- Adapted the package to accommodate the recent update of its dependency
fastshap (version 0.1.0). The usage of the functions remains unchanged.
- Optimized the function examples to meet the requirements for passing the CRAN check.
- Convert Shapley values-based functions to usable by external models (as described in issue # 3), and add examples in function documentation and vignettes to show users how to use these functions.
- Add a function
detect_envi_change to use Shapley values technique to analyze the potential impacts of changing environmental variables in space.
- Modify function
isotree_po to take presence-absence dataset as well (as described in issue #7). To make this happen smoothly, another function format_observation is created to help the users to convert their data to fit into itsdm workflow.
- Reorganized reference section of the online documentation to make it user-friendly.
- Fix a few bugs in the functions.
- Fix a bug in function
print.VariableAnalysis mentioned in issue #2: if any value is negative then it would fail.
- As mentioned in issue #3, add a sampling step in function
plot.ShapDependence when the number of records is larger than 1000. In order to keep the overall trend, the sampling is stratified by bins. So the points cloud can be clearer to interpret.
- Modify some text in POEvaluation plot figure.
- Fix a few bugs in README example.
- Include an independent function to calculate continuous Boyce Index (
.cont_boyce) in utils.R to reduce the pool of dependencies.
- Make the start message simpler.
- Fix the duplicated printout in function
print.POEvaluation.
- Use
inherits function to check “try-error” in dataset functions.
- Updated lines in function
evaluate_po, plot.POEvaluation, and print.POEvaluation related to dependency ecospat 3.2.1. Because now function ecospat.boyce supports Kendall method, itsdm changed to use Kendall method to calculate CBI.
- Merge the pull request made by David Cortes who is the author of package
isotree to use more flexible way for argument passing of isolation.forest.
- According to David Cortes’ reminder, remove argument
sample_rate from isotree_po. Only use sample_size for sub-sampling.
This is the first release. It includes all planned features.