Dmitry Kobak

Results 18 issues of Dmitry Kobak

The docstring here https://github.com/pavlin-policar/openTSNE/blob/master/openTSNE/nearest_neighbors.py#L213 (about pickling Annoy objects into a separate file) seems obsolete: that's not the way this is currently done in the code.

This looks like a nice meta-analysis https://www.medrxiv.org/content/10.1101/2021.07.29.21261282v2 estimating age-stratified IFR based on a hierarchical Bayesian model, using mostly the same data as in Levin et al. Their main focus is...

The terminal output before the gradient descent is a bit of a mess and always annoys me slightly :-) ```=============== t-SNE v1.1.0 =============== fast_tsne data_path: data.dat fast_tsne result_path: result.dat fast_tsne...

enhancement

When preparing a public talk, I often want to include an animation of how t-SNE develops during gradient descent. I thought I can call `fast_tsne` in a loop with `max_iter=1`...

enhancement

Leaving it here so that we don't forget: 2015 data for Oman looks like a low outlier, suggesting maybe incomplete registration in 2015. The subsequent years (2016-2019) do not show...

Waiting for NSO

The input file with the Shakespeare text https://raw.githubusercontent.com/karpathy/char-rnn/master/data/tinyshakespeare/input.txt has exactly 40000 lines and does not contain the entire Shakespeare, e.g. it does not contain string "Hamlet". What exactly is in...

Is there an easy way to use `constraint=Standardized()` with target standard deviation set to some value `sigma` instead of 1? Or would I need to define my own class by...

initial_dims is ignored when whiten=FALSE

See title. We discussed it at Dagstuhl. Should be useful for PyNNdescent, but could potentially be useful for Annoy/HNSW/etc at well. We could potentially allow Affinity classes to take **kwargs...

Hi, thanks for the great project! Currently MTEB has linear classification tasks and kNN-based retrieval tasks. What I feel is missing are kNN classification tasks. For example, I am thinking...