I’m testing out some examples in the examples folder on cmdstanr and this `multi_probit_good.stan`

does not like the input format

Example is at example-models/probit-multi-good.stan at master · stan-dev/example-models · GitHub

Specifically it does not like `y`

as matrix, as an array, or even using `structure(c(...), .Dim = c(500, 4))`

```
library(MASS)
K <- 8;
D <- 4;
N <- 500;
# stick breaking construction to generate a random correlation matrix
L_Omega <- matrix(0,D,D);
L_Omega[1,1] <- 1;
for (i in 2:D) {
bound <- 1;
for (j in 1:(i-1)) {
L_Omega[i,j] <- runif(1, -sqrt(bound), sqrt(bound));
bound <- bound - L_Omega[i,j]^2;
}
L_Omega[i,i] <- sqrt(bound);
}
Omega <- L_Omega %*% t(L_Omega);
x <- matrix(rnorm(N * K, 0, 1), N, K);
beta <- matrix(rnorm(D * K, 0, 1), D, K);
z <- matrix(NA, N, D);
for (n in 1:N) {
z[n,] <- mvrnorm(1, x[n,] %*% t(beta), Omega);
}
z <- matrix(NA, N, D);
for (n in 1:N) {
z[n,] <- mvrnorm(1, x[n,] %*% t(beta), Omega);
}
y <- matrix(0, N, D);
# or y <- array(0, dim = c(N, D));
for (n in 1:N) {
for (d in 1:D) {
y[n,d] <- ifelse((z[n,d] > 0), 1, 0);
}
}
mod_probit <- cmdstan_model("probit_multi_good.stan", force_recompile = T)
fit_probit <- mod_probit$sample(
data = list(K = K,
D = D,
N = N,
y = y,
x = x),
step_size = 0.01,
adapt_delta = 0.99
)
```

Error is dims found = (2000)

```
Chain 1 Exception: mismatch in number dimensions declared and found in context; processing stage=data initialization; variable name=y; dims declared=(500,4); dims found=(2000) (in '/var/folders/30/ddxkfs1x7bscp4_99c4db5jc0000gn/T/RtmpPA33tI/model-8e9526e504c2.stan', line 5, column 2 to column 30)
```