PoisotLab/SpeciesDistributionToolkit.jl: PseudoAbsences-v1.1.0
Bibliographic record
Abstract
PseudoAbsences PseudoAbsences-v1.1.0 Diff since PseudoAbsences-v1.0.1 Merged pull requests: Download API for GBIF (#323) (@tpoisot) SDT new version with pseudo absences (#413) (@tpoisot) SimpleSDMPolygons (#417) (@gottacatchenall) Load and save occurrences from JSON files (#421) (@tpoisot) Default data and plotting for OccInt (#423) (@tpoisot) Remove occurrences plotting from SDT (#424) (@tpoisot) Fix the ways elements are collected from GBIFRecords (#425) (@tpoisot) semver(sdt): v1.6.0 with SimpleSDMPolygons (#427) (@gottacatchenall) Makie for layers (#428) (@tpoisot) SimpleSDMPolygons with re-exports from datasets (#429) (@tpoisot) SDMPolygons Makie.lines (#434) (@gottacatchenall) Phylopic v1 (#435) (@tpoisot) Polygons: Makie.lines a boundingbox (#437) (@gottacatchenall) GADM URL Typo (#440) (@gottacatchenall) SDMPolygons Makie extension supporting lines! and poly! without axis argument (#441) (@gottacatchenall) Release SDT post Phylopic 1 (#444) (@tpoisot) Fix various occint bugs (#447) (@tpoisot) Add bioregions from OneEarth (#448) (@tpoisot) Add bibliography to the documentation (#450) (@tpoisot) Fix GeoInterface implementation for (multi) polygons (#451) (@asinghvi17) OneEarth CSV Path Fix (#453) (@gottacatchenall) SimpleSDMPolygons docs + tests (#454) (@tpoisot) Replace MakieCore by Makie for the layers plotting extensions (#456) (@github-actions[bot]) Variable selection checkpointing issue fixed (#459) (@tpoisot) Iteration over occurrences (#461) (@tpoisot) Replace MakieCore by Makie in OccurrencesInterface (#464) (@tpoisot) Change plotting package for polygons to Makie (#467) (@tpoisot) Update plotting extension to Makie for Phylopic (#468) (@tpoisot) Doc updates pre CSEE workshop (#473) (@tpoisot) Fix masking of occurrences by feature collections (#478) (@tpoisot) Bug fixes for SimpleSDMPolygons (#482) (@tpoisot) Fix the bug when masking a layer by a featurecollection (#483) (@tpoisot) Fix some bugs in SimpleSDMPolygons (#487) (@tpoisot) Experiments with AdaBoost (#488) (@tpoisot) Adding operations to SimpleSDMPolygons (#490) (@gottacatchenall) Support Makie v0.24 in SimpleSDMLayers (#491) (@tpoisot) Add Phylopic dep for 0.24 (#492) (@tpoisot) Clipping Polygons w/ Bounding Boxes (#495) (@gottacatchenall) boundingbox for MultiPolygon hotfix (#496) (@gottacatchenall) Training and thresholding improvements for SDeMo (#498) (@tpoisot) Adding CRS to construct polygons (#500) (@gottacatchenall) Explict specification of delimiter to GBIF CSV.File (#501) (@jeetsukumaran) Update to make decision stubs actually work (#502) (@tpoisot) SDeMo: PCA and Logistic automatically apply a z-score (#514) (@tpoisot) Fix PCA bug (#515) (@tpoisot) Fix for decision stumps (#518) (@tpoisot) Calibration improvements (#519) (@tpoisot) SDeMo v1.6.1 (#524) (@tpoisot) Occurrences with missing coordinates are handled during boundingbox (#526) (@tpoisot) Add a function to use the distance in degrees to generate pseudo-absences (#527) (@tpoisot) Closed issues: Ecoregions (#49) MLJ integration (#83) Use the resolve endpoint to match GBIF (#181) Download API for GBIF (#308) Add a reclassify function (#407) New version using PseudoAbsences (#411) Cellsize was useful... (#416) CrossValidation type (#420) Move plotting function in SimpleSDMLayers (#422) Documentation update for OccurrencesInterface (#426) Phylopic v1 when? (#430) Add methods to do lines in addition to poly (#431) Some demo occurrences in OccurrencesInterface are wrong (#432) Boundingbox for SimpleSDMPolygons (#433) Typo in GADM url in SDMPolygons (#438) OneEarth ecoregions (#442) Indexing Polygons based on properties (#449) Can't tag new SimpleSDMPolygons release (#452) Support iteration for AbstractOccurrences (#460) Cannot mask occurrences by a feature collection (#476) Masking a layer with a FeatureCollection is not working (#477) Bugfixes for polygons (#479) Resolv should be Resolve (fix + alias for compatibility) (#480) EPA polygons (I think I know why) are not working (#481) Cleanup the reference section of the package (#484) EPA polygons are on the side (#485) First polygon plotted by Makie has the wrong attributes (#486) No boundinngbox method for multipolygons (#494) Improvements to model thresholding (#497) Clipped polygons cannot be used to mask layers (#499) bug in virtual species documentation (#511) vcat doesn't work for MultiPolygons (#512) Decision stumps make weird predictions (#517) Add weights to isotonic calibration (#521) Platt calibration is very slow (#522)
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.013 |
| Meta-epidemiology (narrow) | 0.005 | 0.007 |
| Meta-epidemiology (broad) | 0.003 | 0.004 |
| Bibliometrics | 0.006 | 0.005 |
| Science and technology studies | 0.002 | 0.001 |
| Scholarly communication | 0.006 | 0.010 |
| Open science | 0.007 | 0.008 |
| Research integrity | 0.003 | 0.005 |
| Insufficient payload (model declined to judge) | 0.495 | 0.590 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".