--- title: "Examples of additional tidymodels features" output: rmarkdown::html_vignette: toc: true #output: rmarkdown::pdf_document vignette: > %\VignetteIndexEntry{Examples of additional tidymodels features} %\VignetteEngine{knitr::rmarkdown} %\VignetteEncoding{UTF-8} --- ```{r, include = FALSE} knitr::opts_chunk$set( collapse = TRUE, comment = "#>" ) # set options to limit threads used by imported libraries options(cores=2) options(mc.cores=2) # xgboost uses data.table data.table::setDTthreads(2) download_data = FALSE # note that data were created in the overview vignette ``` In this vignette, we illustrate how a number of features from `tidymodels` can be used to enhance a conventional SDM pipeline. We recommend users first become familiar with `tidymodels`; there are a number of excellent tutorials (both introductory and advanced) on its dedicated [website](https://www.tidymodels.org/) We reuse the example on the Iberian lizard that we used in the [`tidysdm` overview](https://evolecolgroup.github.io/tidysdm/articles/a0_tidysdm_overview.html) article. # Exploring models with `DALEX` An issue with machine learning algorithms is that it is not easy to understand the role of different variables in giving the final prediction. A number of packages have been created to explore and explain the behaviour of ML algorithms, such as those used in `tidysdm`. In the [`tidysdm` overview](https://evolecolgroup.github.io/tidysdm/articles/a0_tidysdm_overview.html) article, we illustrated how to use `recipes` to create profiles. Here we demonstrate how to use [DALEX](https://modeloriented.github.io/DALEX/), an excellent package that has methods to deal with `tidymodels`. `tidysdm` contains additional functions that allow use to use the DALEX functions directly on `tidysdm` ensembles. We will use a simple ensemble that we built in the overview vignette. ```{r} library(tidysdm) lacerta_ensemble ``` The first step in DALEX is to create an explainer object, which can then be queried by different functions in the package, to turn the explainer into an explanation (following the DALEX lingo). As a first step, we use the custom function `explain_tidysdm` to generate our explainer: ```{r} explainer_lacerta_ens <- explain_tidysdm(lacerta_ensemble) ``` Now that we have our explainer, we can explore variable importance for the ensemble: ```{r vip, fig.width=6, fig.height=4} library(DALEX) vip_ensemble <- model_parts(explainer = explainer_lacerta_ens) plot(vip_ensemble) ``` Or generate partial dependency plots for a given variable (e.g. bio05): ```{r pdp, fig.width=6, fig.height=4} pdp_bio05 <- model_profile(explainer_lacerta_ens, N = 500, variables = "bio05") plot(pdp_bio05) ``` There are many other functions in DALEX that can be applied to the explainer to further explore the behaviour of the model; see several tutorial on https://modeloriented.github.io/DALEX/ It is also possible to explore the individual models that make up the ensemble: ```{r} explainer_list <- explain_tidysdm(tidysdm::lacerta_ensemble, by_workflow = TRUE) ``` The resulting list can be then used to build lists of explanations, which can then be plotted. ```{r profile, fig.width=6, fig.height=4} profile_list <- lapply(explainer_list, model_profile, N = 500, variables = "bio05" ) plot(profile_list) ``` # The initial split The standard approach in `tidymodels` is to make an initial split of the data into a test and a training set. We will use retain 20% of the data (1/5) for the testing set, and use the rest for training. We start by loading a set of presences and absences and their associated climate, analogous to the one that we generated in the [`tidysdm` overview](https://evolecolgroup.github.io/tidysdm/articles/a0_tidysdm_overview.html) article: ```{r} library(tidysdm) library(sf) lacerta_thin <- readRDS(system.file("extdata/lacerta_climate_sf.RDS", package = "tidysdm" )) ``` We then use `spatial_initial_split` to do the split, using a `spatial_block_cv` scheme to partition the data: ```{r initial_split, fig.width=6, fig.height=4} set.seed(1005) lacerta_initial <- spatial_initial_split(lacerta_thin, prop = 1 / 5, spatial_block_cv ) autoplot(lacerta_initial) ``` And check the balance of presences vs pseudoabsences: ```{r} check_splits_balance(lacerta_initial, class) ``` We can now extract the training set from our `lacerta_initial` split, and sample folds to set up cross validation (note that we set the `cellsize` and `offset` based on the full dataset, `lacerta_thin`; this allows us to use the same grid we used for the `initial_split`). ```{r training_cv, fig.width=6, fig.height=4} set.seed(1005) lacerta_training <- training(lacerta_initial) lacerta_cv <- spatial_block_cv(lacerta_training, v = 5, cellsize = grid_cellsize(lacerta_thin), offset = grid_offset(lacerta_thin) ) autoplot(lacerta_cv) ``` And check the balance in the dataset: ```{r} check_splits_balance(lacerta_cv, class) ``` # Different recipes for certain models Only certain type of models (e.g. glm, svm) struggle with correlated variables; other algorithms, such as random forests, can handle correlated variables. So, we will create two recipes, one with all variables, and one only with the variables that are uncorrelated: ```{r recipe} lacerta_rec_all <- recipe(lacerta_thin, formula = class ~ .) lacerta_rec_uncor <- lacerta_rec_all %>% step_rm(all_of(c( "bio01", "bio02", "bio03", "bio04", "bio07", "bio08", "bio09", "bio10", "bio11", "bio12", "bio14", "bio16", "bio17", "bio18", "bio19", "altitude" ))) lacerta_rec_uncor ``` And now use these two recipes in a `workflowset` (we will keep it small for computational time), selecting the appropriate recipe for each model. We will include a model (polynomial support vector machines, or SVM) which does not have a wrapper in `tidysdm` for creating a model specification. However, we can use a standard model spec from `yardstick`: ```{r workflow_set} lacerta_models <- # create the workflow_set workflow_set( preproc = list( uncor = lacerta_rec_uncor, # recipe for the glm all = lacerta_rec_all, # recipe for the random forest all = lacerta_rec_uncor # recipe for svm ), models = list( # the standard glm specs glm = sdm_spec_glm(), # rf specs with tuning rf = sdm_spec_rf(), # svm specs with tuning svm = parsnip::svm_poly( cost = tune(), degree = tune() ) %>% parsnip::set_engine("kernlab") %>% parsnip::set_mode("classification") ), # make all combinations of preproc and models, cross = FALSE ) %>% # tweak controls to store information needed later to create the ensemble # note that we use the bayes version as we will use a Bayes search (see later) option_add(control = stacks::control_stack_bayes()) ``` We can now use the block CV folds to tune and assess the models. Note that there are multiple tuning approaches, besides the standard grid method. Here we will use `tune_bayes` from the `tune` package (see its help page to see how a Gaussian Process model is used to choose parameter combinations). This tuning method (as opposed to use a standard grid) does not allow for hyper-parameters with unknown limits, but `mtry` for random forest is undefined as its upper range depends on the number of variables in the dataset. So, before tuning, we need to finalise `mtry` by informing the set dials with the actual data: ```{r} rf_param <- lacerta_models %>% # extract the rf workflow extract_workflow("all_rf") %>% # extract its parameters dials (used to tune) extract_parameter_set_dials() %>% # give it the predictors to finalize mtry finalize(x = st_drop_geometry(lacerta_thin) %>% select(-class)) # now update the workflowset with the new parameter info lacerta_models <- lacerta_models %>% option_add(param_info = rf_param, id = "all_rf") ``` And now we can tune the models: ```{r tune_grid} set.seed(1234567) lacerta_models <- lacerta_models %>% workflow_map("tune_bayes", resamples = lacerta_cv, initial = 8, metrics = sdm_metric_set(), verbose = TRUE ) ``` We can have a look at the performance of our models with: ```{r} autoplot(lacerta_models) ``` # Stack ensembles Instead of building a simple ensemble with the best version of each model type, we can build a stack ensemble, as implemented in the package `stacks`. Stacking uses a meta-learning algorithm to learn how to best combine multiple models, including multiple versions of the same algorithm with different hyper-parameters. ```{r build_stack, fig.width=6, fig.height=4} library(stacks) set.seed(1005) lacerta_stack <- # initialize the stack stacks() %>% # add candidate members add_candidates(lacerta_models) %>% # determine how to combine their predictions blend_predictions() %>% # fit the candidates with non-zero weights (i.e.non-zero stacking coefficients) fit_members() autoplot(lacerta_stack, type = "weights") ``` We can see that three versions of the SVM and one of the random forests were selected; the stacking coefficients give an indication of the weight each model carries within the ensemble. We can now use the ensemble to make predictions about the testing data: ```{r predict_test} lacerta_testing <- testing(lacerta_initial) lacerta_test_pred <- lacerta_testing %>% bind_cols(predict(lacerta_stack, ., type = "prob")) ``` And look at the goodness of fit using some commonly used sdm metrics. Note that `sdm_metric_set` is first invoked to generate a function (with empty `()`) that is then used on the data. ```{r assess_test} sdm_metric_set()(data = lacerta_test_pred, truth = class, .pred_presence) ``` We can now make predictions with this stacked ensemble. We start by extracting the climate for the variables of interest ```{r eval=download_data} download_dataset("WorldClim_2.1_10m") climate_vars <- lacerta_rec_all$var_info %>% filter(role == "predictor") %>% pull(variable) climate_present <- pastclim::region_slice( time_ce = 1985, bio_variables = climate_vars, data = "WorldClim_2.1_10m", crop = iberia_poly ) ``` ```{r echo=FALSE, results='hide', eval=!download_data} climate_present <- terra::readRDS( system.file("extdata/lacerta_climate_present_10m.rds", package = "tidysdm") ) climate_vars <- lacerta_rec_all$var_info %>% filter(role == "predictor") %>% pull(variable) if (!all(climate_vars %in% names(climate_present))) { stop("mismatched variables in the raster") } ``` ```{r plot_present, fig.width=6, fig.height=4} prediction_present <- predict_raster(lacerta_stack, climate_present, type = "prob" ) library(tidyterra) ggplot() + geom_spatraster(data = prediction_present, aes(fill = .pred_presence)) + scale_fill_terrain_c() + # plot presences used in the model geom_sf(data = lacerta_thin %>% filter(class == "presence")) ``` # Using multi-level factors as predictors Most machine learning algorithms do not natively use multilevel factors as predictors. The solution is to create dummy variables, which are binary variables that represent the levels of the factor. In `tidymodels`, this is done using the `step_dummy` function. Let's create a factor variable with 3 levels based on altitude. ```{r} library(tidysdm) # load the dataset lacerta_thin <- readRDS(system.file("extdata/lacerta_climate_sf.RDS", package = "tidysdm" )) # create a topography variable with 3 levels based on altitude lacerta_thin$topography <- cut(lacerta_thin$altitude, breaks = c(-Inf, 200, 800, Inf), labels = c("plains", "hills", "mountains")) table(lacerta_thin$topography) ``` We then create the recipe by adding a step to create dummy variables for the `topography` variable. ```{r} # subset to variable of interest lacerta_thin <- lacerta_thin %>% select(class, bio05, bio06, bio12, bio15, topography) lacerta_rec <- recipe(lacerta_thin, formula = class ~ .) %>% step_dummy(topography) lacerta_rec ``` Let's us see what this does: ```{r} lacerta_prep <- prep(lacerta_rec) summary(lacerta_prep) ``` We have added two "derived" variables, *topography_hills* and *topography_mountains*, which are binary variables that allow us to code topography (with plains being used as the reference level, which is coded by both hills and mountains being 0 for a given location). We can look at the first few rows of the data to see the new variables by baking the recipe: ```{r} lacerta_bake <- bake(lacerta_prep, new_data = lacerta_thin) glimpse(lacerta_bake) ``` We can now run the sdm as usual: ```{r} # define the models lacerta_models <- # create the workflow_set workflow_set( preproc = list(default = lacerta_rec), models = list( # the standard glm specs glm = sdm_spec_glm(), # rf specs with tuning rf = sdm_spec_rf() ), # make all combinations of preproc and models, cross = TRUE ) %>% # tweak controls to store information needed later to create the ensemble option_add(control = control_ensemble_grid()) # tune set.seed(100) lacerta_cv <- spatial_block_cv(lacerta_thin, v = 3) lacerta_models <- lacerta_models %>% workflow_map("tune_grid", resamples = lacerta_cv, grid = 3, metrics = sdm_metric_set(), verbose = TRUE ) # fit the ensemble lacerta_ensemble <- simple_ensemble() %>% add_member(lacerta_models, metric = "boyce_cont") ``` We can now verify that the dummy variables were used by extracting the model fit from one of the models in the ensemble: ```{r} lacerta_ensemble$workflow[[1]] %>% extract_fit_parsnip() ``` We can see that we have coefficients for *topography_hills* and *topography_mountains*. Let us now predict the presence of the lizard in the Iberian Peninsula using the ensemble. Note that, for `predict_raster()` to work, the name and levels for a categorical variable need to match with those used when training the models (i.e. in the recipe with `step_dummy()`): ```{r} climate_present <- terra::readRDS(system.file("extdata/lacerta_climate_present_10m.rds", package = "tidysdm")) # first we add a topography variable to the climate data climate_present$topography <- climate_present$altitude climate_present$topography <- terra::classify(climate_present$topography, rcl = c(-Inf, 200, 800, Inf), include.lowest=TRUE, brackets=TRUE) library(terra) levels(climate_present$topography) <- data.frame(ID = c(0,1,2), topography = c("plains", "hills", "mountains")) # now we can predict predict_factor <- predict_raster(lacerta_ensemble, climate_present) plot(predict_factor) ```