Package: ssdsims 0.0.0.9015

Joe Thorley

ssdsims: Simulation Analyses for Species Sensitivity Distributions

Runs reproducible simulation studies for species sensitivity distribution (SSD) models built on the 'ssdtools' package. Expands a declarative scenario into per-step task tables, draws data, fits distributions, and estimates hazard concentrations, with a 'targets'-based, Hive-partitioned shard pipeline for running studies in parallel or on a cluster.

Authors:Joe Thorley [aut, cre], Rebecca Fisher [aut], Kirill Müller [aut]

ssdsims_0.0.0.9015.tar.gz
ssdsims_0.0.0.9015.zip(r-4.7)ssdsims_0.0.0.9015.zip(r-4.6)ssdsims_0.0.0.9015.zip(r-4.5)
ssdsims_0.0.0.9015.tgz(r-4.6-any)ssdsims_0.0.0.9015.tgz(r-4.5-any)
ssdsims_0.0.0.9015.tar.gz(r-4.7-any)ssdsims_0.0.0.9015.tar.gz(r-4.6-any)
ssdsims_0.0.0.9015.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
ssdsims/json (API)

# Install 'ssdsims' in R:
install.packages('ssdsims', repos = c('https://poissonconsulting.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/poissonconsulting/ssdsims/issues

Pkgdown/docs site:https://poissonconsulting.github.io

Datasets:

On CRAN:

Conda:

quarto

6.18 score 1 stars 45 scripts 43 exports 81 dependencies

Last updated from:3374ac88fd. Checks:9 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-devel-x86_64OK254
source / vignettesOK376
linux-release-x86_64OK256
macos-release-arm64OK187
macos-oldrel-arm64OK158
windows-develOK271
windows-releaseOK250
windows-oldrelOK214
wasm-releaseOK165

Exports:local_dqrng_backendlocal_dqrng_statescenario_datasetscenario_distsetscenario_min_pmixscenario_results_dirssd_analyse_costssd_calibrate_costssd_calibrate_cost_from_runssd_compare_costssd_cost_calibrationssd_define_scenariossd_designssd_design_targetsssd_distsetssd_estimate_costssd_genssd_open_uploadedssd_pmixssd_run_fit_stepssd_run_hc_stepssd_run_sample_stepssd_run_scenario_baselinessd_run_scenario_shardsssd_scenario_datassd_scenario_fit_shardsssd_scenario_fit_tasksssd_scenario_hc_shardsssd_scenario_hc_tasksssd_scenario_sample_shardsssd_scenario_sample_tasksssd_scenario_targetsssd_scenario_tasksssd_summarisessd_summarise_designssd_summarise_memberssd_summarise_uploadedssd_test_uploadssd_upload_azuressd_upload_dryrunssd_upload_shardtask_primerwith_dqrng_state

Dependencies:BHbitbit64cachemchkclicliprcodetoolscollectionscommonmarkcpp11crayoncurlDBIdigestdplyrdqrngduckdbduckplyrfarverfastmapfurrrfuturegenericsggplot2ggtextglobalsgluegoftestgridtextgtablehmsisobandjpegjsonlitelabelinglatticelifecyclelistenvlitedownmagrittrmarkdownMatrixmemoiseparallellypillarpkgconfigpngprettyunitsprogresspurrrR6rbibutilsRColorBrewerRcppRcppEigenRdpackreadrrlangS7scalessessioninfositmossddatassdtoolsstringistringrtibbletidyrtidyselectTMBtzdbuniversalsutf8vctrsVGAMviridisLitevroomwithrxfunxml2

Analysing a Run's Observed Compute Cost
Prediction, then measurement | A small worked run | Analysing observed cost | The shard envelope | Comparing predicted against observed | Recalibrating from the run | Across a design | The loop | See also
Rendered fromcost-analysis.qmdusingquarto::html

Last update: 2026-06-29
Started: 2026-06-14

Defining a Scenario
Why a declarative scenario? | Stage 1: assemble the data | Generated datasets with ssd_gen() | Stage 2: declare the scenario | Dataset input is the collection | min_pmix is referenced by name | dists is a collection of distribution sets | The ci flag | Stage 3: expand into task tables | Stage 4: run the baseline loop | Next: shards, designs, and the targets pipeline | See also
Rendered fromdefining-a-scenario.qmdusingquarto::html

Last update: 2026-06-29
Started: 2026-06-02

Estimating a Scenario's Compute Cost
Why estimate cost? | The cost model | The calibration object | Estimating a scenario | From serial total to wall-time | Recalibrating for your machine | Caveats | See also
Rendered fromcost-estimation.qmdusingquarto::html

