Skip to content

Commit

Permalink
initial commit :)
Browse files Browse the repository at this point in the history
  • Loading branch information
andrewGhazi committed Jun 30, 2024
0 parents commit 12c0c00
Show file tree
Hide file tree
Showing 23 changed files with 1,250 additions and 0 deletions.
11 changes: 11 additions & 0 deletions .Rbuildignore
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
^.*\.Rproj$
^\.git$
^\.github$
^\.gitignore$
^\.Rbuildignore$
^\.RData$
^\.Rhistory$
^\.Rproj\.user$
^NOTICE$
^LICENSE\.md$
^README\.Rmd$
51 changes: 51 additions & 0 deletions .github/workflows/check.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,51 @@
# Workflow derived from https://github.com/r-lib/actions/tree/v2/examples
# Need help debugging build failures? Start at https://github.com/r-lib/actions#where-to-find-help
on: [push, pull_request]

name: check

jobs:
check:
runs-on: ${{ matrix.config.os }}

name: ${{ matrix.config.os }} (${{ matrix.config.r }})

strategy:
fail-fast: false
matrix:
config:
- {os: ubuntu-latest, r: 'release'}

env:
GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
NOT_CRAN: true
R_KEEP_PKG_SOURCE: yes

steps:
- uses: actions/checkout@v3

- uses: r-lib/actions/setup-pandoc@v2

- uses: r-lib/actions/setup-r@v2
with:
r-version: ${{ matrix.config.r }}
extra-repositories: 'https://mc-stan.org/r-packages/'
http-user-agent: ${{ matrix.config.http-user-agent }}
use-public-rspm: true

- name: Install CmdStan
shell: Rscript {0}
run: |
install.packages("cmdstanr", repos = c("https://mc-stan.org/r-packages/", getOption("repos")))
cmdstanr::check_cmdstan_toolchain(fix = TRUE)
cmdstanr::install_cmdstan()
- uses: r-lib/actions/setup-r-dependencies@v2
with:
extra-packages: any::rcmdcheck, local::.
needs: check
cache-version: 2

- uses: r-lib/actions/check-r-package@v2
with:
upload-snapshots: true
11 changes: 11 additions & 0 deletions .gitignore
Original file line number Diff line number Diff line change
@@ -0,0 +1,11 @@
*~
.Rapp.history
.RData
.Rhistory
.Rproj.user/
inst/stan/**
!inst/stan/**/*.*
inst/stan/**/*.exe
inst/stan/**/*.EXE
*.dll
dyingforacup.Rproj
23 changes: 23 additions & 0 deletions DESCRIPTION
Original file line number Diff line number Diff line change
@@ -0,0 +1,23 @@
Package: dyingforacup
Type: Package
Title: Example Package With Instantiate
Version: 0.0.1
Author: You
Maintainer: The package maintainer <[email protected]>
Description: Demonstrate how to create an R package with pre-compiled Stan
models.
Depends:
R (>= 4.0.0)
Imports:
cli,
collapse,
data.table,
instantiate,
rlang
Additional_repositories:
https://mc-stan.org/r-packages/
SystemRequirements: CmdStan (https://mc-stan.org/users/interfaces/cmdstan)
Encoding: UTF-8
LazyData: true
RoxygenNote: 7.3.1
License: GPL (>= 3)
595 changes: 595 additions & 0 deletions LICENSE.md

Large diffs are not rendered by default.

15 changes: 15 additions & 0 deletions NAMESPACE
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
# Generated by roxygen2: do not edit by hand

export(run_gp)
export(suggest_next)
import(collapse)
importFrom(data.table,":=")
importFrom(data.table,.BY)
importFrom(data.table,.EACHI)
importFrom(data.table,.GRP)
importFrom(data.table,.I)
importFrom(data.table,.N)
importFrom(data.table,.NGRP)
importFrom(data.table,.SD)
importFrom(data.table,data.table)
importFrom(instantiate,stan_package_model)
34 changes: 34 additions & 0 deletions NOTICE
Original file line number Diff line number Diff line change
@@ -0,0 +1,34 @@
This package includes components from other open-source software. The projects and licenses are listed below.

