Package can't be installed

Hi there,

Last time I used this package it was still on cran. I tried to install from here but not having any luck. The error is also kinda non descriptive. Any advice or help?

Bayesmodels Project Roadmap

Future Work

Add more models

Error in library(bayesmodels)

Session Information:

> sessionInfo()
R version 4.0.3 (2020-10-10)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 19041)

Matrix products: default

locale:
[1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United States.1252   
[3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C                          
[5] LC_TIME=English_United States.1252    

attached base packages:
[1] parallel  stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] bsts_0.9.6                 BoomSpikeSlab_1.2.4        Boom_0.9.7                
 [4] MASS_7.3-54                bayesforecast_0.0.1        garchmodels_0.1.1         
 [7] rmgarch_1.3-7              rugarch_1.4-4              modeltime.h2o_0.1.1       
[10] h2o_3.32.1.3               modeltime.ensemble_0.4.1   modeltime.resample_0.2.0  
[13] tidyquant_1.0.3            quantmod_0.4.18            TTR_0.24.2                
[16] PerformanceAnalytics_2.0.4 xts_0.12.1                 zoo_1.8-9                 
[19] janitor_2.1.0              DBI_1.1.1                  odbc_1.3.2                
[22] timetk_2.6.1               lubridate_1.7.10           forcats_0.5.1             
[25] stringr_1.4.0              readr_1.4.0                tidyverse_1.3.1           
[28] modeltime_0.6.0            yardstick_0.0.8            workflowsets_0.0.2        
[31] workflows_0.2.2            tune_0.1.5                 tidyr_1.1.3               
[34] tibble_3.1.2               rsample_0.1.0              recipes_0.1.16            
[37] purrr_0.3.4                parsnip_0.1.6              modeldata_0.1.0           
[40] infer_0.5.4                ggplot2_3.3.3              dplyr_1.0.6               
[43] dials_0.0.9                scales_1.1.1               broom_0.7.6               
[46] tidymodels_0.1.3           pacman_0.5.1              

loaded via a namespace (and not attached):
  [1] utf8_1.2.1                  ks_1.13.1                   tidyselect_1.1.1           
  [4] htmlwidgets_1.5.3           grid_4.0.3                  gmp_0.6-2                  
  [7] pROC_1.17.0.1               munsell_0.5.0               codetools_0.2-18           
 [10] xgboost_1.4.1.1             future_1.21.0               withr_2.4.2                
 [13] Brobdingnag_1.2-6           colorspace_2.0-1            rstudioapi_0.13            
 [16] stats4_4.0.3                bayesplot_1.8.0             listenv_0.8.0              
 [19] labeling_0.4.2              GeneralizedHyperbolic_0.8-4 rstan_2.21.2               
 [22] TeachingDemos_2.12          DistributionUtils_0.6-0     bit64_4.0.5                
 [25] DiceDesign_1.9              farver_2.1.0                bridgesampling_1.1-2       
 [28] coda_0.19-4                 parallelly_1.25.0           vctrs_0.3.8                
 [31] generics_0.1.0              ipred_0.9-11                R6_2.5.0                   
 [34] LICHospitalR_0.2.0          bitops_1.0-7                lhs_1.1.1                  
 [37] assertthat_0.2.1            nnet_7.3-16                 forecast_8.15              
 [40] gtable_0.3.0                globals_0.14.0              processx_3.5.2             
 [43] timeDate_3043.102           rlang_0.4.11                Bessel_0.6-0               
 [46] splines_4.0.3               lazyeval_0.2.2              earth_5.3.0                
 [49] SkewHyperbolic_0.4-0        inline_0.3.19               yaml_2.2.1                 
 [52] modelr_0.1.8                crosstalk_1.1.1             backports_1.2.1            
 [55] tools_4.0.3                 lava_1.6.9                  ellipsis_0.3.2             
 [58] ff_4.0.4                    Rsolnp_1.16                 ggridges_0.5.3             
 [61] Rcpp_1.0.6                  plyr_1.8.6                  RCurl_1.98-1.3             
 [64] ps_1.6.0                    prettyunits_1.1.1           rpart_4.1-15               
 [67] fracdiff_1.5-1              haven_2.4.1                 fs_1.5.0                   
 [70] furrr_0.2.2                 magrittr_2.0.1              data.table_1.14.0          
 [73] lmtest_0.9-38               reprex_2.0.0                GPfit_1.0-8                
 [76] truncnorm_1.0-8             mvtnorm_1.1-1               matrixStats_0.59.0         
 [79] hms_1.1.0                   mclust_5.4.7                readxl_1.3.1               
 [82] rstantools_2.1.1            gridExtra_2.3               compiler_4.0.3             
 [85] KernSmooth_2.23-20          V8_3.4.2                    crayon_1.4.1               
 [88] StanHeaders_2.21.0-7        htmltools_0.5.1.1           corpcor_1.6.9              
 [91] pcaPP_1.9-74                Formula_1.2-4               RcppParallel_5.1.4         
 [94] dbplyr_2.1.1                Matrix_1.3-4                cli_2.5.0                  
 [97] quadprog_1.5-8              gower_0.2.2                 pkgconfig_2.0.3            
[100] spd_2.0-1                   numDeriv_2016.8-1.1         plotly_4.9.3               
[103] xml2_1.3.2                  foreach_1.5.1               hardhat_0.1.5              
[106] plotmo_3.6.0                prodlim_2019.11.13          rvest_1.0.0                
[109] snakecase_0.11.0            callr_3.7.0                 digest_0.6.27              
[112] pracma_2.3.3                cellranger_1.1.0            curl_4.3.1                 
[115] urca_1.3-0                  nloptr_1.2.2.2              lifecycle_1.0.0            
[118] nlme_3.1-152                jsonlite_1.7.2              tseries_0.10-48            
[121] viridisLite_0.4.0           fansi_0.5.0                 pillar_1.6.1               
[124] loo_2.4.1                   lattice_0.20-44             plotrix_3.8-1              
[127] httr_1.4.2                  pkgbuild_1.2.0              survival_3.2-11            
[130] glue_1.4.2                  iterators_1.0.13            bit_4.0.4                  
[133] class_7.3-19                stringi_1.6.2               prophet_1.0                
[136] Quandl_2.10.0               blob_1.2.1                  Rmpfr_0.8-4

