| bd4e27f | Kshitij Chawla | 02 April 2020, 00:49:51 UTC | Add Nistats retirement notices (#447) * Updated README.rst * Corrected Warning section tag * Added retirement warning to the NIstats website * Updated retirement warning to the NIstats README.rst * REformatting link to work in stripped README | 02 April 2020, 00:49:51 UTC |
| fdc0b37 | Kshitij Chawla | 31 March 2020, 00:44:57 UTC | Preparation for last release (#446) * Preparation for last release - Added New Deprecation warning - Removed old deprecation warnings and test. - Bumped version to 0.0.1rc - Added a max_version for Nilearn to 0.6.3, preventing installation with Nilearn 0.7.0 . - Updated news section & whats_new. * Removed now obviated tests, added test for library deprecation * Testing installation fails when Nilearn >= 0.7.x * Fixed: Install Nistats wheel, not Nilearn wheel * Normalized version number; removed non-implemented key * Added warnings for Nistats retirements & the unit tests * Added partial test for checking import times warnings & their count * Fixed error in partial test for checking import times warnings & their count * Fixed errors in tests * Always build Universal wheels * Test for expected number of correct warning * Trying to raise error on Grep mismatch * Trying to raise error on warn count mismatch | 31 March 2020, 00:44:57 UTC |
| a2a3928 | Kshitij Chawla | 27 February 2020, 11:11:20 UTC | Updated version for release (#443) | 27 February 2020, 11:11:20 UTC |
| 7b17e51 | Peer Herholz | 05 February 2020, 14:24:28 UTC | Address typos, missing words and structure throughout tutorials (#436) * address typos, missing words and structure throughout tutorials - fix some typos - add missing words - make structure more homogeneous across examples - address further questions and unclear parts in comments * add punctuation to steps - add a dot to all steps in the enumerations * fix too short title underline - circleci exited with errors due to a too short title underline in plot_first_level_model_details.py * Update examples/02_first_level_models/plot_bids_features.py Co-Authored-By: bthirion <bertrand.thirion@inria.fr> * Update examples/05_complete_examples/plot_bids_analysis.py Co-Authored-By: bthirion <bertrand.thirion@inria.fr> * Update examples/03_second_level_models/plot_oasis.py Co-Authored-By: bthirion <bertrand.thirion@inria.fr> * Update examples/05_complete_examples/plot_surface_bids_analysis.py Co-Authored-By: bthirion <bertrand.thirion@inria.fr> Co-authored-by: bthirion <bertrand.thirion@inria.fr> | 05 February 2020, 14:24:28 UTC |
| a7f8da9 | Kshitij Chawla | 27 January 2020, 15:29:48 UTC | Removed 8 space continuation indent to stay inline with Nilearn style (#434) | 27 January 2020, 15:29:48 UTC |
| 4ae4cd6 | Kshitij Chawla | 24 January 2020, 15:26:51 UTC | Fixing Flake8 problems (#430) * Fixing Flake8 problems - 1 * Fixing Flake8 problems - 2 * Fixed incorrect addition of output_type * Placate flake8 - 3 * Placate flake8 - 4 * Fixed Numpydoc build failure * More flake8 fixes | 24 January 2020, 15:26:51 UTC |
| 6f4b962 | Kshitij Chawla | 24 January 2020, 14:16:15 UTC | Added missing whats_new for PR #425 (#433) | 24 January 2020, 14:16:15 UTC |
| 4c3d0e0 | Kshitij Chawla | 24 January 2020, 13:08:15 UTC | Renamed resid methods to resdiuals, w to whitened_ (#425) * Renamed resid methods to resdiuals * Fixed typo and call * Added residuals and resid to tests * Renamed norm_resid to normalized_resuiduals * Undid inadvertent option string change * Reanmed *_resid to *_residuals and added deprecation warning * Reanmed *_resid to *_residuals and .resid to .residuals * Added tests for renamed attributes and associated warnings - Reduced verbosity in warning message. - Renamed wresiduals to whitened_residuals. * Renamed wX and wY to whitened_X, whitened_Y * Replaced wdesign with whitened_design, improved tests | 24 January 2020, 13:08:15 UTC |
| 5be4c0d | Kshitij Chawla | 22 January 2020, 14:33:51 UTC | Replace packages installed from conda-forge by PyPI (#431) We need Nilearn 0.6.1 which will be available on conda-forge in a day or two. No time to wait. | 22 January 2020, 14:33:51 UTC |
| c365505 | jeromedockes | 16 January 2020, 18:19:14 UTC | Design matrices may have different columns (#416) * zip contrasts and design matrices * Added a test * Addressed comments of @kchawla-pi Co-authored-by: bthirion <bertrand.thirion@inria.fr> | 16 January 2020, 18:19:14 UTC |
| 6447ce1 | bthirion | 14 January 2020, 17:41:30 UTC | Improve reporting: Correct alpha default value & enable 2 sided map_threshold (#418) * Changed default for alpha * Added two-sided tests * small improvement * Updated whats_new & Merge 'master' * 'master' of https://github.com/nistats/nistats: Replace numpy's deprecated dec with pytest's skipif (#423) Replace Nosetest library calls with Pytest in code (#422) Replace Nosetests with Pytest as test runner (#419) Removed Appveyor config (Azure Pipelines does the windows tests now) (#420) All-resolution inference (#414) Fix: Missing variables in localizerd dataset are no longer None, but n/a (#415) # Conflicts: # nistats/tests/test_thresholding.py # nistats/thresholding.py Co-authored-by: Kshitij Chawla <kc.insight.pi@gmail.com> | 14 January 2020, 17:41:30 UTC |
