Tony Duan

You need to install [dvisvgm](https://dvisvgm.de) onto your system (it does not come pre-installed with this package).

This is still an issue. One workaround is to use the following, with the latest version of this package: ``` python3 -m readme2tex --branch master --nocdn --readme INPUT.md --output README.md...

I'm able to reproduce this issue by attempting to run the file [1]. The issue seems to arise due to the way Scikit-Learn is using `GridSearchCV`; any `BaseEstimator` sub-class must...

> @tonyduan do you think this might be due to a combination of the log(scale) parametrization and the scale-up part of the line search? Maybe we could try implementing a...

Thank you @ryan-wolbeck!

> @tonyduan are you planning on doing a re-review here? I see some relevant changes have been made. There's an identifiability issue so I'm recommending against implementing the BetaBernoulli. We...

Hi Oleg, At the moment the Multivariate Normal distribution is still under development and is intended for use in survival regression (i.e. with censorship). That said, it should not be...

Hi Dorukhan, Unfortunately I don't have the bandwidth right now to dive into contemporary lines of work on matrix completion. However, if you're willing to do the heavy lifting I'd...

@jasonsun0310 Wow, thanks for citing! You can use the following BibTeX: ``` @software{tonyduan_matrix_completion_github, title = {Lightweight Python library for in-memory matrix completion.}, copyright = {EPL-1.0 License}, url = {https://github.com/tonyduan/matrix-completion}, author...

Thanks for making the requested changes. > The parameters in OneByOneConv are not registered correctly as parameters, and are not moved to the correct device when we call model.to(device). So...