David Muhr

Wheel not pre-built for manylinux1 (AWS Lambda)

9

Interestingly, using the newly release version `0.3.2`, the install fails on the default Linux Github Actions runnerm, because it attempts a source install (no appropriate wheel found?). ``` × Preparing...

Add option to watch all files in a directory.

4

Hey! This pull request adds an option to watch for all file changes in the specified or default directory. #### `-w, --watch` Watches all files in the specified or default...

Adresses https://github.com/KristofferC/NearestNeighbors.jl/issues/99

Benchmarking parallel implementation leads to segfaults

2

Hi! I'm currently trying to benchmark our parallel implementation of `knn` at different numbers of samples and dimensions. ```julia using BenchmarkTools: @benchmark import NearestNeighbors as NN const input_samples = Int.(exp2.(9:14))...

Unsupervised models and prediction

7

Hi, I would like to use ELKI to validate our [OutlierDetection.jl](https://github.com/OutlierDetectionJL/OutlierDetection.jl) algorithms. It appears that for unsupervised algorithms there is no facility to learn a model from data and predict...

Correctly load state from pretraining

1

# Filter out decoder network keys ae_net_dict = {k: v for k, v in ae_net_dict.items() if k in net_dict} Did not match any keys. The encoders seem to live in...

Enable stacking of outlier detection models

8

support probabilistic and deterministic outlier detection models.

Allow multidimensional array inputs

2

How could we extend JavaCall to allow passing of multidimensional arrays? E.g. ```java public class Test { public static double[][] testDoubleArray2D(double[][] array) { return array; } } ``` The problem...

Adressing , as mentioned we use this setup in .

SOD is (possibly) delivering wrong results

2

The shared nearest neighbors implementation in SOD leads to questionable results, probably because of a race condition in the [parallel snn implementation](https://github.com/yzhao062/pyod/blob/c4aecd1dbd19b7386514d1af8978115a2dab695a/pyod/models/sod.py#L17). The parallel snn leads to different results compared...