| 429f346 | Kshitij Chawla | 14 January 2020, 16:30:44 UTC | Prevent duplication & insertion of reports from damaging website layout (#429) | 14 January 2020, 16:30:44 UTC |
| 27c37cb | Kshitij Chawla | 14 January 2020, 16:06:02 UTC | Fix whatsnew and the website's News/updates section (#428) * Corrected formatting, moved entries to correct version, added Missing entries * Corrected the Updates section for the website | 14 January 2020, 16:06:02 UTC |
| dacd545 | Kshitij Chawla | 14 January 2020, 15:07:44 UTC | GLM Reports: Computed masks are used if there is no user supplied one (#424) * Computed masks are used if there is no user supplied one * Corrected computed mask variable name | 14 January 2020, 15:07:44 UTC |
| 839f8c9 | James Kent | 13 January 2020, 10:38:16 UTC | First level residuals (#410) * Allow for storing residuals in First Level Models * import missing modules and update calls to functions * fix flake8 errors * remove @set_attr_on_read * modify tests to reflect updated code base * fix typo and simplifiy loop * respond to review comments: - add docstrings - re-add @setattr_on_read - change rsq to r_square * change rsq to r_square * change rsq to r_square in tests * fix function calls * Example of how to use for . * Made ValueError for storing model attributes more verbose. * Also include R-squared * fix heading underlines in example * fix grammar Co-Authored-By: bthirion <bertrand.thirion@inria.fr> * fix code formatting and do not standardize * change parameter timeseries to result_as_time_series * attempt to address @bthirion comments * split imports statements * always return list get_voxelwise_model_attribute_ * change docstrings for output to always be a list * modify tests to treat output as list * make _get_voxelwise_model_attribute private and improve documentation * fix formatting of function call * add empty line back in * revert regression.py to master * make result_as_time_series mandatory * add newlines to docs * add newline to end of file * fix missing newline * add James Kent to .mailmap * add entry for the new attributes to FirstLevelModel Co-authored-by: Gilles de Hollander <Gilles.de.Hollander@gmail.com> Co-authored-by: bthirion <bertrand.thirion@inria.fr> Co-authored-by: Kshitij Chawla <kc.insight.pi@gmail.com> | 13 January 2020, 10:38:16 UTC |
| a10c5fe | Kshitij Chawla | 09 January 2020, 16:50:23 UTC | Replace numpy's deprecated dec with pytest's skipif (#423) | 09 January 2020, 16:50:23 UTC |
| 755039f | Kshitij Chawla | 08 January 2020, 12:00:59 UTC | Replace Nosetest library calls with Pytest in code (#422) * Removed use of nose tester from test_check_events_file_uses_tab_separators.py & test_datasets.py * Removed use of nose tester from test_dmtx.py * Removed use of nose tester from test_first_level_model.py * Removed use of nose tester from test_model.py * Removed use of nose tester from test_second_level_model.py * Removed use of nose tester from 3 modules * Removed assert_true & assert_raises from test_utils.py * Removed assert_true & assert_raises from test_thresholding.py * Removed more nose test based calls from * Placate Flake8 * Fixed incorrect word * Pinned Pytest version to 3.9 in certain builds to use tmp_path fixture * Pinned Pytest version to 3.9 in certain builds to use tmp_path fixture - 2 * Tweak version spec syntax * request_mocker fixture available for module wide access * Use pip to force install compatible pytest version - Removed compatibility shim for tmp_path. * Corrected syntax of specifying Pypi version * Tetsing with pytest 4.0 * Pytest installation spec are no longer at multiple places * Pytest installation spec in ☺travis.yml ensuring one place * Removed pathlib import (unnecessary, & Py2 incompatible) * Pinned Pytest version to 3.9 in certain builds to use tmp_path fixture -3 * Replaced lingering pld tmpdir with pytests tmp_path * Replaced assert_equal calls with Pytest compatible assert * Converted Path object to str to comply with existing tests * Now using pytest to skip a test to fix incorrect call duing testing * Py2 DeprecationWarning not raised with Pytest==3.9.1 so increasing it * Added proper spacing | 08 January 2020, 12:00:59 UTC |
| 3bafb76 | Kshitij Chawla | 03 January 2020, 16:37:48 UTC | Replace Nosetests with Pytest as test runner (#419) * Replace nosetests with pytest as test runner * Replaced yield asserts with standard asserts (pytest unsupported) * Replaced nilearn with nistats references * Fixed: Makefile uses tabs not spaces * Moved pytest-cov to always install so --cov option always works * Skip testing reporting module when no matplotlib * Skip testing reporting module when no matplotlib - 2 * Skip testing reporting module when no matplotlib - 3 | 03 January 2020, 16:37:48 UTC |
| 39f789d | Kshitij Chawla | 03 January 2020, 13:32:49 UTC | Removed Appveyor config (Azure Pipelines does the windows tests now) (#420) | 03 January 2020, 13:32:49 UTC |
