jnsbck

Results 22 comments of jnsbck

So I ran pytest --durations=0 to determine which tests to prioritise and this was the output: [pytest_output.txt](https://github.com/mackelab/sbi/files/7931986/pytest_output.txt) The main two takeaways: 1. linearGaussian_snle_test.py takes by far the longest (see top...

Made a quick 1st draft for an RNN embedding net. ```python class DefaultEmbeddingRNN(nn.Module): """Preconfigured RNN for embedding of sequence data. Attributes: n_layers: Number of rnn layers. rnn: RNN layers. fc:...

Same here. In my case I was using `Evolutionary.jl` with [`Optimization.jl`](https://docs.sciml.ai/Optimization/stable/optimization_packages/evolutionary/), where you have to set a `x0` when defining the `OptimizationProblem`. Got it to work by initializing with an...

Yea of course. Was planning to continue working on it. Left the desc blank since its in draft state. Yes, can rename it to `MixtureDensityEstimator`. :) EDIT: updated the desc....

Have too much of other stuff on my plate currently, so I have not been actively working on this. Still on my todo though. I hope I will get around...

I think abstracting model, loss and optimization/training seperately makes a lot of sense. Would require a ton of changes to the code in `inference` though I think.

Hey @janfb, would very much appreciate your input at this stage: Currently the PR adds the [`CategoricalMADE`](https://github.com/jnsbck/sbi/blob/bcc75dbdf4ab39533b69d9466dc1e4737468d4cf/sbi/neural_nets/estimators/categorical_net.py#L17) and builder [`build_autoregressive_categoricalestimator`](https://github.com/jnsbck/sbi/blob/bcc75dbdf4ab39533b69d9466dc1e4737468d4cf/sbi/neural_nets/net_builders/categorial.py#L69) + some minor modifications to `build_mnle` and `MixedDensityEstimator`. This enables...

Hey @janfb, you might have missed this, but I would be happy about feedback :)

Cool, thanks for all the feedback! A quick call would be great, also to discuss suitable tests for this. Will reach out via email and tackle the straight forward things...

After discussion with @janfb I will: 1. adapt the simulator of `Example_01_DecisionMakingModel.ipynb` to multiple discrete variables. 2. Get this to run for 1D and ND 3. fix remaining comments/issues 4....