-
-
Hi Neel,
Thanks for this powerful tool for interpretation and it helps me a lot in understanding the mechanics of LLMs. But I found there is slight difference in gpt-neo-125m weights loading with h…
-
We should think about how best to resolve the partial duplications between `sktime` and `skpro` related to probability distributions and metrics, in terms of end state and transition strategy. Quickly…
-
The paper [Weight Uncertainty in Neural Networks](https://arxiv.org/abs/1505.05424) proposes the usage of a mixture of two Gaussians as the prior distribution. By default, the priors (of the kernel an…
-
# Node Support Definition
Node support is an estimate of the confidence that a clade is in the true tree. Formally, this is the sum over all posterior probabilities for trees that the node appears …
-
**Is your feature request related to a problem? Please describe.**
n/a
**Describe the solution you'd like**
Would like to see the following features added
- re-scaling of height of PDF
For N(0…
-
The following is a very simple and easy way to add a noise model to Bioscrape inference so cost functions are actually probabilities.
Allow two noise models, normal and log-normal (distribution typ…
-
I was thinking something like:
listOfLength: Int -> Fuzzer a -> Fuzzer (List a)
This looks like it shouldn't be too hard to generalize from the existing list function. The major benefit would be t…
-
Hi,
I have a clusterer trained with many clusters identified. I used
` hdbscan.prediction.membership_vector(clusterer, points_to_predict)`
to get the probability distribution of the points o…
-
The [ASH](https://github.com/joshday/AverageShiftedHistograms.jl/pull/40) package wants to add `logpdf` and `logcdf`, but adding a dependency on Distributions might be a little bit overboard. Should w…