| 3ffd142 | bthirion | 18 December 2019, 20:13:33 UTC | All-resolution inference (#414) * introduced h value for ARI * small update * Added a bit of code * Made it work + examples * removed unused code * added a test * added documentation and explanations * coerced oversampling factor to int * fixed current comments * added whatsnew entry * added tests * Small improvements in variable names, typo fixes * Minor corrections in typing * Some refinements to var names & user messages * Replaced Nibabel's discontinued .get_data() method with get_data() function | 18 December 2019, 20:13:33 UTC |
| d86362f | bthirion | 18 December 2019, 11:58:07 UTC | Fix: Missing variables in localizerd dataset are no longer None, but n/a (#415) * fixed failing example: missing variables in localizerd dataset are no longer None, but n/a * Removed incorrect referrnce causing CircleCI failure | 18 December 2019, 11:58:07 UTC |
| 32fe4d4 | bthirion | 07 December 2019, 23:45:17 UTC | Added example of beta series regression (#411) * added example of beta series regression * fixing circle ci * Added report * a few more fixes * Small tweaks to code, comment & quality * Tweaking the insertion of the report * Tweaked the documentation language about rhe report * Tweaked the documentation commentation to block unwanted output * added whats_new * remove spurious link * Added link to new example in whats_new * Corrected link * Converted url to anchor (fails when changed) | 07 December 2019, 23:45:17 UTC |
| 464c849 | bthirion | 07 December 2019, 23:16:52 UTC | Integer oversampling value (#412) * coerced oversampling factor to int * cleaning docstring * Small tweaks for docstring & code quality * Revert (ambiguous) truth test * Updated whats_new | 07 December 2019, 23:16:52 UTC |
| e88a5f3 | bthirion | 29 November 2019, 10:07:15 UTC | Introduced fixed effects model (#386) * introduced fixed effects model * Trying to fix circleci * attempt to fix wondows error * started to address comments * addressed a series of comments * fixing windows failure * make mask optional in fixed effects * systematically fit the masker * fixed failing test * Changed var names for clarity; fixed some typos * Only objects necessary for cleanup are deleted in the test * Added proper link & title to example * Changed link role type to troubleshoot sphinx error * Changed link role type to troubleshoot sphinx error - 2 * Corrected hyperlinking in doc | 29 November 2019, 10:07:15 UTC |
| 45894a7 | bthirion | 10 November 2019, 20:10:02 UTC | Merge pull request #403 from jeromedockes/use_a_get_data_function Use a get data function | 10 November 2019, 20:10:02 UTC |
| ff8c31b | bthirion | 10 November 2019, 20:08:27 UTC | Merge pull request #408 from adelavega/reduce-memory Reduce memory by deleting unmasked run_img | 10 November 2019, 20:08:27 UTC |
| 420d072 | delavega4 | 08 November 2019, 18:07:40 UTC | Add comment | 08 November 2019, 18:07:40 UTC |
| 1d485ab | delavega4 | 08 November 2019, 00:26:27 UTC | Add change to whats new | 08 November 2019, 00:26:27 UTC |
| 94c45cf | delavega4 | 07 November 2019, 02:12:49 UTC | Delete unnecessary run_img object after masking | 07 November 2019, 02:12:49 UTC |
| b74b496 | Kshitij Chawla | 29 October 2019, 16:20:13 UTC | Release Nistats 0.0.1b1 prep (#407) * Updated list of contributors & version number for release * Updated mailmap * Updated mailmap - 2 * Set version to 0.0.1b1 | 29 October 2019, 16:20:13 UTC |
| c611ed0 | Jerome Dockes | 27 October 2019, 14:51:35 UTC | Merge remote-tracking branch 'upstream/master' into use_a_get_data_function | 27 October 2019, 14:51:35 UTC |
| c67ec89 | Jessica Thompson | 24 October 2019, 15:36:55 UTC | Comply with BIDS v1.2.0 and BEP003: Common Derivatives in first_level_models_from_bids (#373) * improvement to plot_spm_auditory.py; use make html in doc to generate the autoexample doc * added a fetcher to download the language localizer demo dataset adapted to the BIDS 1.2.0 specification * essai * Reflect latest BIDS specification for derivatives Make fake BIDS dataset according to BEP 3: Common Derivatives * Update to BIDS 1.2.0 and BEP 3: Common Derivatives * remove mention of defunct BIDS key from docstring * Remove references to defunct BIDS preproc keyword * reverted `plot_bids_analysis.py` to use the older version of the langloc dataset * avoid the nested folder struc for langloc example * corrected the langloc dataset fetcher and renamed it `fetch_language_localizer_demo_dataset()` * fix path to langloc dataset so as to prevent re-downloading the dataset if previously downloaded * fix path to langloc dataset returned by fetcher * update to use new BIDS 1.2 version of langloc dataset by calling new fetcher * update to use new BIDS 1.2 version of langloc dataset via new fetcher * add 'res' and 'den' to allowable BIDS keys * download dataset if data_dir is empty instead of if data_dir does not exist * Added test for fetch_language_localizer_demo_dataset * Replaced linux only os.mknod with portable open('w') * Updated whats_new * Corrected fetcher name * Corrected fetcher name in docs reference * Corrected module name in whats_new | 24 October 2019, 15:36:55 UTC |
