rlworkgroup/garage
Split GaussianMLPModel into several variants
Aberta
#587 aberto em 19 de mar. de 2019
good first issuetf
Métricas do repositório
- Stars
- (1.728 estrelas)
- Métricas de merge de PR
- (Nenhuma PRs mesclada em 30d)
Description
The current GaussianMLPModel actually represents several architecture variants, which makes it difficult to use and to test. We should split it into several variants based on architecture (and possibly parameterization)
| description | mean | std_param |
|---|---|---|
| fixed variance | MLP1 | parameter (trainable or not) |
| split network | MLP1 | MLP2 |
| shared network/multi-headed | MLP1/head1 | MLP1/head2 |