如何使用新的预测替代方案修改 apollo_prediction() 和函数 apollo_probabilities() 的参数？

如何解决如何使用新的预测替代方案修改 apollo_prediction() 和函数 apollo_probabilities() 的参数？

我有从 apollo_estimate() 估计的基本模式选择模型

    model = apollo_estimate(apollo_beta,apollo_fixed,apollo_probabilities,apollo_inputs)

在预测步骤中，我想添加一个或多个备选方案并使用此估计模型进行预测。以下是我遵循的步骤。

使用替代数据更新数据库（数据集）：

data.list <- lapply(1,function(x) {
 nrep <- 1;
 cost_bike <- c(replicate(nrep,(runif(1000,100,300))));
 time_bike <- c(replicate(nrep,0.5,17.4))));
 access_bike <- c(replicate(nrep,5,50))));
 data.frame(cost_bike,time_bike,bike_access);
});
database$time_bike <-  as.numeric(as.character(trunc(data.frame(data.list[[1]])[,1])))
database$cost_bike <-  as.numeric(as.character(trunc(data.frame(data.list[[1]])[,2])))
database$bike_access <- as.numeric(as.character(trunc(data.frame(data.list[[1]])[,3])))

更新 apollo_probabilities

apollo_probabilities=function(apollo_beta,apollo_inputs,functionality="prediction"){
  
  ### Attach inputs and detach after function exit
  apollo_attach(apollo_beta,apollo_inputs)
  on.exit(apollo_detach(apollo_beta,apollo_inputs))
  
  ### Create list of probabilities P
  P = list()
  
  ### List of utilities: these must use the same names as in mnl_settings,order is irrelevant
  V = list()
  V[['car']]  = asc_car  + b_tt_car  * time_car                           
  V[['bus']]  = asc_bus  + b_tt_bus  * time_bus  + b_access * access_bus  
  V[['air']]  = asc_air  + b_tt_air  * time_air  + b_access * access_air   
  V[['rail']] = asc_rail + b_tt_rail * time_rail + b_access * access_rail
  V[['bike']] = b_access * access_bike # newly added
 
  
  ### Define settings for MNL model component
  mnl_settings = list(
    alternatives  = c(car=1,bus=2,air=3,rail=4,bike=5),avail         = list(car=av_car,bus=av_bus,air=av_air,rail=av_rail),choiceVar     = choice,V             = V
  )
  
  ### Compute probabilities using MNL model
  P[['model2']] = apollo_mnl(mnl_settings,functionality)
  
  ### Take product across observation for same individual
  P = apollo_panelProd(P,functionality)
  
  ### Prepare and return outputs of function
  P = apollo_prepareProb(P,functionality)
  return(P)
}


apollo_inputs   = apollo_validateInputs()

# prediction on new alternatives
predictions_new = apollo_prediction(model,prediction_settings = "model2")

此步骤因参数类型无效而失败。追溯：

Error in !a : invalid argument type 
6.
apollo_setRows(v,!a,0) 
5.
(function (v,a) 
apollo_setRows(v,0))(dots[[1L]][[5L]],dots[[2L]][[5L]]) 
4.
mapply(function(v,a) apollo_setRows(v,0),V,avail,SIMPLIFY = FALSE) 
3.
apollo_mnl(mnl_settings,functionality) 
2.
apollo_probabilities(apollo_beta,functionality = "prediction") 
1.
apollo_prediction(model,prediction_settings = "model2")

如何在 apollo 中添加新的替代方案并运行预测？