| a8824fa | jeromedockes | 23 October 2019, 20:47:07 UTC | Remove file local variables (for emacs, vim etc)(#400) | 23 October 2019, 20:47:07 UTC |
| f9ab380 | Kshitij Chawla | 23 October 2019, 20:45:36 UTC | Added error tests for utils functions for code coverage (#404) | 23 October 2019, 20:45:36 UTC |
| c7039f8 | Jerome Dockes | 15 October 2019, 12:32:11 UTC | missing import | 15 October 2019, 12:32:11 UTC |
| 1b951bd | Jerome Dockes | 15 October 2019, 12:29:26 UTC | iter | 15 October 2019, 12:29:26 UTC |
| 335ec31 | Jerome Dockes | 15 October 2019, 12:20:34 UTC | add get_data | 15 October 2019, 12:20:34 UTC |
| ea22644 | Jerome Dockes | 15 October 2019, 09:07:03 UTC | .get_data -> get_data function for most cases | 15 October 2019, 09:07:03 UTC |
| 4b7adc7 | jeromedockes | 11 October 2019, 13:02:21 UTC | Remove dependency on patsy (#399) * remove dependency on patsy * use engine="python" to avoid pandas error when numexpr not installed * special-case passing a single column name as contrast * escape * character * whitespace for sphinx * revert inadvertant unrelated changes * remove patsy from .circleci/config.yml * whitespace in blank line | 11 October 2019, 13:02:21 UTC |
| c3028e5 | Kshitij Chawla | 04 October 2019, 09:19:48 UTC | Remove Boto3 dependency (#393) * Removed Boto3 need by directly downloading urls.json file we uploaded to osf.io * Changed tst to reflect new name for old opneneuro index download function * Added original downloader as a separate utility for maintainers; its test passes - Improved fetch_openneuro_dataset_index() - Imporved docstring & messaging to user for the new utility. * Updated copy of openneuro url index is downloaded independent of existing index - make_openneuro_dataset_urls_index() by default downloads to Desktop. - Simplified make_openneuro_dataset_urls_index() a bit. * Moved test for make_openneuro_dataset_urls_index to nistats.tests.test_utils * Formatting & docstring tweaks * Tests for make_openneuro_dataset_urls_index not run if boto not intalled * Boto3 non-installation raises an error in make_openneuro_dataset_urls_index * Added test for the new openneuro dataset index downloader * Fixed test fail on windows due to improper path separator use * Removed boto3 package from CIs (except one build config); updated docstrings * Fixed numpy decorator import * Conditonal installation of boto3 to run related tests * Removed unreferenced doc links as they error; symbols monospaced in docstring * Fix attempt 1:Conditional install of boto3 in travisci * Fix attempt 2: Conditional install of boto3 in travisci * Rewrote docstrings, added comments, renamed function for clarity * Removed user irrelevant docstring * Fixed docstring typo | 04 October 2019, 09:19:48 UTC |
| 9679bfa | Kshitij Chawla | 03 October 2019, 09:33:11 UTC | HTML Scraper for GLM Reports (#387) * Added Elizabeth's version of HTML Scraper from Nilearn * Added Elizabeth's test of HTML Scraper from Nilearn * Added attribution to Elizabeth Dupre as original author; tweaked an import - tweaked an import in visual inspection suite script * Changed template CSS for Sphinx Gallery scraper (better report presentation) * Changed Bunch import path to fix ImportError * Fixed flake8 non-compliance in test_sphinx_reports.py | 03 October 2019, 09:33:11 UTC |
| f7e1056 | bthirion | 26 September 2019, 07:31:49 UTC | Merge pull request #385 from bthirion/high_pass_misspecification [MRG] warning + upper bound on number of drift columns | 26 September 2019, 07:31:49 UTC |
| 2e9dce7 | bthirion | 23 September 2019, 21:39:01 UTC | better string formatting | 23 September 2019, 21:39:01 UTC |
| 4cec382 | Kshitij Chawla | 13 September 2019, 12:48:11 UTC | Fix Flake8 non-compliance in GLM Reporter PR #352 (#391) * Fixed flake8 non-compliance in glm_reporter.py * Fixed flake8 non-compliance in test_glm_reporter.py * Removed a coverage branch | 13 September 2019, 12:48:11 UTC |
| 4142e00 | bthirion | 13 September 2019, 09:30:43 UTC | Removed unused code (#384) | 13 September 2019, 09:30:43 UTC |
| 7ce6d4e | bthirion | 13 September 2019, 07:33:41 UTC | Merge pull request #390 from bthirion/fix-failing-fir-example [MRG] fixed failing example | 13 September 2019, 07:33:41 UTC |
| 3c094f8 | bthirion | 12 September 2019, 22:09:30 UTC | fixed failing example | 12 September 2019, 22:09:30 UTC |
| 93869f3 | bthirion | 11 September 2019, 14:30:27 UTC | Merge pull request #314 from bthirion/fir_example [WIP] Fir example | 11 September 2019, 14:30:27 UTC |
