Feras Saad

Results 115 issues of Feras Saad

As it stands, conditional component models are created by using `View.update_cctype`, and the allowed inference operator are everything except column transitions. When a user invokes `View.update_cctype(a)` and then `View.update_cctype(b)`: -...

Algorithm 2: Particle learning from Grosse, R. _Model selection in compositional spaces_. Ph.D. thesis, 2014.

It seems to work on uncollapsed beta and truncated normal models, but the code is a mess.

Initializing a Dim without the regression metadata will crash (entry points are `State.incorporate_dim` and `State.__init__` with a conditional model. The ViewGPM is smart enough to take the metadata from its...

For instance, the posterior predictive of Normal-Gamma Normal is a Student t. Do we achieve identical simulation if we instead simulate rho ~ Gamma(Hn), mu ~ Normal(rho,Hn), then x ~...

Raises a `NotImplementedError` for the SPAux. ``` venture[script] > define gp = make_gp(gp_mean_const(0.), gp_cov_se(1)) venture[script] > gp([1,2]) /scratch/fsaad/Venturecxx/build/lib.linux-x86_64-2.7/venture/lite/psp.py in simulate(self, args) 598 599 def simulate(self, args): --> 600 return self._disptach(args).simulate(args)...

- [ ] Add prune to trace_set.py - [ ] Add prune to engine.py (four flavors) - [ ] Add prune to inference.py (four flavors) - [ ] Add prune...

Here is a sample trace, noting that `y` has dynamic integer scope and string block, whereas `z` has string scope and dynamic integer block. ``` $ cat case.vnts assume x...

``` ripl.execute_program('assume x1 = tag("x", "x", normal(0,1))') ripl.execute_program('assume y1 = tag("y", "y", normal(x1,1))') ripl.observe('y1', 2) ripl.execute_program('infer(mh("x", 1, 100))') ``` The final line crashes noisily since block `1` does not exist...