Error Message:

> library("bayesmodels")
Error: package or namespace load failed for ‘bayesmodels’ in get(method, envir = home):
 lazy-load database 'C:/Users/bha485/Documents/R/win-library/4.0/bayesmodels/R/bayesmodels.rdb' is corrupt
In addition: Warning messages:
1: package ‘bayesmodels’ was built under R version 4.0.5 
2: In .registerS3method(fin[i, 1], fin[i, 2], fin[i, 3], fin[i, 4],  :
  restarting interrupted promise evaluation
3: In get(method, envir = home) :
  restarting interrupted promise evaluation
4: In get(method, envir = home) : internal error -3 in R_decompress1

The package can't be installed

Hi, I am trying to install the bayesmodels package with devtools::install_github("AlbertoAlmuinha/bayesmodels")

I get the following error:

Error: package or namespace load failed for ‘bayesmodels’:
.onLoad failed in loadNamespace() for 'bayesmodels', details:
call: check_mode_for_new_engine(model, eng, mode)
error: 'regression' is not a known mode for model gen_additive_reg().
Error: loading failed
Execution halted

It seems that the parsnip defined modes are not properly passed during the installation.

Any idea how to fix that?

Session info:

R version 4.2.0 (2022-04-22)
Platform: x86_64-apple-darwin17.0 (64-bit)
Running under: macOS Monterey 12.2.1

Matrix products: default
LAPACK: /Library/Frameworks/R.framework/Versions/4.2/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] parsnip_0.2.1 rlang_1.0.2

