good first issuetf
仓库指标
- 星标
- (1,728 个星标)
- PR 合并指标
- (30 天内没有已合并 PR)
描述
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 |