flex
flex copied to clipboard
xxxnell
Probabilistic deep learning for data streams.
sbt `experiment` command in root should execute all experiments (c.f. `flex.experiment` package). However, for now, only one experiment is executed (with `arg0`). Therefore, the `experiment` command that does not have...
Now RecurSketch only override update, and count updates when the update is called. Therefore, RecurSketch doesn't update its count when it calls the narrowUpdate only. narrowUpdate of RecurSketch must be...
`sample` of `Sketch` returns boundary values (e.g. List(..., -1.3346329812349141E307, 7.927339694866348E307, ..., 4.2420349412446703E307, ..., -1.1082857601763558E308...)).
`smoothing` operations are used in several places. The use of UpdateCmap and DeepUpdate is especially important. As part of refactoring the smoothing operation, several methods should be applicable dynamically.
When `Sketch` estimates the density distribution, too low a KL-divergence value is obtained because the boundary is not processed properly. Therefore, as a way of smoothing the edges, we use...
There is some error (about 10-15%) between `interpolationPdf` and `fastPdf`. See `fastPdf` part of `AdaPerSketchSpec`.