loaded via a namespace (and not attached):
[1] tidyselect_1.1.2 remotes_2.4.2 purrr_0.3.4 colorspace_2.0-3
[5] vctrs_0.4.1 generics_0.1.2 testthat_3.1.4 usethis_2.1.5
[9] utf8_1.2.2 pkgbuild_1.3.1 pillar_1.7.0 glue_1.6.2
[13] withr_2.5.0 DBI_1.1.2 sessioninfo_1.2.2 foreach_1.5.2
[17] lifecycle_1.0.1 munsell_0.5.0 gtable_0.3.0 devtools_2.4.3
[21] codetools_0.2-18 evaluate_0.15 memoise_2.0.1 callr_3.7.0
[25] fastmap_1.1.0 doParallel_1.0.17 ps_1.7.0 parallel_4.2.0
[29] curl_4.3.2 fansi_1.0.3 scales_1.2.0 cachem_1.0.6
[33] desc_1.4.1 pkgload_1.2.4 fs_1.5.2 brio_1.1.3
[37] ggplot2_3.3.6 png_0.1-7 digest_0.6.29 processx_3.5.3
[41] dplyr_1.0.9 grid_4.2.0 rprojroot_2.0.3 hardhat_0.2.0
[45] cli_3.3.0 tools_4.2.0 magrittr_2.0.3 tibble_3.1.7
[49] crayon_1.5.1 tidyr_1.2.0 pkgconfig_2.0.3 ellipsis_0.3.2
[53] prettyunits_1.1.1 assertthat_0.2.1 rstudioapi_0.13 iterators_1.0.14
[57] R6_2.5.1 compiler_4.2.0

Formula Interface: causes loss of Transformation Information

Calibration needs to be able to determine if transformations were applied.

Cannot determine if transformation is required on 'actual_data'

Problem:

library(bayesmodels)

library(tidymodels)

library(timetk)

library(modeltime)

library(modeltime.resample)

library(modeltime.ensemble)



data(iclaims)



names(initial.claims)



df <- timetk::tk_tbl(initial.claims)



df %>% plot_time_series(.date_var = index,
                        
                        .value = iclaimsNSA,
                        
                        .smooth = FALSE)





# Issue 1

# split

splits <- time_series_split(
    
    data = df,
    
    # date_var = 'date',
    
    assess     = 52,
    
    cumulative = TRUE
    
)


# splits %>% tk_time_series_cv_plan() %>% plot_time_series_cv_plan(index, iclaimsNSA)


ss <- AddLocalLinearTrend(list(), training(splits)$iclaimsNSA)

ss <- AddSeasonal(ss, training(splits)$iclaimsNSA, nseasons = 52)



modelo <- bayesian_structural_reg() %>%
    
    set_engine("stan", state.specification = ss, niter = 1000) %>%
    
    fit(iclaimsNSA ~ index, data = training(splits))



modeltime_tbl <- modeltime_table(modelo)


calib_tbl <- modeltime_table(modelo) %>% modeltime_calibrate(testing(splits))

Results in this:

Warning message:
Problem with `mutate()` column `.nested.col`.
i `.nested.col = purrr::map2(...)`.
i Cannot determine if transformation is required on 'actual_data'

Date Dependence in Bayesian Structural Models

There is a bug where the date is being included in the formula when calling the bsts::bsts() model. This behavior should be corrected.

`sarima_reg` parameters

@AlbertoAlmuinha Is it mandatory that the seasonal parameters be known beforehand for the sarima_reg or will the algorithm take what you give and adjust?

Problems predicting univariate state space models

A problem has been detected when calculating predictions in the bayesian_structural_reg() function for univariate problems. Actually this function was only working properly for problems where external regressors were used in the model.

Bayesmodels with newer versions of R

Hello Alberto, first of all thanks for the amazing work with the package.

I would like to know if there is any prospects for the package to be usable with the newer versions of R. Thanks

cant activate package

Hello Alberto. Suddenly the package has stopped working. When trying to load the package it returns :

Error: package or namespace load failed for ‘bayesmodels’:
.onLoad failed in loadNamespace() for 'bayesmodels', details:
call: check_model_doesnt_exist(model)
error: Model sarima_reg already exists

Unable to install the package

I am just trying to install your package with no success. I did try on mac and on windows 10 under R version 4.1.1 and 4.1.2. When I try to install via cran it states, that this package is not available for my version of R.

When I try to install the development version, I get the following error:

Restarting R did not help.

Bug in Bayesian Structural Reg Univariate Predictions

When there are no external regressors and the variable has no missing values, the "if" condition must be changed so that it does not go through the second condition. This causes that the predictions are not well calculated and a bug is produced.

bayesmodels::exponential_smoothing() runs forever...

Ubuntu LTS, R latest, boostime latest, modeltime latest.
exponential_smoothing() runs forever...
Is there a way to reach rlgt.control() to get access to iterations etc.?
I tried set_engine('stan',iter=100) with no success.

Please don't give up with maintaining this package! It has great potential, since the bayes-thingy can be a nightmare manually.

