wrongu

Clarifying comments in model_base, namely 'set'-->'appended' parameters

1

#### Submission Checklist - [x] Run unit tests: `./runTests.py src/test/unit` - [x] Run cpplint: `make cpplint` - [ ] Declare copyright holder and open-source license: see below #### Summary Minor...

Create a JSON Protocol

6

The SC format is temporary. I believe the consensus is that JSON is a good direction to take it. Use this issues thread to start thinking about the actual JSON...

Graphics pipeline could be more efficient + more flexible

14

I wrote up my thoughts on it [here](https://github.com/wrongu/Sigma/wiki/Graphics-Pipeline-%28Current-and-Proposed%29) EDIT: turns out I pretty much just described a scene graph.

__Description of the bug:__ The `partial_credit` and `leaderboard` decorators are class-based and use a class-based decorator pattern with `@functools.wraps` around an inner decorator method, along with a local callback function...

Make tests deterministic from seed

7

Add conftest.py and use it to set a random seed for all pytest tests. This PR makes pytest deterministic and solves the problem of tests randomly failing due to unlucky...

I created a local test suite that tries to run `bayes_kit.HMCDiag(..., stepsize=0.01, steps=50)` and `bayes_kit.MALA(epsilon=0.01)` on suite of problems from posteriordb. I am just testing that outputs are not NaN...

Expose instantiator API

2

## 🚀 Feature request Expose instantiator logic outside of `jsonargparse`: ```python from jsonargparse import instantiate spec = { "class_path": "foo.Bar", "init_args": {"a": 1, "b": 2, "c": 3} } my_bar =...

there is still some appreciable bias with large flattened conv layers and `m=1000`. Idea: - we keep `m` moderate for geometric calculations - ...but initially estimating what the top `p`...