0.0.9rc2

Release Notes

This release candidate adds the ImageTransformer class for the OHBM tutorial.

Changes

• [ENH] Add ImageTransformer class (#513) @tsalo

0.0.9rc1

Release Notes

This release candidate adds an overwrite option in preparation for the OHBM 2021 NiMARE tutorial.

Changes

• [ENH] Add overwrite option to transform_images (#509) @tsalo

0.0.8

Release Notes

This release includes a number of bug-fixes, along with enhancements to how many tools within NiMARE implement low-memory options. In addition, we have renamed the CBMA estimators' null methods. The "analytic" method is now "approximate" and the "empirical" method is now "montecarlo".

Changes

• [REF] Rename CBMA null distribution generation methods (#494) @tsalo
• [FIX] Add informative error when NeuroVault collection is not found (#500) @tsalo
• [ENH] Support symmetric GCLDA topics with more than two subregions (#499) @tsalo
• [DOC] Add sphinx-copybutton to docs requirements (#502) @tsalo
• [ENH] Incorporate information about valid masking approaches into IBMA Estimators (#495) @tsalo
• [FIX] Deal with extreme t-values in t_to_z by truncating associated p-values (#498) @tsalo
• [TST] Add flake8-isort to test dependencies (#493) @tsalo
• [REF] Miscellaneous GCLDA cleanup (#486) @tsalo
• [DOC] Add new functions and classes to API documentation (#490) @tsalo
• [ENH] add images_to_coordinates (#446) @jdkent
• [ENH] Add check_type function (#480) @tsalo
• [REF] Add low_memory option to Estimators and add function for moving metadata from Dataset to DataFrame (#476) @tsalo
• [FIX] Set Dataset.basepath using absolute path (#474) @tsalo
• [FIX] Find common stem in find_stem instead of largest common substring (#472) @tsalo
• [FIX] Replace misspelled "log_p" with "logp" (#468) @tsalo
• [FIX] Assume non-symmetric null distribution in ALESubtraction (#464) @tsalo
• [TST] Add memmap test. (#463) @jdkent
• [REF] Write temporary files to the NiMARE data directory (#460) @tsalo
• [REF] Use saved MA maps, when available, in CBMA estimators (#462) @tsalo
• [FIX] Neurovault name collisions (#457) @jdkent
• [FIX] Update niftimasker in dataset blob (#459) @jdkent
• [FIX] Add work-around for maskers that do not accept 1D input (#455) @jdkent
• [ENH] Add low-memory option for kernel transformers (#453) @tsalo
• [ENH] add function to convert neurovault collections to a NiMARE dataset (#432) @jdkent
• [FIX] Ensure IBMA results have the expected number of dimensions (#450) @jdkent
• [STY, TST] Add flake8-docstrings to requirements (#435) @tsalo

0.0.7

Release Notes

This release involves two changes worth mentioning. First, we have fixed a bug in how permutation-based p-values are calculated (thanks to @alexenge for identifying and reporting). Second, we have changed how the "empirical" null method is performed. The "empirical" method is now much slower, but more accurate, than the "analytic" approach.

This release should deploy to PyPi, unlike 0.0.6.

Changes

• [FIX] Permutation p-values (#447) @tyarkoni
• [FIX,REF] start changing how to handle resampling (#439) @jdkent
• [FIX] transform_images extra dimension (#445) @jdkent
• [DOC] Add decoding description page (#443) @tsalo
• [MAINT] Switch to GitHub Actions for PyPi deployment (#441) @tsalo
• [ENH] Implement full coordinate-set empirical null method (#424) @tsalo
• [DOC] Fix NeuroStars link (#434) @tsalo
• [DOC] Add specialized issue templates (#433) @tsalo
• [MAINT] Add indexed_gzip as a dependency (#431) @tsalo

0.0.7rc1

Release Notes

This release involves two changes worth mentioning. First, we have fixed a bug in how permutation-based p-values are calculated (thanks to @alexenge for identifying and reporting). Second, we have changed how the "empirical" null method is performed. The "empirical" method is now much slower, but more accurate, than the "analytic" approach.