* CmdStan (https://github.com/stan-dev/cmdstan) by Stan Developers and their Assignees.

BSD 3-Clause License
=====================

Copyright (c) 2014, Stan
All rights reserved.

Redistribution and use in source and binary forms, with or without modification,
are permitted provided that the following conditions are met:

* Redistributions of source code must retain the above copyright notice, this
list of conditions and the following disclaimer.

* Redistributions in binary form must reproduce the above copyright notice, this
list of conditions and the following disclaimer in the documentation and/or
other materials provided with the distribution.

* Neither the name of the {organization} nor the names of its
contributors may be used to endorse or promote products derived from
this software without specific prior written permission.

THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR
ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
13 changes: 13 additions & 0 deletions R/check_input.R
Original file line number Diff line number Diff line change
@@ -0,0 +1,13 @@
#' Check input data.frame
#' @description
#' The input data frame should have a limited number of columns and at least two rows
#'
check_df = function(dat, call = rlang::caller_env()) {
cn = colnames(dat)

if (!("rating" %in% cn)) cli::cli_abort("No `rating` column detected in input.", call = call)

cli::cli_alert("Detected brew parameters {.val {cn[!(cn %in% 'rating')]}}")

# if (nrow(dat) < 2) cli::cli_abort("Input needs at least two existing observations.", call = call)
}
20 changes: 20 additions & 0 deletions R/dyingforacup-package.R
Original file line number Diff line number Diff line change
@@ -0,0 +1,20 @@
#' @keywords internal
#' @name dyingforacup
#' @description A package using `instantiate` for coffee optimization.
#' @family help
#' @importFrom instantiate stan_package_model
#' @import collapse
"_PACKAGE"

## usethis namespace: start
#' @importFrom data.table :=
#' @importFrom data.table .BY
#' @importFrom data.table .EACHI
#' @importFrom data.table .GRP
#' @importFrom data.table .I
#' @importFrom data.table .N
#' @importFrom data.table .NGRP
#' @importFrom data.table .SD
#' @importFrom data.table data.table
## usethis namespace: end
NULL
33 changes: 33 additions & 0 deletions R/model.R
Original file line number Diff line number Diff line change
@@ -0,0 +1,33 @@
#' @title Fit the GP model.
#' @family models
#' @description Fit the GP Stan model and return posterior summaries.
#' @returns a draws data frame
#' @param X a matrix of scaled brew parameters
#' @param y Numeric vector of ratings (NORMAL scale, not 0-10!)
#' @param X_pred a matrix of grid points of in scaled brew parameter space to evaluate the GP at
#' @param ... Named arguments to the `sample()` method of CmdStan model
#' @param verbose logical indicating whether to print messages
#' objects: <https://mc-stan.org/cmdstanr/reference/model-method-sample.html>
#' @examples
#' if (instantiate::stan_cmdstan_exists()) {
#' run_gp_model(y = rnorm(5))
#' }
run_gp_model = function(X, y, X_pred, ..., verbose) {

model = instantiate::stan_package_model(
name = "gp_mod",
package = "dyingforacup"
)

data_list = list(N = length(y),
N_pred = nrow(X_pred),
D = ncol(X),
x = X,
x_pred = X_pred,
y = y)

fit = model$sample(data = data_list,
...)

fit$draws(format = "data.frame")
}
83 changes: 83 additions & 0 deletions R/run.R
Original file line number Diff line number Diff line change
@@ -0,0 +1,83 @@
#' @export
run_gp = function(dat, ...) {

check_df(dat)

# TODO adapt centering/scaling, generalize to arbitrary # of parameters
dat = dat |>
mtt(gs_cent = (grinder_setting - 9) / 5 * 3,
temp_cent = (temp - 190) / (20) * 3,
bloom_cent = (bloom_time - 30) / 30 * 3) |>
qDT()

g_map = data.table(g = seq(4,14, by = .5)) |>
mtt(gc = (g - 9) / 5 * 3)

t_map = data.table(t = seq(170, 210, by = 5),
tc = (seq(170, 210, by = 5) - 190) / 20 * 3)

b_map = data.table(b = seq(0, 60, by = 10),
bc = ((seq(0, 60, by = 10) - 30) / 30) * 3 )

x_grid = expand.grid(gc = g_map$gc,
tc = t_map$tc,
bc = b_map$bc) |>
qM()