| 9a8cc92 | Kshitij Chawla | 10 September 2019, 13:34:02 UTC | [MRG] Simple GLM reports (#352) * Initial groundwork for simple GLM reports * Reports show rudimentary contrast info & design matrix; added reporter example * Reports show stat map - Reverted (changeable) to using plot_design_matrix, saved to disk, instead of custom code to embed image into html. * Revert example, moving generate_report() o custom example * Added more parameters for generate_report() - Moved report example script to stop sphinx-gallery error. * Added cluster table to the HTML report * Simple GLM reporter is not hard coded for a specific contrast map, uses the first one - Added reporting example based on FLM Plot Bids Features. - Moved reporting examples to their own directly to prevent sphinx builds. * ENH: Multiple Design matrices, stat maps, cluster tables can be displayed - Refactored Design matrix reporting code - Moved each example script to its own directory to prevent generated file overwrites. - Added example for SLM Oasis Vbm. - Stat map plot files are named by their zmap/contrast name. * Stats maps are displayed with corresponding cluster table - Refactored a few functions. - Using strings.Template functionality creates challenges in easily tweaking layouts. * Cleared up the function interface; added docstrings; reordered functions - Simplified the signature of the reporter function. - Reordered function definitions for easier comprehension. - Changed function parameter names. - Added default values to function parameters. - Added partial docstring to the functions. - Removed arbitrary keyworded arguments support. * GLM Reporter uses model+contrasts to make z-maps; displays model attributes - Design matrix & z_maps are no longer accepted as arguments. - * Reports are generated without excessive HTML strings within python - Reports are now generated by filling a components template which is inserted into the overall page template. * Improved the docstring for how the components are generated * Reports now have titles and headings auto-generated or user-specified - Some docstring corrections * Contrasts can now be strings or collection of strings * Restored: All contrasts listed in the beginning show their arrays if available * Added: make_glm_report(plot_type='glass') to plot glass brain: UI, features TBD * Made elements collapsible using jquery, much WIP * Made cluster table a popup, WIP * Collapsible cLuster table using HTML's <details> * Design Matrices are now svg images inserted directly into the HTML page * Fixed size of Design Matrix image to 640x480 * Stat Map plots are created as SVGs & Directly inserted into the HTML - Size currently set to 1660x229px * make_glm_report() now returns an HTML Document object which can be opened or saved * Added docstrings for all the functions; moved imports into global namespace * Corrected docstring for make_glm_report() * Added some unit tests (WIP) * Correction: Added some unit tests (WIP) * Prepared Cluster Tables to show more info; pared down precision for display * Corrected the Cluster size threshold field's data in cluster tables * If no Cluster size threshold specified, it is displayed as 1 * Additional clustering params; cluster table info ordered; Changed display_modes - The height_control, alpha parameters can be specified for clustering. - Default display_mode for plot_type 'stat' is 'z', for 'glass' is 'ortho'. - Info for cluster table changed, order fixed. - Fixed report size to 1600x800 when displaying in notebooks. - Private function's calls use keyworded args. - Docstrings entries for params updated (WIP). * Changed the attributes displayed in the model attribute table * Refactored preparing contrast info * The Region of Interest is now plotted if the ROI mask img is provided * COntrast matrices are now plotted (WIP) * Contrast plots returned as Dict[contrast name, svg img] for versatile reuse - Added a test for contrat plotting. - Pickled report_slm_oasis data for faster reruns. * Improved layout for contrast plots - Default display_type for glass brain is lzry not ortho. - Added parameter for size for use in Jupyter Notebooks. - glass brain plotting now plots non-absolute values & shows color bar. - Number of subjects in slm oasis report example down to 5 for speed. * Added file for testing contrast plotting functionality * Corrected filepath in test * Added a FirstLevelModel Fiac example * Deactivated some code in reporting/test_glm_reporter.py [WIP] * Added docstring, some reformatting, removed some unused code * Added some tests for the GLM Reporter functionality (WIP) - cluster_threshold is converted to 0 if None to avoid Py3 comparision error. - single contrast_plot is now correctly tagged as svg, not <p>. - all_contrast_plots is now tagged as an object (most compact, visually). * Removed some unused code which was causing fail * Added report templates and module paths to nistats package * Changed the HTML template path to relative * Changed the HTML template path to relative; Attempt 2 * Grafted path to dir of templates * Added reporting as a package in setup.py * include_package_data=True in setup() to include template html files * Handled exception for pandas<0.19 * Removed pandas version dependent test (border parameter in to_html) * Skipping test_flm_fiac_test in Matplotlib v1.5.1 (EOL'd) - CReating the correct environment for this version is difficult now. * Skipping test_flm_fiac_test if