Changes

• [FIX] Permutation p-values (#447) @tyarkoni
• [FIX,REF] start changing how to handle resampling (#439) @jdkent
• [FIX] transform_images extra dimension (#445) @jdkent
• [DOC] Add decoding description page (#443) @tsalo
• [MAINT] Switch to GitHub Actions for PyPi deployment (#441) @tsalo
• [ENH] Implement full coordinate-set empirical null method (#424) @tsalo
• [DOC] Fix NeuroStars link (#434) @tsalo
• [DOC] Add specialized issue templates (#433) @tsalo
• [MAINT] Add indexed_gzip as a dependency (#431) @tsalo

0.0.6

Release Notes

WARNING: This release was not deployed to PyPi. However, 0.0.7 is the same as 0.0.6, so just use that one.

This release involves two changes worth mentioning. First, we have fixed a bug in how permutation-based p-values are calculated (thanks to @alexenge for identifying and reporting). Second, we have changed how the "empirical" null method is performed. The "empirical" method is now much slower, but more accurate, than the "analytic" approach.

Changes

• [FIX] Permutation p-values (#447) @tyarkoni
• [FIX,REF] start changing how to handle resampling (#439) @jdkent
• [FIX] transform_images extra dimension (#445) @jdkent
• [DOC] Add decoding description page (#443) @tsalo
• [MAINT] Switch to GitHub Actions for PyPi deployment (#441) @tsalo
• [ENH] Implement full coordinate-set empirical null method (#424) @tsalo
• [DOC] Fix NeuroStars link (#434) @tsalo
• [DOC] Add specialized issue templates (#433) @tsalo
• [MAINT] Add indexed_gzip as a dependency (#431) @tsalo

0.0.5

Release Notes

This release is focused on fixing two bugs in v0.0.4. One bug affected which files were packaged with the library, such that some templates were missing. The other bug was introduced in v0.0.4 and invalidates cluster-level Monte Carlo-based FWE-correction in coordinate-based meta-analyses.

Changes

• [FIX] Convert histogram weights to null distribution in p-to-stat conversion (#430) @tsalo
• [FIX] Fix packaging of resources (#428) @tsalo
• [FIX] Include resources in library data files (#427) @tsalo

0.0.4

Release Notes

This release includes a number of substantial changes to NiMARE.

Major changes

1. We've added PyMARE as a dependency! PyMARE is a general-purpose meta-analysis library in Python that we now use to perform our image-based meta-analyses.
2. For image-based meta-analyses, we also now have a transforms module to calculate new image types from available data.
3. Datasets now have a number of attributes retained as properties, which will break compatibility with Datasets from older versions of NiMARE.
4. We now have multiple methods for converting summary statistics (e.g., ALE, OF) to p-values in all of our major CBMA algorithms, thanks to @tyarkoni! The two current methods for each algorithm are a fast, but slightly less accurate, "analytic" method and a slower, but more accurate, "empirical" method. For ALE, We generally recommend the "analytic" method for maximum compatibility with GingerALE. The implementations of these algorithms have also been streamlined and sped up somewhat.
5. We have a new generate module for simulating coordinate-based datasets, thanks to @jdkent!
6. A number of modules, classes, and functions that were not yet implemented have been pruned from the API to make it easier to work with. Don't worry, we're still planning to get around to them at some point.

