Scott Sievert
Scott Sievert
This pull request allows an easy option to enable/disable the sticky header. In _config.yml, the variable `sticky_header: true` or `sticky_header: false` can be included. By default, the sticky header remains...
In an org file with neovim, `cat` should show the TODO list. Instead, ``` traceback Error detected while processing function provider#python3#Call: line 18: Traceback (most recent call last): File "",...
Having a brief visualization on the readme of basic use would be nice. It’s the first thing I looked for when looking at this repo.
**What happened**: When I set `solver=lbfgs`, I get a warning: > /Users/scott/anaconda3/lib/python3.8/site-packages/dask/config.py:588: UserWarning: Configuration key "fuse_ave_width" has been deprecated. Please use "optimization.fuse.ave-width" instead **What you expected to happen**: I did...
This implements the popular mini-batch stochastic gradient descent in dask-glm. At each iteration, it grabs `batch_size` examples, computes the gradients for these examples, and then updates the parameter accordingly. The...
Optimization can take less time is the gradient is approximated using a subset of the examples. The approach in detailed in "[Hybrid deterministic-stochastic methods for data fitting][1]", but more examples...
Scikit-learn estimators have a `n_iter_` parameter. Dask-GLM doesn't. For example, the Scikit-Learn logistic regression implementation has the `n_iter_` parameter ([docs][sklearn]). However, the Dask-GLM implementation doesn't ([docs][dask]). This issue presented itself...
On [the arXiv version][1] of this paper, there's a missing citation: > ... Monte Carlo sampling where appropriate (sections 3.6, ??). [1]:https://arxiv.org/abs/1907.07587
When I run makemeasandwich, it's completely broken. It can't even log in. ``` shell $ sudo makemeasandwich --debug Logging in. $ ``` Maybe something changed in what Jimmy Johns returns?...
What's the reasoning behind "Note that pytorch might need about 1GB of disk space." on [the user guide][1]. When does that happen? Is it possible to use PyTorch without GPUs?...