Expand arrays #8
Conversation
| N_age <- parameter() | ||
| dim(S0) <- N_age | ||
| dim(I0) <- N_age | ||
| dim(S) <- N_age | ||
| dim(I) <- N_age | ||
| dim(R) <- N_age | ||
| dim(n_SI) <- N_age | ||
| dim(n_IR) <- N_age | ||
| dim(p_SI) <- N_age | ||
| dim(m) <- c(N_age, N_age) | ||
| dim(s_ij) <- c(N_age, N_age) | ||
| dim(lambda) <- N_age |
There was a problem hiding this comment.
Wes' fixes here will make this nicer, we'll want to go back and add it
arrays.qmd
Outdated
|
|
||
| Of course you may want to use higher-dimensional arrays! We currently supporting up to 8 dimensions, with index variables `i`, `j`, `k`, `l`, `i5`, `i6`, `i7` and `i8`. | ||
|
|
||
| Be careful with array equations! It's possible that in an array equation you end up going out of bounds with an array used on the RHS. This can crash the program when running. |
There was a problem hiding this comment.
I've got some thoughts on improving this, but it's a way off
arrays.qmd
Outdated
|
|
||
| ```r | ||
| m_col_totals[] <- sum(m[, i]) | ||
| ``` |
There was a problem hiding this comment.
The other reduction functions that might be useful here are prod, min and max, all of which work the same way
There was a problem hiding this comment.
Added a note about those
| # Force of infection | ||
| m <- parameter() # age-structured contact matrix | ||
| s_ij[, ] <- m[i, j] * sum(I[j, ]) | ||
| lambda[] <- beta * sum(s_ij[i, ]) |
There was a problem hiding this comment.
We should spell out in the text that this is a matrix multiplication, and give the appropriate R code (lambda <- beta * (m %*% I) I think) and say that we're looking to simplify this in future?
There was a problem hiding this comment.
Added a note about this!
arrays.qmd
Outdated
|
|
||
|
|
||
| ## Summing over arrays | ||
| Frequently, you will want to take a sum over an array, or part of an array, using `sum`. To sum over all elements of an array, use `sum()` with the name of the array you would like to sum over: |
There was a problem hiding this comment.
This section comes after the first bit of array modelling, so suggest a reordering? Perhaps build up to the SIR model you have by writing a simple model (e.g., something like SIR by hand for two cases, manually summing for the FOI, then merging, or even simpler two logistic growth curves then introduce the idea of summing)
Co-authored-by: Rich FitzJohn <r.fitzjohn@imperial.ac.uk>
No description provided.