Zach Deane-Mayer

R / Rcpp

3

Seeing as you're planning a future Julia wrapper, I thought I'd ask for an R one too, so I can try FALCONN instead of [RcppAnnoy](https://github.com/eddelbuettel/rcppannoy).

data on number of tests?

4

There's some data sources out there on number of tests by state/county— do you have any plans to try to organize some of this data too?

Make a PR to the main keras repo?

4

I think it might make sense to add this optimizer to the main keras repo!

A Simple but Tough-to-Beat Baseline for Sentence Embeddings

9

Paper: http://104.155.136.4:3000/pdf?id=SyK00v5xx Blog post: http://www.offconvex.org/2016/02/14/word-embeddings-2/ Looks like an interesting idea

[request] model trained on wikipedia articles

1

The pre-trained imagenet model is super useful. A model pre-trained on wikipedia (or another large text corpus) would be super useful too!

AnomalyDetectionTs drops timezone from POSIXct objects and converts POSIXlt to POSIXct

29

AnomalyDetectionTs should keep the timestamp column of the output dataset as-is, rather than converting to POSIXct and dropping the timezone attribute: ``` r library(AnomalyDetection) data(data_raw) data

OneHotEncoder(sparse=True)

11

[sklearn.preprocessing.OneHotEncoder](https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.OneHotEncoder.html) has the option `sparse=True`, to return the output in a scipy.sparse matrix. This can be really useful if you have categories with high cardinality. Would it be possible to...

It's broke

Lots of testthat tests have warnings— need to fix them