Matplotlib is not installed * Checnged default cluster_size value to 0 * Added a number to each plotted design matrix * Removed HTML code mixed in with python code * Removed some unused code/text * Pandas options now change within context only * Reporter now works with design matri(x_/ces_) of FLM & SLM * Contrast plot are more consistent; their titles now part of figure - Activate all contrasts in FIAC report example to better see various contrast plot possibilities. * Showcased reporter functionality in the plot_oasis example - Sphinx is rendering all generated images from the report into the example. TODO: figure out hoe to stop that. * Showcased reporter functionality in the plot_adhd_dmn example - Sphinx is rendering all generated images from the report into the example. TODO: figure out hoe to stop that. * Showcased reporter functionality in the plot_fiac_analysis example - Sphinx is rendering all generated images from the report into the example. TODO: figure out hoe to stop that. * Showcased reporter functionality in the plot_bids_features example - Sphinx is rendering all generated images from the report into the example. TODO: figure out how to stop that. * Some reformatting in reporter example & removed unused code * Fixed failing test * Summary contrasts, design matrices, ROI are plotted as SVG URL images - Summary contrasts, design matrices have their own html sub-template. * Component contrast plots are now SVG URL images & correctly positioned * Component stat maps are now SVG URL images * Providing a list of contrast arrays now works * Fixed error during document generation - Tweaked some example code * Changed organization fo report examples used in dev * Renamed 'stat' arg to 'slice'; single list labels_ are no longer nested - Removed unused imports in test * Contrast names with < & > symbols are now escaped - Renamed fiac report bench func to prevent discovery by test runners, disambiguate from the same named test. * Added docstring for _make_html_for_stat_maps - Renamed statistical_map_img to stat_img. - Tweaked some other docstrings. - Some reformatting. * Improved _make_html_for_stat_maps docstring; renamed vars for pithiness - Invalid plot_type arg raises ValueError. * Renamed _make_html_for_stat_maps > _make_stat_map_svg - Some docstring improvement. * Added Doctring; Renamed _make_html_for_cluster_table > _make_cluster_table_html - Changed parameter order to more closely resemble calling functions. - Renamed some variables for pithiness. * Added doctring for _make_report_components(); renamed statistical_maps to stat_img * Renamed _make_roi_plot > _make_roi_plot_svg & added docstring, comment * Renamed _make_roi_plot_svg > _make_roi_svg * Contrast names & Title are now escaped for HTML special characters * Correction to Contrast names & Title when being escaped for HTML special chars * Corrected parameter arg, changed slices to slice * Fixed a bug in test * Added more docstrings * Added docstrings for make_glm_report, tweaked existing ones * Renamed functions for pithiness * Tweaked the docstrings * Moved GLM Reporter html templates to their own directory * Rename report_components_template.html to components_template.html * Attempting to fix inaccessible templates when nistats is installed as a package * Attempt 0 to Include HTML templates in package * Attempt : 1 include templates in package * Attempt 2: include templates in package * Attempt 3: include templates in package * Added __init__.py * Attempt 4: include template files in package * Attempt 5: include template files in package * Improved _make_contrasts_dict() & added more tests for it * Imporved code readability via refactoring, removed unnecessary test * Fixed error; slightly modified test function * Escaped the `|` character in docstring for RestructuredText * Internally linked references, tweaked docstring formatting * Added an SLM end-to-end smoke test for testing out remaining options - Missing 'labels_' is now correctly handled in model attribute table. - Simplified the report_slm_oasis report example. - Removed commented out code. * KeyError during utf8 encode/decode across Python2/3 is now handled correctly * Added 2 more tests for report headings - Renamed plot_svg to svg_plot * Deactivated a test to assess its utility * Added 2 more tests for test_make_html_for_stat_maps() options - Restored deactivated test. - Removed a file that was (un)used for testing. * Renamed some functions for brevity; Improvements based on feedback - Renamed tests to reflect new names of renamed functions. - Improved docstrings, added missing parameters. - Tweaked some args to avoid confusion. * Init implementation of including thresholding params as part of stat_map_plots * Corrected a function call * Casted dict.values() to list to enable indexing * Generated reports are now (almost) valid HTML5; Applied CSS Styling to tables - Added CSS Styles to tables. - Remove improper <p> tags. - Added alt-text to images. - Removed border attribute from tables. * Generated reports are now valid HTML5; Page