* Drop Python 3.6 and 3.7. Support Python 3.11

* Drop 3.11

* regex=True for pandas 2.0.0

remove codecov

change readthedocs badge from latest to stable

* fix latest to stable...

* add easier link to neurolibre publication

* Update examples/README.txt

Co-authored-by: Taylor Salo <tsalo006@fiu.edu>

---------

Co-authored-by: Taylor Salo <tsalo006@fiu.edu>

* Support pre-generated maps in CorrelationalDecoder

* [skip ci] fix circular import

* Test new functionality

* Add load_imgs method to CorrelationDecoder

* fix linting issues

* Update continuous.py

* Return a MetaResult object

* Improve coverage

* Update base.py

* Apply @adelavega suggestions. change `fit` and `load_imgs` to work inplace

* Update continuous.py

* Apply @adelavega suggestions.

make indexed_gzip install optional

* Remove RC versions from Changelog

* Update .readthedocs.yml

* Replace _get_clusters_table with nilearn's get_clusters_table

* Update utils.py

* Pin minimum version of nibabel

* Pin minimum version of numpy

* Pin minimum version of pandas

* Pin minimum version of scikit-learn

* Pin minimum version of scipy

* Support Python 3.11

* Pin minimum version of numpy

* Pin minimum version of nibabel

* update setup.cfg

* Update setup.cfg

* [skip ci] add setuptools to doc build

* [skip ci] do not use system packages

* add cbmr.py file

* create a design matrix function for cbmr

* add test file for cbmr

* modify pre-process and training function in cbmr

* modify the dataset.anotations in cbmr

* add documentation in utils functions

* update model structure

* update optimizer function

* [skip ci][wip] update loss function

* update _fit function in CBMR

* [wip][skip ci] allow other data types as pre-process inputs

* use a sparse array instead of numpy

* [skip ci][wip] allow for multiple-group cbmr

* [skip ci][wip] fix conflict to merge

* [skip CI][wip] modify settings in pre_process

* [skip ci][wip] implemented group-wise CBMR and fix problems

* [skip ci][wip] add results as inputs to MetaResults

* [skip ci][wip] modify standardization of group moderators

* [skip ci][wip] implement NB regression

* [skip ci][wip]remove vox2idx function and simplify the code

* [skip ci][wip]develp CNB model

* adjustment to Firth penalty

* [skip CI][wip] implement index2voxel function

* [skip CI][wip] add implementation for SE of regression coefficient

* [skip CI][WIP] implementing CBMRInference

* [skip CI][wip] implement spatial homogeneity test

* [skip ci][wip] implement CBMRInference group-wise comparison

* formalize GLH contrast variable

* [skip ci][wip] implemented Cov in GLH for all three models

* [skip CI][wip] add a demonstration for CBMREstimator & CBMRInference

* [skip CI][wip] modify example files for demonstrating CBMR

* add documentation to functions.

* solve some issues suggested by flake8

* [skip CI][WIP] fix a bug in log-likelihood function of CNB model

* [skip CI][WIP] Update code according to comments

* [skip CI][WIP] solve conflicts in code

* restructure code

* [skip CI][WIP] replace variable name and remove study_level_moderators

* [skip CI][WIP] changed variables names to be more intuitive.

* reorganize model classes to be partially initialized

* [skip CI][WIP] set some params as attribute of CBMREstimator Class.

* restruct inference code to models

* add some code for overdispersion model class.

* change model to use optimizer

* change model names

* refactor the optimizer functions into the model class

* create a fit method for models

* add summary to model fit

* function name suggestions

* make square_root an attribute

* allow categorical variables in CBMR

* fix  a bug

* new changes on inference class

* solve conflict

* restruct code in CBMRInference

* add documentation foor create_contrast function

* add new steps: remove duplicate rows in contrast matrix

* modify documentation and comments

* change function name to snake case

* restruct code and remove repetition

* reconstruct code, remove repeated code

* correct testing cases of z_to_p function

* add regular expression code to CBMRInference

* [skip CI][WIP] update example file based on reconstructed code

* [skip CI][WIP] Tried standardized categorical covariates

* Raise deprecation warnings with Python 3.6 and 3.7 (#754)

