Summary/Context I am attempting to deploy a model from R to Google Cloud.
I am following these steps/tutorial from the vetiver GitHub.
I am using the sample data provided in the tutorial except I am attempting to deploy the model to Google Cloud, see code & error below:
library(parsnip)
library(workflows)
library(vetiver)
library(pins)
data(Sacramento, package = "modeldata")
rf_spec <- rand_forest(mode = "regression")
rf_form <- price ~ type + sqft + beds + baths
rf_fit <-
workflow(rf_form, rf_spec) %>%
fit(Sacramento)
v <- vetiver_model(rf_fit, "sacramento_rf")
v
model_board <- board_register_gcloud( name = "lead-time-model"
,bucket = "pro_lead_time_prediction_bucket" )
model_board
The error I get when I run model_board is:
Error in order(results$name) : argument 1 is not a vector
Part of my confusion is that my vetiver model object ran without issue:
v
-- sacramento_rf - <bundled_workflow> model for deployment
A ranger regression modeling workflow using 4 features
Other error:
model_board %>% pin(v)
Error in path_tidy(.Call(fs_path_, lapply(args, function(x) enc2utf8(as.character(x))), :
Total path length must be less than PATH_MAX: 260
Could the problem stem from authenticating or not connecting correctly to gcloud? I have SDK installed and authenticated as well.