headings indicate model type - Validation performed at https://validator.w3.org/ . - Updated tests & docstrings * Added CSS styling to reports - Indentation of HTML file for clarity * Added comments in HTML template indicating which function builds what - Attributes table only shows mask data type or path not the mas object. * make_glm_report(title) accepts a 3 element sequence for Page headings - Added test, modified example accordingly. * Added Nistats repo links to the report footer * Address text & links no longer stand out * ROIs are now labelled masks; function renamed accordingly * Changed cmap for plottingmasks, added additional params for mask plotting - display_mode, thresholding, cmap added to _mask_to_svg() * Removed alpha, threshold in mask plot; reordered cluster details; - Removed alpha, threshold in mask plotting, since different alpha, mask from in rest of function. - alpha now greek symbol. * Model attributes table now shows units where appropriate. * Model attributes table is now sorted alphabetically * Changed examples so each height_control param is used in one example * Refactored & improved several functions, organization, thresholding parameters - Correct alpha & thresholding are used in computation & presentation based on height_control. - Unnested several function calls by one level, improving code organization & comprehension. - This also simplified function signatures. - Changed docstrings accordingly. * Renamed template file; renamed functions & moved their order for clarity - Adjusted tests & comments according to refactorings. * Lots of visual adjustments; refactored plot resizing functionality * Refactor dataframe to html table functionality - Tweaked docstrings & comments. * Reverted some changes/junk code * Replaced a user supplied roi/mask to the one provided by the model * Text "alpha" is displayed instead of symbol in Py2 - Py2 Notebook reports were failing due to HTMLDocument not handling this encoding case. * nb_width, b_height replaced by report_dims; checked for correct values - Added test for this. - Adjusted dev examples. - Added reporting module to nistats init file docstring * Removed min_onset from reported model attributes * Removed mask_img from reported model attributes (can look for mask in model) * Lots of fixes based on code review * Lots of fixes based on code review - 2 - hasattr, getattr instead of object._dict__[] - Changed section about viewing the report in each example. * Tweaked section about viewing the report in each example. * Tweaked section about viewing the report in each example - 2 * More details about contrasts info in docstring * Simplified end-to-end tests for FLM & SLM reports * Simplified title generation function * Set mask display_mode to always be 'z' * Fixed the inadvertent opening of the GLM Report in tests * Removeddisplay_mode parameter from _mask_to_svg() & docstring * Fixed docstring leading to sphinx error * Removed file & config entry for MANIFEST.in * Added whats_new entry * Tweaked docstring; tested non-default args * Added section heading for the GLM report in examples * Added module docstring; tweaked the tests, added comments, & spacing * Consolidated examples for visual inspection of reports, +parallelism * Sphinx-gallery & Jupyter no longer insert all images after function call * Incorporate bthirion's review comments * Contrast plots now display a color bar * Tests artifacts are created within temporary, self-destructing directories - Some comments & docstring tweak. * Delete open objects to fix windows test cleanup failure | 10 September 2019, 13:34:02 UTC |
| 4fd00b6 | bthirion | 06 September 2019, 10:11:58 UTC | added a warning and upper bound on number of drift columns | 06 September 2019, 10:11:58 UTC |
| bc0886b | bthirion | 29 August 2019, 19:20:52 UTC | Merge pull request #381 from gifuni/minor_doc_fix minor corrections of typos | 29 August 2019, 19:20:52 UTC |
| 3af6f3e | Anthony Gifuni | 29 August 2019, 16:46:40 UTC | Update examples/01_tutorials/plot_first_level_model_details.py Co-Authored-By: bthirion <bertrand.thirion@inria.fr> | 29 August 2019, 16:46:40 UTC |
| ae68fba | Anthony Gifuni | 28 August 2019, 18:08:27 UTC | minor corrections of typos | 28 August 2019, 18:08:27 UTC |
| 8f0b606 | Kshitij Chawla | 31 July 2019, 00:40:45 UTC | PRt #374 Directly download & install conda==4.6 [WIP] Directly download & install conda==4.6 | 31 July 2019, 00:40:45 UTC |
| a3e242d | Kshitij Chawla (kchawla-pi) | 31 July 2019, 00:33:26 UTC | Conda==4.6 is now directly downloaded & installed in Travis | 31 July 2019, 00:33:26 UTC |
| 7cb2951 | Kshitij Chawla (kchawla-pi) | 31 July 2019, 00:21:46 UTC | Added a mechanism to force flush package & dataset cache | 31 July 2019, 00:21:46 UTC |
| ffe71e8 | Kshitij Chawla (kchawla-pi) | 31 July 2019, 00:01:46 UTC | Directly downloading & iNstalling conda==4.6 | 31 July 2019, 00:01:46 UTC |
| 82560ff | Kshitij Chawla | 30 July 2019, 11:14:08 UTC | PR #371 Fixed incorrect docstring in run_glm() Fixed incorrect docstring in run_glm() | 30 July 2019, 11:14:08 UTC |
