-
Some of the examples are broken because of distribution interface changes. Identify and fix these broken examples.
-
The attend-infer-repeat ([AIR](https://arxiv.org/abs/1603.08575)) model is a great example of stochastic recursion in deep probabilistic models. It is going to be one of the anchor examples for the fi…
-
I was experimenting with the following model and came across some surprising behaviour. Running on [webppl.org](http://webppl.org) as of today:
```javascript
var dist = Uniform({a:0, b:1})
var gu…
-
As we build more advanced gradient estimators ( #84 ) and change particular algorithm interfaces to fit our needs (e.g. #74 and #32 and #104 ) we should make sure our abstractions are lightweight and …
-
this is useful for constructing marginals in particular.
it should be a very straightforward extension of Categorical (especially if implemented as a subclass?).
-
Suppose I am working with a sample-based inference result for a continuous distribution, e.g.:
```
var dist = Infer({method: 'SMC', particles: 10000}, function() {
var x = poisson(.5);
var y = ga…
-
Suppose I have a likelihood that can only be written as: `getLikelihood(x, callback)`. Specifically, I'd like to have a child process continually running, which I send data off to for likelihood evalu…
-
npm install -g webppl
sh: 1: node: not found
npm ERR! Linux 4.4.0-75-generic
npm ERR! argv "/usr/bin/nodejs" "/usr/bin/npm" "install" "-g" "webppl"
npm ERR! node v4.2.6
npm ERR! npm v3.5.2
np…
-
This program using [webppl-nn](github.com/null-a/webppl-nn) fails:
```js
var mlp = stack([
tanh,
affine('mlp', { "in": 2, "out": 1 })
]);
var model = function() {
var xs = par…
-
Using [`process.argv`](https://nodejs.org/docs/latest/api/process.html#process_process_argv) we can access additional command line arguments in webppl scripts. For example, with `test.wppl` only conta…