Changes

• [FIX] Fix the warnings about mismatched kernels and estimators (#425) @tsalo
• [FIX] Add nullhist_to_p and crop invalid p-values (#409) @tsalo
• [TST] Do not download test peaks2maps to tmpdir (#419) @tsalo
• [FIX] Restructure Peaks2MapsKernel to operate like other kernels (#410) @tsalo
• [ENH] Improve convergence between ALE null methods (#411) @tsalo
• [DOC] Add warnings for CBMA kernel/estimator mismatch (#416) @tsalo
• [FIX] Remove rows with empty abstract before running LDAModel (#414) @JulioAPeraza
• [FIX] Sort all arrays and DataFrames in Dataset by ID (#402) @tsalo
• [FIX] Allow no coordinates in a dataset (#407) @jdkent
• [ENH] Add analytic null method to KDA estimator (#397) @tsalo
• [FIX] Use unzipped mask as temporary fix (#401) @tsalo
• [DOC] Update API and examples (#395) @tsalo
• [REF] CBMA re-organization and improvement (#393) @tyarkoni
• [MAINT] Pin to PyMARE 0.0.2 (#391) @tsalo
• [TST] Test both analytic and empirical methods in ALE and MKDA (#380) @jdkent
• [FIX] Change default seed to None (#392) @jdkent
• [PERF] Various performance improvements (#386) @tyarkoni
• Add performance tweaks to ALE analytical null generation (#390) @tyarkoni
• fix tests (#387) @tyarkoni
• [FIX] respect n_noise_foci value (#382) @jdkent
• [ENH] Add analytic null method to MKDADensity (#375) @tsalo
• [ENH] Add empirical null method to density-based CBMA Estimators (#372) @tsalo
• [REF] Refactor KernelTransformer hierarchy (#369) @tyarkoni
• [ENH] Add generate module (#343) @jdkent
• [FIX] enforce correct lowest p-value (#365) @jdkent
• [FIX] Treat vfwe as an array of floats for KDA (#362) @jdkent
• [DOC] Update roadmap.rst (#359) @tsalo
• [DOC] Add example of combining kernels and CBMA estimators (#346) @koudyk
• [MAINT] Add Dorota Jarecka to Zenodo file (#358) @djarecka
• [MAINT] Add Enrico Glerean's affiliation and ORCID (#357) @eglerean
• [ENH] Clip p-values based on number of permutations (#353) @tsalo
• [REF] Remove unused alpha argument in statsmodels call (#354) @tsalo
• [ENH] Replace TTest with PermutedOLS (#304) @tsalo
• [REF] Reduce dependencies (#345) @tsalo
• [ENH] Add Neurosynth data fetcher (#342) @tsalo
• [INFRA] Add json describing filename convention (#338) @tsalo
• [DOC] Enable CBMA example (#337) @tsalo
• [FIX] Add private setter method for Dataset.ids (#336) @tsalo
• [REF] More low-memory work (#334) @tsalo
• [FIX, DOC] Change natural log to base-ten and document output naming convention (#333) @tsalo
• [FIX] Pin setuptools again (#331) @tsalo
• [FIX] Update setuptools version (#330) @tsalo
• [FIX] Add setuptools to requirements (#329) @tsalo
• [TST] Add test for peaks2maps (#328) @tsalo
• [FIX, TST] Fix and test CorrelationDistributionDecoder (#327) @tsalo
• [TST] Use temporary directories with automatic teardown (#326) @tsalo
• [REF] Speed up CorrelationDecoder (#324) @tsalo
• [ENH] Support Dataset transformations in kernel transformers (#320) @tsalo
• [ENH] Add PairwiseCBMAEstimator class and add low_memory option to ALESubtraction (#319) @tsalo
• [TST] Improve meta-analysis tests (#318) @tsalo
• [DOC] Fix Lancaster xform and Sleuth conversion docstrings (#317) @tsalo
• [TST] Improve nimare.io test coverage (#314) @tsalo
• [REF] Reduce duplication by calling _check_ncores (#313) @tsalo
• [REF] Remove generate_cooccurrence (#312) @tsalo
• [REF] Operate on arrays in ALESubtraction (#311) @tsalo
• [TST] Add flake8-black to test requirements (#300) @akimbler
• [FIX] Support multiple header lines in Sleuth text files (#310) @tsalo
• [FIX] Operate on copy of df in extract_cogat() (#306) @tsalo
• [MAINT] Update setup configuration (#303) @tsalo
• [REF] Sort imports alphabetically (#299) @tsalo
• [REF] Run automated code formatting with black (#296) @tsalo
• [DOC] Remove whitespace from README (#295) @tsalo
• [MAINT, TST] Drop 3.5 support. Add tests for Python 3.7 and 3.8. (#293) @tsalo
• [MAINT] Delete unused files (#291) @tsalo
• [MAINT] Increase minimum tensorflow to 2.0.0 (#290) @tsalo
• [FIX] Update peaks2maps w.r.t. recent changes in the API (#287) @tsalo
• [FIX] Raise an error in Decoders if no features remain (#284) @tsalo
• [REF] Move CBMA methods up a level (#283) @tsalo
• [REF] Rename RandomEffectsGLM to TTest (#282) @tsalo
• [ENH] Split DerSimonianLaird and Hedges IBMA estimators (#281) @tsalo
• [DOC] Expand IBMA example (#280) @tsalo
• [ENH] Use PyMARE for image-based meta-analyses (#273) @tsalo
• [FIX] Replace NaNs in Datasets with Nones (#276) @tsalo
• [ENH] Support initialized and uninitialized kernels for CBMA (#275) @tsalo
• [ENH] Add functions to convert image types (#272) @tsalo
• [REF] Convert Dataset attributes to properties (#270) @tsalo
• [REF] Drop unimplemented annotators (#269) @tsalo
• [REF] Drop unimplemented parcellate module and meta-ICA workflow (#264) @tsalo
• [ENH] Use nearest-neighbor interpolation for masks (#258) @tsalo