Loading error

Hello @AlbertoAlmuinha, I have installed bayesmodels from CRAN. But getting the following error when loading it.


library(bayesmodels)
Loading required package: parsnip
Loading required package: bayesforecast
Registered S3 method overwritten by 'quantmod':
  method            from
  as.zoo.data.frame zoo 
Registered S3 methods overwritten by 'bayesforecast':
  method      from    
  autoplot.ts forecast
  forecast.ts forecast
  fortify.ts  forecast
  print.garch tseries 
Error: package or namespace load failed for ‘bayesforecast’ in .doLoadActions(where, attach):
 error in load action .__A__.1 for package bayesforecast: Rcpp::loadModule(module = "stan_fit4SVM_mod", what = TRUE, env = ns, : Unable to load module "stan_fit4SVM_mod": function 'Rcpp_precious_remove' not provided by package 'Rcpp'
Error: package ‘bayesforecast’ could not be loaded
In addition: Warning messages:
1: package ‘bayesmodels’ was built under R version 4.1.1 
2: package ‘bayesforecast’ was built under R version 4.1.1

Predictions: Only one predicted value being returned when should be length(testing(split))

I'm pretty sure this is related to Issue #8 - Something weird happens when using the formula interface. This is another reason I prefer the data.frame interface when developing parsnip functions.

Problem

Only 1 predicted value is being returned. Should be 52 predictions. - Again, I think the solution can be resolved by switching to data.frame interface as discussed in #8.

predict(modelo, testing(splits)) # will return only 1 predicted value
# A tibble: 1 x 1
  .pred
  <dbl>
1 0.220

Reproducible Example

library(bayesmodels)

library(tidymodels)

library(timetk)

library(modeltime)

library(modeltime.resample)

library(modeltime.ensemble)



data(iclaims)



names(initial.claims)



df <- timetk::tk_tbl(initial.claims)



df %>% plot_time_series(.date_var = index,
                        
                        .value = iclaimsNSA,
                        
                        .smooth = FALSE)





# Issue 1

# split

splits <- time_series_split(
    
    data = df,
    
    # date_var = 'date',
    
    assess     = 52,
    
    cumulative = TRUE
    
)


# splits %>% tk_time_series_cv_plan() %>% plot_time_series_cv_plan(index, iclaimsNSA)


ss <- AddLocalLinearTrend(list(), training(splits)$iclaimsNSA)

ss <- AddSeasonal(ss, training(splits)$iclaimsNSA, nseasons = 52)



modelo <- bayesian_structural_reg() %>%
    
    set_engine("stan", state.specification = ss, niter = 1000) %>%
    
    fit(iclaimsNSA ~ index, data = training(splits))



modeltime_tbl <- modeltime_table(modelo)


calib_tbl <- modeltime_table(modelo) %>% modeltime_calibrate(testing(splits))



a <- calib_tbl %>%
    
    modeltime_forecast(
        
        new_data = testing(splits),
        
        actual_data = training(splits),
        
    )



# make the values NA in the test split

testing_tmp <- testing(splits)[c('index', 'iclaimsNSA')]

testing_tmp$iclaimsNSA <- NA



calib_tbl2 <- modeltime_tbl %>%
    
    modeltime_calibrate(new_data = testing_tmp,
                        
                        actural_data = training(splits))



a2 <- calib_tbl %>%
    
    modeltime_forecast(
        
        new_data = testing_tmp,
        
        actual_data = training(splits),
        
    )



# compare Tables a and a2: Table a has only one prediction value for all timestaps in the test split but a2 has different predicted values in testing_tmp



predict(modelo, testing(splits)) # will return only 1 predicted value

Bayesmodels Models Roadmap

Actual Models

Sarima: bayesforecast package
Garch: bayesforecast package
Random Walk (Naive): bayesforecast package
Stochastic Volatility: bayesforecast package
State Space Model: bayesforecast package
Adaptive Spline Surface: BASS package
GAMs: brms package
Structural Time Series (State Space Models): bsts package
Exponential Smoothing: Rlgt package

Future Models

Additive Regression Trees: BART package (no rJava dependencies)

albertoalmuinha / bayesmodels Goto Github PK

bayesmodels's Introduction

bayesmodels

Tutorials

Installation

Why Bayesmodels?

bayesmodels's People

Contributors

Stargazers

Watchers

Forkers

bayesmodels's Issues