time indices, concatenated blocks with model_output with stats_only=FALSE about redm

[SoftwareLiteracy]

Thanks for illustrating these issues.

Since this was done on an old version (pre 1.7.5), and encapsulates multiple issues, I'm going to close it and reopen a new issue focusing on the library break predictiion output.

Please note it is recommended to use the new API, as it is consistent across implementations (C++, Python, R) and directly interfaces to cppEDM. Nonetheless, we certainly want to address the legacy (0.7.X) API compatibility.

I believe that the NaN alignment issue has been resolved. Here's output of the version 1.8

> block <- data.frame( time=1:10, x=sin((1:10)/pi), y=cos((1:10)/pi) )
> out <- block_lnlp( block, tp=2, columns=c("x","y"), target_column = "x", stats_only = FALSE )
> out $ model_output
   Index Observations Predictions Pred_Variance Const_Predictions
1      1      0.31296         NaN           NaN               NaN
2      2      0.59448         NaN           NaN               NaN
3      3      0.81627     0.96559     0.0003982           0.31296
4      4      0.95606     0.91359     0.0072469           0.59448
5      5      0.99978     0.92841     0.0024076           0.81627
6      6      0.94307     0.84251     0.0241079           0.95606
7      7      0.79160     0.79119     0.0389745           0.99978
8      8      0.56060     0.59761     0.0790172           0.94307
9      9      0.27328     0.34399     0.1140424           0.79160
10    10     -0.04149     0.39507     0.0318934           0.56060
11    11          NaN     0.09529     0.0413000           0.27328
12    12          NaN     0.17593     0.0557830          -0.04149