0.0.3

Release Notes

This release consolidates changes prior to PyMARE integration. In addition to a number of bug fixes, this release also includes substantial changes to ALESubtraction, annotation storage, and Dataset size.

Changes

• [FIX] Preallocate ALE cFWE p-value array with ones instead of zeros (#254) @tsalo
• [ENH] Convert decoders to classes (#252) @tsalo
• [ENH] Add prefix to annotations to delineate sources (#250) @tsalo
• [REF] Drop Dataset.data attribute (#249) @tsalo
• [ENH] Remove voxel selection in ALESubtraction (#245) @tsalo
• [REF] Eliminate duplication in ALESubtraction (#244) @tsalo
• [FIX] Add minimum nibabel version (#237) @tsalo
• [TST] Fix CodeCov config file (#240) @tsalo
• [ENH] Add transforms module (#239) @tsalo
• [TST] Add workflow tests (#235) @tsalo
• [REF, ENH] Add CBMAEstimator base class (#232) @tsalo
• [MAINT] Add @nicholst's info to Zenodo file (#228) @tsalo
• [MAINT] Consolidate requirements files (#230) @tsalo
• [ENH] Dataset.get_X methods return available types when type is not provided (#205) @tsalo
• [FIX] Update workflows given recent changes (#226) @tsalo
• [REF] Reorganize submodules (#225) @tsalo
• [REF, DOC] Update meta-analysis output map names (#224) @tsalo
• [REF] Rename "permutation" to "montecarlo" (#223) @tsalo
• [DOC] Fix annotate API docs (#222) @tsalo
• [DOC] Improve API rendering and run some examples (#219) @tsalo
• [REF, DOC] Remove unused base classes and improve docs (#216) @tsalo
• [REF] Rename kernel_estimator attribute to kernel_transformer (#197) @tsalo
• [ENH] Make convert_sleuth_to_dataset more flexible (#166) @62442katieb

0.0.2

Changes

• [FIX] Mimic PyMARE's config more completely (#214)
• [FIX] Retain ALESubtraction results (#213)
• [TST] Add CI step for building docs (#212)
• [MAINT] Add workflow to autodeploy to PyPi (#210)
• [FIX] Fix bugs in LDAModel and paths to MALLET (#202)
• [DOC] Fix examples and add gallery back in (#201)
• [REF] Consolidate Dataset loading methods and refactor nimare.extract (#200)
• [ENH] Add extract submodule (again) (#199)
• [ENH] Improve Corrector transparency (#192)
• [TST] Expand test coverage and refactor LDA/CogAt (#193)
• [DOC] Update API documentation (#191)
• [FIX] Fix IO bug (#190)
• [FIX] Fix Neurosynth conversion. (#188)
• [MAINT] Add badges to README and RTD (#187)
• [MAINT] Add PyPi badges
• [MAINT] Add Zenodo DOI badge