Michael Deistler
Michael Deistler
Closing this due to inactivity. Feel free to re-open if concrete questions come up.
We will move to flowtorch after the hackathlon
Hi Francois, thanks for the pointer, and we are indeed on the lookout for a good repo for flows. We'll have a look! Michael PS: amazing package name and logo
I'm thinking of sth like this: ```python from sbi.neural_nets.preconfigured_embeddings import CNNEmbedding from sbi.utils import posterior_nn cnn_embedding = CNNEmbedding(num_layers=5, num_channels=[1, 3, 5, 5, 5]) density_estimator = posterior_nn("maf", embedding_net=cnn_embedding) inference = SNPE(prior=prior,...
Maybe add one fully connected layer.
Removing the `hackathlon` label because we will deal with this after porting to flowtorch
Removing the label because we will do this after the flowtorch port
Let's keep this open until the `__getstate__` is merged.
No, not fixed yet. SNPE-B raises a `NotImplementedError`.