-
One thing I've noticed in playing with implementations of two new PushGP problems (Smallest and Median) is that while the individual errors on individual training cases fit in a `i64`, adding up 100 o…
-
Add a third representation: PushGP
* http://helios.hampshire.edu/lspector/pubs/ace.pdf
* https://dl.acm.org/doi/pdf/10.1145/3067695.3082468
-
(*After some consideration, I think that instead of implementing genetic algorithm on a linked graph, as [this "Evolving Neural Networks through Augmenting Topologies" implementation](https://github.c…
-
Once the evolutionary process has terminated, for instance, after you have called the `run` method of a `GeneticAlgorithm` object, you get an instance of the `Individual` class, which supposedly repre…
-
My question is in the title. After a quick search on the documentation page with the keywords "genetic programming", there does not seem to be any support for GP, which seems to be confirmed by the Fu…
-
Opening up this issue to have a place for discussion - in case I miss out any libraries.
The current list is ...
## Clojure
#### Natural Language Processing
* [Clojure-openNLP](https:/…
-
Programs in output look something like this:
```
Program(signature=ProgramSignature(output_stacks=['str'], arity=1, push_config=PushConfig(numeric_magnitude_limit=1000000000000.0, step_limit=500, …
-
I feel like the `number` function's name has two major problems:
1. It's entirely unclear what it does.
2. It makes me worry that it will collide with the same name from somewhere else (recall the…
-
Starting a fresh repl in my project (which depends on clojush 3.19.0) and inputting the following excerpt from the tutorial worksheet throws an exception:
```
(use 'clojush.ns)
(use-clojush)
(pu…
-
When running a modified UBall5D example on a training set with around 4000 test cases of the form:
```
[[0.0 58.992325]
[1.0 60.346346]
[2.0 71.436636]
[3.0 80.2523]
...
[4000.0 46.23525…