* Add deprecation warnings for Python 3.6 and 3.7

* Remove arrays from exclude in Github action

* Run tests and minimum dependencies on python 3.8

* Run linting and publish on 3.8

* Ignore D401 warnings

* Update setup.cfg

* Update testing.yml

* [MAINT] Fix various errors due to major version changes in dependencies (#757)

* bump matplotlib version to handle new nilearn release

* restrict numpy versions due to numba issue

* make nibabel less than version 5.0

* Remove "dataset" `return_type` option from kernel transformers (#752)

* Remove "dataset" `return_type` option from kernel transformers

* Drop tests ma_map_reuse

* Update test_meta_kernel.py

* Update test_meta_kernel.py

* Update nimare/meta/kernel.py

Co-authored-by: Taylor Salo <tsalo90@gmail.com>

* Update nimare/meta/kernel.py

Co-authored-by: Taylor Salo <tsalo90@gmail.com>

* Add versionchanged to child classes

Co-authored-by: Taylor Salo <tsalo90@gmail.com>

* [skip ci] Update CHANGELOG

* Support nibabel 5.0.0 (#762)

* Add header to `Nifti1Image` when passing an int64 array

* @tsalo Apply suggestions from code review

* Update test_annotate_gclda.py

* Do not zero out one-tailed z-statistics for p-values > 0.5 (#693)

* Do not zero out one-tailed z-statistics for p-values > 0.5

* add comment

* Replace negative values in z with values estimated by CDF

* Add test for p to z conversion

* @yifan0330 Apply suggestions from code review

* Random tests failing. Run black

* Link to NeuroStars software support category instead of neuro questions (#768)

* Link to software support category.

* Add links to GitHub repo.

* Update nilearn URL.

* Revert "Do not zero out one-tailed z-statistics for p-values > 0.5" (#769)

* Revert "Do not zero out one-tailed z-statistics for p-values > 0.5 (#693)"

This reverts commit 87964b8.

* Solve `black` issues

* Add a note about why we zero out negative z-scores

* create a design matrix function for cbmr

* [skip CI][WIP] solve conflicts

* update model structure

* use a sparse array instead of numpy

* [skip CI][WIP] solve conflict

* solve conflicts.

* [skip ci][wip] modify standardization of group moderators

* [skip CI][wip] implement index2voxel function

* [skip CI][wip] add implementation for SE of regression coefficient

* [skip CI][wip] add a demonstration for CBMREstimator & CBMRInference

* [skip CI][WIP] fix a bug in log-likelihood function of CNB model

* [skip CI][WIP] Update code according to comments

* refactor the optimizer functions into the model class

* create a fit method for models

* allow categorical variables in CBMR

* restruct code in CBMRInference

* [skip CI][WIP] update example file based on reconstructed code

* solve conflict

* [skip CI][WIP] solve conflicts

* solve conflicts

* [skip CI][WIP] work on example file

* [skip CI][WIP] complete example file for cbmr.

* [skip CI][WIP] implement an option to specify the reference subtype for categorical moderators.

* [skip CI][WIP] rewrite cbmr example in py file.

* [skip CI][WIP] modify corrector class to be consistent with cbmr outputs

* [skip CI][WIP] add FDR/FWE correction methods to test

* add testing cases with more coverage for CBMREstimator

* [skip CI] [WIP] added new changes

* run black and isort

* wip: working through refactor

* more refactor

* remove debug info

* fix errors

* test firth penalty

* black formating

* more formatting

* remove peaks2maps

* remove redundant def

* change documentation line

* move patsy into function

* add necessary installs

* update example notebook with api

* increase spacing and tolerance

* fix estimator name

* sync utils with main

* update to main on z_to_p test

* remove conperm workflow

* remove whitespace

* fix some errors

* make explicit where to document

* make StandardizeField a transformer

* add functorch for python 3.6

* try to use older version of functorch

* loosen restriction

* fix bugs in cbmr example file

* [skip CI][WIP] fix bugs in testing function for cbmr_update

* add documentation for models.py

* add documentation for cbmr.py

* add documentation for utils.py

* add description for CBMREstimator

* change lr to a smaller value

* edit description function and add reference.