Last update: 2026-06-29
Started: 2026-06-07

From a Single Scenario to a Design
Start with a single scenario | Migrate: wrap it in a design | Cache-preserving upgrade | Refine: zoom into one region | Notes | See also
Rendered fromscenario-to-design.qmdusingquarto::html

Last update: 2026-06-29
Started: 2026-06-14

Get Started with ssdsims
What ssdsims is for | Two tracks | A recommended reading order | A 30-second on-ramp
Rendered fromssdsims.qmdusingquarto::html

Last update: 2026-06-29
Started: 2026-06-29

Running a Sharded Pipeline
partition_by and bundle: how tasks group into shards | distset: bundled by default, promotable to the path | Run it single core, over shards | Results match the in-memory runner | Scaling up: the targets pipeline | Combining scenarios into a design | Comparing settings in one design | See also
Rendered fromsharded-pipeline.qmdusingquarto::html

Last update: 2026-06-29
Started: 2026-06-04

Running on a SLURM Cluster
The cluster template | Step 1 — Map your site's SLURM instructions to the controller | Backend prerequisite — the worker install path | Step 2 — Confirm the mapping with the preflight | Step 3 — Run a minimal first job | Step 4 — Swap in your own scenario | Targeting a non-SLURM cluster (untested) | Shard ↔ SLURM-job packing | See also
Rendered fromcluster-pipeline.qmdusingquarto::html

Last update: 2026-06-29
Started: 2026-06-07

Uploading Shards to Cloud Storage
The destination objects | Run it locally with a dry run | Extend the same call to Azure on a cluster | Verify the upload, in place | What to pay attention to | See also
Rendered fromcloud-upload.qmdusingquarto::html

Last update: 2026-06-29
Started: 2026-06-07

Readme and manuals

Help Manual

Help pageTopics
Local dqrng pcg64 Backendlocal_dqrng_backend
Local/With dqrng Statelocal_dqrng_state with_dqrng_state
Isolate a Materialised Dataset from a Scenario by Namescenario_dataset
Isolate a Distribution Set from a Scenario by Namescenario_distset
Isolate a Materialised 'min_pmix' Function from a Scenario by Namescenario_min_pmix
Seed- and Layout-keyed Results Root for a Scenarioscenario_results_dir
Analyse a Run's Observed Compute Costssd_analyse_cost
Calibrate the Per-task Cost Model on the Current Machinessd_calibrate_cost
Recalibrate the Cost Model from an Observed Runssd_calibrate_cost_from_run
Compare Predicted Against Observed Compute Costssd_compare_cost
Default Cost Calibrationssd_cost_calibration
Default Cost Calibration Objectssd_cost_calibration_default
Define a Simulation Scenariossd_define_scenario
Assemble and Validate a Design of Scenariosssd_design
Build the Targets Pipeline for a Designssd_design_targets
Assemble One or More Distribution Setsssd_distset
Estimate a Scenario's Compute Cost and Longest Taskssd_estimate_cost
Materialise Generator Datasets for a Simulation Scenariossd_gen
Open Uploaded Results for Querying, In Placessd_open_uploaded
Assemble and Validate 'min_pmix' Functions for a Simulation Scenariossd_pmix
Run a Scenario with the Baseline Loop Runnerssd_run_scenario_baseline
Run a Scenario over Hive-partitioned Parquet Shards (single core)ssd_run_scenario_shards
Run a Step Shardssd_run_fit_step ssd_run_hc_step ssd_run_sample_step ssd_run_step
Assemble and Validate Datasets for a Simulation Scenariossd_scenario_data
Group Tasks into Shardsssd_scenario_fit_shards ssd_scenario_hc_shards ssd_scenario_sample_shards ssd_scenario_shards
Build the Targets Pipeline for a Scenariossd_scenario_targets
Expand a Scenario into Task Tablesssd_scenario_fit_tasks ssd_scenario_hc_tasks ssd_scenario_sample_tasks ssd_scenario_tasks
Summarise a Run's hc Estimates Across Shardsssd_summarise
Combine Per-scenario Summaries into One Design Summaryssd_summarise_design
Summarise One Design Member from the Shared hc Shardsssd_summarise_member
Summarise Uploaded Results, In Place (the cloud 'ssd_summarise()')ssd_summarise_uploaded
Probe an Upload Destination's Credentials and Connectivityssd_test_upload
Upload Destinations for a Scenario's Shardsssd_upload_azure ssd_upload_dryrun
Ship Shard (or Summary) Parquet Files to an Upload Destinationssd_upload_shard
Derive a Per-task Primer from its Parameterstask_primer