I added a new tutorial#141
Conversation
|
|
||
| } | ||
|
|
||
| for (j in 1:NUM_HIDDEN) { |
There was a problem hiding this comment.
The indentation seems a bit weird here.
| nStates = NUM_RATES) | ||
|
|
||
| # clamp the model with the "observed" tree | ||
| hisse.clamp(observed_phylogeny) |
There was a problem hiding this comment.
Did you mean hisse4.clamp( )? Right now the script will not run.
| moves.append(mvSlice(q_1A1B, window = 0.1, weight=2, search_method = "stepping_out")) | ||
| moves.append(mvSlice(q_1B1A, window = 0.1, weight=2, search_method = "stepping_out")) | ||
|
|
||
| # first initialize the matrix with all zeros: |
There was a problem hiding this comment.
The indentation seems wrong here.
| moves.append(mvSlide(extinction_beta[i],delta=0.20,tune=true,weight=2.0)) | ||
|
|
||
| } | ||
|
|
There was a problem hiding this comment.
Wrong indentation here also.
| rate_sd <- 2 * half_sd | ||
|
|
||
| # Create a normal distributed variable for the diversification and turnover rates | ||
|
|
| moves.append(mvSlice(extinction_alpha[i],window = 0.1, weight=2, search_method = "stepping_out")) | ||
| } | ||
|
|
||
| for (i in 1:(NUM_HIDDEN-1)) { |
|
|
||
| } | ||
|
|
||
| for (j in 1:NUM_HIDDEN) { |
|
|
||
| # first initialize the matrix with all zeros: | ||
| for (i in 1:NUM_RATES) { | ||
| for (j in 1:NUM_RATES) { |
| q[3][4] := q_0B1B | ||
| q[3][1] := q_0B0A | ||
| q[4][3] := q_1B0B | ||
| q[4][2] := q_1B1A |
There was a problem hiding this comment.
What about this? I'm not entirely sure it has exactly the same speed, but maybe it is easier to read? Up to you.
q := [ [0, q_0A1A, q_0A0B, 0 ],
[q_1A0A, 0, 0, q_1A1B],
[q_0B0A, 0, 0, q_0B0B],
[0, q_1B1A, q_1B0B, 0 ] ]
|
Hi Nicolas, thank you for the new tutorial! It would be really nice if we could add a test for this tutorial. To add a test, we would do something like:
The last part is a bit tricky, since you don't specify an iteration number for your mcmc.run command. |
|
Hi Ben! Thank you for your review! I fixed the indentation and added |
| moves.append(mvSlice(q_1D1C, window = 0.1, weight=2, search_method = "stepping_out")) | ||
|
|
||
| # first initialize the matrix with all zeros: | ||
| for (i in 1:NUM_RATES) { |
There was a problem hiding this comment.
The indentation is still wrong here
|
Hi Nicolas, some of the other comments are not addressed yet. If you add the current script to the tests without modifying it, then the tests will become very slow. So we need to figure out how to make the For other scripts, the |
3 participants
Hi, I added a tutorial entitled "Hypothesis testing of state-dependent diversification with a full HiSSE model" for the SSE section.