%
dimension =   1   mpirank =   10   progress = 100.0%
dimension =   1   mpirank =   11   progress = 100.0%
dimension =   1   mpirank =   14   progress = 100.0%
dimension =   1   mpirank =    9   progress = 100.0%
dimension =   1   mpirank =   12   progress = 100.0%
dimension =   1   mpirank =
  1D descriptor/model(y=sum(ci*di)+c0) :
================================================================================
     d001 = (ETS/IPS)    feature_ID:000001

        coeff.(ci):     0.7944894697E+01
                c0:    -0.2091604189E+00
    RMSE and MaxAE:     0.1643216197E+00    0.6685446201E+00
==============================
SISSO done successfully!

=======================
Time (second) used for this DI:            0.02
Total time (second):            0.13
Have a nice day !

使用SISSO_predict读取上述文件:

dimension: 1 feature: forrtl: severe (24): end-of-file during read, unit 1, file SISSO.out Image PC Routine Line Source SISSO_predict 0000000000428536 Unknown Unknown Unknown SISSO_predict 0000000000407275 Unknown Unknown Unknown SISSO_predict 000000000040474D Unknown Unknown Unknown libc-2.17.so 0000153C7A7E5555 __libc_start_main Unknown Unknown SISSO_predict 0000000000404676 Unknown Unknown Unknown

rouyang2017 commented 3 days ago

你是对的,之前的没更新。回头我更新下