* check if result.__description is a string.

* resolve merge conflict

* set random seed

* simplify the log-likelihood function of NB model

* simplify the log-likelihood function of NB model

* implement wald test for CBMRInference

* edit testing function for cbmr

* edit testing function for correctors

* fix linting error

* fix linting error

* fix linting error

* fixed linting error

* fix linting error

* fix linting error

* fix linting error

* fix linting error

* fix linting error

* fix linting error

* remove unused test datasets

* fix linter error

* add cbmr to docs/api.rst

* edit example file for cbmr

* remove the standardize_field function as it's replicated in the StandardizeField class

* use pass instead of return in the abstract methods

* added a test for StandardizeField class.

* edit example file of cbmr methods.

* fix linter error.

* fix a linter error.

* fix a linter error

* fix names of notebooks

* remove functorch (it was absorbed into torch)

---------

Co-authored-by: Yifan Yu <pra123@rescomp1.hpc.in.bmrc.ox.ac.uk>
Co-authored-by: Yifan Yu <pra123@compg007.hpc.in.bmrc.ox.ac.uk>
Co-authored-by: Yifan Yu <pra123@compg005.hpc.in.bmrc.ox.ac.uk>
Co-authored-by: James Kent <jamesdkent21@gmail.com>
Co-authored-by: Yifan Yu <pra123@compg006.hpc.in.bmrc.ox.ac.uk>
Co-authored-by: Julio A. Peraza <52050407+JulioAPeraza@users.noreply.github.com>
Co-authored-by: Taylor Salo <tsalo90@gmail.com>
Co-authored-by: jdkent <jdkent@users.noreply.github.com>
Co-authored-by: Taylor Salo <salot@pennmedicine.upenn.edu>

…rlo"` (#803)

* Set n_iters=10000 default only for null_method="montecarlo"

* update default

* handle case of no sample size being reported

* handle sample size metadata

* fix math mode and links in cbmr example.

* fix math mode

* Add Corrector and Diagnostics attributes to MetaResult object

* Update diagnostics.py

* fix documentation

* Update ale.py

* add new bids-like modalities

…to dataset (#808)

allow analysis to have null points

* Add reports module

* Add template

* Add config file

* Add generate_figues script

* Add test

* Update test_reports.py

* Update setup.cfg

* Fix lint error

* Add yaml to setup file

* Update setup.cfg

* Update default.yml

* Update setup.cfg

* Add report to example gallery

* check online path

* Plot coordinates

* Update 10_plot_cbma_workflow.py

* Update 10_plot_cbma_workflow.py

* Update figures.py

* Update 10_plot_cbma_workflow.py

* [skip ci] update figures

* [skip ci] add static plot

* [skip ci] update path to report

* [skip ci] update path

* [skip ci] explore rtd directory tree

* [skip ci] high res images

* [skip ci] update path to report

* [skip ci] close fig after saving

* [skip ci] update path

* [skip ci] try another path the report.html

* [skip ci] change the path again.............

* [skip ci] update path

* [skip ci] update path

* [skip ci] add cluster table

* [skip ci] add title, caption, and description

* [skip ci] update example

* [skip ci] increase size of heatmap

* [skip ci] add ids of studies excluded to summary

* [skip ci] update table style

* [skip ci] add dynamic coordinates plot

* [skip ci] add legend

* [skip ci] embed heatmaps wit iframe

* [skip ci] change names to fit new bids-like format

* [skip ci] pin pandas version

* [skip ci] add summary diagnostics

* [skip ci] work on copy of diagnostics

* [skip ci] get threshold from diagnostics

* [skip ci] add all possible combinations of results

* [skip ci] add message when tables and map are empty

* [skip ci] add alpha to dictionary

* [skip ci] add label maps

* [skip ci] add tail label

* [skip ci] add mask

* add documentation

* fix typo

* add tables naming convention to output

* only plot coordinates for CBMAEstimator

* fix typo

* improve coverage

* reduce report and yml file