X = dat |> slt(gs_cent, temp_cent, bloom_cent) |> qM()

list(run_gp_model(X, dat$rating, x_grid, ...),
x_grid)
}

#' @export
suggest_next = function(dat, x_grid, ...) {

run_res = run_gp(dat, ...)
gp_res = run_res[[1]]
x_grid = run_res[[2]]

obs_max = max(dat$rating)

offset = .25
minus_max = qM(gp_res |> get_vars("f_star", regex = TRUE)) - obs_max - offset

w = 1*(minus_max > 0)

acq = minus_max * w

max_pred_dens = fsum(acq) |> which.max()

if (max_pred_dens == 1) cli::cli_warn("Selected the first grid point as maximum of the acquisition function. You may need to run the chains for longer.")

pred_g = x_grid[max_pred_dens,,drop=FALSE][,"gc"]

acq_post = data.table(variable = colnames(acq),
mean = acq |> colMeans(),
i = 1:ncol(acq))

post_range = acq_post$mean |> range()

# qDT(x_grid) |> mtt(i = 1:nrow(x_grid)) |>
# sbt(dplyr::near(gc, pred_g)) |>
# join(acq_post, on = "i", validate = "1:1") |>
# ggplot(aes(tc, bc)) +
# geom_tile(aes(fill = mean)) +
# scale_fill_viridis_c(limits = post_range)

g_map = data.table(g = seq(4,14, by = .5)) |>
mtt(gc = (g - 9) / 5 * 3)

t_map = data.table(t = seq(170, 210, by = 5),
tc = (seq(170, 210, by = 5) - 190) / 20 * 3)

b_map = data.table(b = seq(0, 60, by = 10),
bc = ((seq(0, 60, by = 10) - 30) / 30) * 3 )

x_grid[max_pred_dens,,drop=FALSE] |>
qDT() |>
join(g_map, verbose = FALSE) |>
join(t_map, verbose = FALSE) |>
join(b_map, verbose = FALSE)

}
63 changes: 63 additions & 0 deletions README.Rmd
Original file line number Diff line number Diff line change
@@ -0,0 +1,63 @@
---
output: github_document
---

<!-- README.md is generated from README.Rmd. Please edit that file -->

```{r, include = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>",
fig.path = "man/figures/README-",
out.width = "100%"
)
```

# dyingforacup

[DO YOU FOLKS LIKE COFFEE?!](https://www.youtube.com/watch?v=RJC9DXQAd7U)

<!-- badges: start -->
<!-- badges: end -->

This is the package I use to optimize my coffee brewing with [Bayesian Optimization](https://www.youtube.com/watch?v=wZODGJzKmD0)

## Installation

You need to install [`cmdstanr`](https://mc-stan.org/cmdstanr/index.html) to use this package:

```{r eval=FALSE}
# we recommend running this is a fresh R session or restarting your current session
install.packages("cmdstanr", repos = c("https://stan-dev.r-universe.dev", getOption("repos")))
```

You can install the development version of `dyingforacup` like so:

```{r eval=FALSE}
remotes::install_github('andrewGhazi/dyingforacup', type = "source")
```

## Example

Give the `run_gp()` function a data frame of brew parameters with an

```{r eval=FALSE}
library(dyingforacup)
dat = tibble::tribble(
~grinder_setting, ~temp, ~bloom_time, ~rating,
# 8, 193, 25, 1.1,
7, 195, 20, -.7,
9, 179, 45, -1,
9, 195, 25, -.5,
)
suggest_next(dat,
iter_sampling = 4000,
refresh = 1250,
show_exceptions = FALSE,
adapt_delta = .95,
parallel_chains = 4)
```
Loading

0 comments on commit 12c0c00

Please sign in to comment.