#819 closed doubt (fixed)
Error message in BSR estimate
Reported by: | imendez | Owned by: | Jorge |
---|---|---|---|
Priority: | high | Milestone: | Mantainance |
Component: | BSR | Version: | 2.0.1 |
Severity: | major | Keywords: | |
Cc: |
Description
Hi, BSR returns this error message:
[BSR.Parse] Parsing BSR file C:/users/imendez/bsd/entity/PrjBysForMeMoPr/data/output/Modelo_CEN02.ModeloEMU20091217_AllInputs_BSRHLM_y2009m12d17h16i44s03.00.bsr with 1625578 bytes <E> ERROR: [24] [BSR.Parse] Unexpected symbol not matching none of BSR (Bayesian Sparse Regression) declaration syntax options: Linear block variable declaration with initial values: my_linear_block_name <- number; Missing block variable declaration with initial values: my_linear_block_name ? <- number; Missing block variable declaration with prior info: my_linear_block_name ? ~ {Normal(nu,sigma^2)|TruncatedNormal(nu,sigma^2,minBound,maxBound)}; White noise distribution declaration with known homocedastic diagonal covariance: my_noise_block_name[length] ~ Normal(0,sigma^2); White noise distribution declaration with known heterocedastic diagonal covariance: my_noise_block_name[length] ~ Normal(0,SetDiag([[sigma_1^2, sigma_2^2, ...]])); White noise distribution declaration with unknown variance: my_noise_block_name[length] ~ Normal(0,my_sigma_block_name); ARMA noise distribution declaration with unknown variance: my_noise_block_name[length] ~ Normal(0,my_sigma_block_name * ARMA.COV([[ARIMAStruct(period,ar,ma,dif),...]])); Linear regressing equation: {number|?} == my_noise_block_name[index] {+|-}{{number|missing_block_name}*}my_linear_block_name {+|-}{{number|missing_block_name}*}my_linear_block_name ... ; Linear constraining inequation number {<=|>=} {+|-}{{number}*}my_linear_block_name {+|-}{{number}*}my_linear_block_name ... ; File:'C:/users/imendez/bsd/entity/PrjBysForMeMoPr/data/output/Modelo_CEN02.ModeloEMU20091217_AllInputs_BSRHLM_y2009m12d17h16i44s03.00.bsr' Line:113 Col:83 Puede ver un ejemplo de la sintaxis de BSR en http://cvs.tol-project.org/viewcvs.cgi/*checkout*/tol/stdlib/math/stat/models/bayesian/bysMcmc/test_01/parse.bsr</E> [Call stack] [3] Set StdLib::BysMcmc::Bsr::Gibbs::EstimAscii (Text ascii_, Set nonLinFilter, NameBlock config, NameBlock notifier) [2] Set StdLib::BysMcmc::Bsr::Gibbs::EstimAndNotify (Anything modelDef, NameBlock config, NameBlock notifier) [1] Set StdLib::BysMcmc::Bsr::DynHlm::DBApi::Estim (Text dbName, Text model, Text session, Text resultRootPath, NameBlock seriesHandler, NameBlock config_) <E>
However, the file seems to be correct at the indicated point (Line:113 Col:83). I attach .log and .bsr files.
Regards.
Attachments (1)
Change History (3)
Changed 15 years ago by
Attachment: | Ticket#819.7z added |
---|
comment:1 Changed 15 years ago by
Resolution: | → fixed |
---|---|
Status: | new → closed |
comment:2 Changed 15 years ago by
Component: | Database → BSR |
---|---|
Milestone: | → Manteinance |
Version: | 1.1.7 → 2.0.1 |
Note: See
TracTickets for help on using
tickets.
En alguna parte parece que hay una limitación de 100 caracteres para los identificadores por lo que está cortando los nombres. En esa línea concreta aparece unidentificador que acaba en
:
porque ha sido recortado.Ahora mismo no sé exactamente dónde se pone ese límite pero a mí no me parece nada malo que lo haya pues un identificador de 100 caracteres ya es algo absolutamente ilegible para un humano medio.