Implements an algorithim for Latent Dirichlet Allocation using style conventions from the [tidyverse](https://style.tidyverse.org/) and [tidymodels](https://tidymodels.github.io/model-implementation-principles/index.html).
Other
41
stars
3
forks
source link
refit.tidylda does not propagate eta forward correctly #66
The model resulting from a call to refit.tidylda should always have eta as a matrix. However, I have a model where eta is a scalar. This affects all downstream results for sequential calls to refit, as in the case of incremental learning.
TommyJones
commented
2 years ago
The model resulting from a call to
refit.tidyldashould always haveetaas a matrix. However, I have a model whereetais a scalar. This affects all downstream results for sequential calls torefit, as in the case of incremental learning.