Comments (2)
I agree that the notation is a bit weird, that was basically the motivation for playing around with #64 (but it's a bit outdated now). The approach there was to use the fixed variable names _varinfo
(or _vi
), _model
, _sampler
, and _context
, so one has access to everything during model evaluation and could write in the model definition
_vi.logp[] += logsumexp(gamma)
or
acclogp!(_vi, logsumexp(gamma))
IMO it would be nice to reduce the use of macros and just use standard Julia syntax, whenever possible and reasonable.
from dynamicppl.jl.
It makes sense to reduce macros in the model block. Ideally @model
should be the only macro exposed to users from Turing.
from dynamicppl.jl.
Related Issues (20)
- @model macro breaks on conditional branches with return values HOT 2
- MCMCChains for Julia < 1.9
- `logprior`, `loglikelihood`, `logjoint` are not working with `MCMCChains` in Julia 1.9 HOT 1
- Derived variables from data on the LHS of tilde HOT 10
- Decouple from Distributions.jl HOT 8
- ThreadSafeVarInfo(::SimpleVarInfo) bug with ForwardDiff HOT 1
- Need alternative to `NamedTuple` for `SimpleVarInfo` HOT 12
- Storing returned values in `VarInfo`.
- `BangBang.setindex!!(vi::SimpleVarInfo{<:AbstractDict}, val, vn::VarName)` returns eltype Any arrays when `vn` is multidimensional HOT 1
- Remove `tonamedtuple`
- Can't use `@info` in models HOT 1
- Move TestUtils to an extension
- `SimpleVarInfo` sometimes returns a `SimpleVarInfo` with `Int`-type `logp` and creates down-stream errors HOT 1
- Roadmap for removal of `Selector` shenanigans HOT 2
- change defaults: @addlogprob! not included in PriorContext() HOT 5
- Type-instability when using type-parameters HOT 1
- Type instability in unflatten for medium / large tuples HOT 6
- DynamicPPL..jl tests 23 Broken. HOT 3
- Introduce a convenience function, `AutoType`, to hide `TypeWrap`.
- Issue with ReverseDiff and undef on Julia v1.7 HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from dynamicppl.jl.