| 2aa7e6d | Kshitij Chawla (kchawla-pi) | 29 July 2019, 08:50:35 UTC | Corrected read_csv() call parameter | 29 July 2019, 08:50:35 UTC |
| fe3ed6d | Kshitij Chawla (kchawla-pi) | 26 July 2019, 09:15:06 UTC | Corrected read_csv() call | 26 July 2019, 09:15:06 UTC |
| c596fe2 | Kshitij Chawla (kchawla-pi) | 26 July 2019, 09:13:23 UTC | Docstring correction | 26 July 2019, 09:13:23 UTC |
| 8bc44d7 | Kshitij Chawla (kchawla-pi) | 26 July 2019, 09:08:47 UTC | Fixed failing tests | 26 July 2019, 09:08:47 UTC |
| c24d60f | Kshitij Chawla (kchawla-pi) | 26 July 2019, 08:57:44 UTC | Corrected attribute names in docstring, fixed failing test | 26 July 2019, 08:57:44 UTC |
| 309bf28 | Kshitij Chawla (kchawla-pi) | 25 July 2019, 09:51:16 UTC | Fixed incorrect docstring in FirstLevelModel | 25 July 2019, 09:51:16 UTC |
| c2e3808 | Kshitij Chawla (kchawla-pi) | 24 July 2019, 16:35:19 UTC | Fixed incorrect docstring | 24 July 2019, 16:35:19 UTC |
| 9fa2e0e | Kshitij Chawla | 05 July 2019, 16:41:49 UTC | PR #357 average rather than sum effect sizes in fixed effect contrasts average rather than sum effect sizes in fixed effect contrasts | 05 July 2019, 16:41:49 UTC |
| a844810 | Kshitij Chawla (kchawla-pi) | 05 July 2019, 16:06:15 UTC | Merge branch 'fix_effect_sizes_are_summed' of git://github.com/jeromedockes/nistats into fix_effect_sizes_are_summed | 05 July 2019, 16:06:15 UTC |
| b866d3f | Jerome Dockes | 05 July 2019, 15:54:29 UTC | Merge remote-tracking branch 'upstream/master' into fix_effect_sizes_are_summed | 05 July 2019, 15:54:29 UTC |
| 0f67789 | Kshitij Chawla (kchawla-pi) | 05 July 2019, 15:35:07 UTC | Merge 'master' after unvendoring sphinx-gallery, numpydoc * 'master' of github.com:nistats/nistats: FIXUP: whats_new.rst FIXUP: fixes in the documentation FIXUP: padding contrast vectors with 5 more zeros ENH: improve comment about drift model DOC: add txt label warning for fmri_glm MISC: change high_pass to default value in plot_single_subject_single_run.py FIXUP: plot_first_level_model_details.py CLEANUP: remove unnecessary files FIXUP: examples FIXUP: replace period cut with high-pass in 1st level model and design matrices corrected pip option to --prefer-binary pip installing nibabel, nilearn binaries to attempt fix for nibabel build error sphinx-gallery is now correctly installed from conda-forge Added sphinx-gallery numpydoc to full-builds Removed nummpydocs; changed ref links in conf.py per new versions Switched plot_gallery arg from bool to string as per new sphinx-gallery version Unvendored sphinx-gallery | 05 July 2019, 15:35:07 UTC |
| a25973b | Kshitij Chawla | 05 July 2019, 14:12:54 UTC | PR #351 Replace period cut with high-pass in 1st level model and design matrices [WIP]replace period cut with high-pass in 1st level model and design matrices | 05 July 2019, 14:12:54 UTC |
| df845d8 | Ana Luisa Pinho | 05 July 2019, 13:42:25 UTC | Merge branch 'master' of github.com:nistats/nistats into low_high_pass_flm_attempt2 * 'master' of github.com:nistats/nistats: corrected pip option to --prefer-binary pip installing nibabel, nilearn binaries to attempt fix for nibabel build error sphinx-gallery is now correctly installed from conda-forge Added sphinx-gallery numpydoc to full-builds Removed nummpydocs; changed ref links in conf.py per new versions Switched plot_gallery arg from bool to string as per new sphinx-gallery version Unvendored sphinx-gallery | 05 July 2019, 13:42:25 UTC |
| f93675c | Kshitij Chawla | 05 July 2019, 12:03:36 UTC | PR #326 Unvendor sphinx-gallery & numpydoc [WIP] Unvendor sphinx-gallery & numpydoc | 05 July 2019, 12:03:36 UTC |
| 54c93bc | Jerome Dockes | 05 July 2019, 10:09:58 UTC | Merge remote-tracking branch 'upstream/master' into fix_effect_sizes_are_summed | 05 July 2019, 10:09:58 UTC |
| a33fd71 | Jerome Dockes | 05 July 2019, 10:05:59 UTC | updtate whatsnew | 05 July 2019, 10:05:59 UTC |
| b804236 | Kshitij Chawla (kchawla-pi) | 05 July 2019, 09:59:45 UTC | Merge branch 'unvendor-sphinx-gallery' of github.com:kchawla-pi/nistats into unvendor-sphinx-gallery * 'unvendor-sphinx-gallery' of github.com:kchawla-pi/nistats: corrected pip option to --prefer-binary pip installing nibabel, nilearn binaries to attempt fix for nibabel build error | 05 July 2019, 09:59:45 UTC |
| d587169 | Kshitij Chawla (kchawla-pi) | 05 July 2019, 09:55:27 UTC | Merged master after pinning conda==4.6 | 05 July 2019, 09:55:27 UTC |
| b8af36d | Ana Luisa Pinho | 04 July 2019, 15:26:31 UTC | Merge branch 'master' of github.com:nistats/nistats into low_high_pass_flm_attempt2 * 'master' of github.com:nistats/nistats: Fixed conda env creation for master Replaced Sphinx's autodoc_default_flags list with autodoc_default_options dict Pinned Conda to version 4.6 due to incompatibilities & errors in 4.7.5 Restored Matplotlib==3.0.3 fixed version Resored CircleCI config to before April 2019 sprint CircleCI: MPL==3.0.3; circumvents Nilearn 0.5.2 breaking change in MPL3.1 fetch_surf_fsaverage5 -> fetch_surf_fsaverage | 04 July 2019, 15:26:31 UTC |
| 3249ec5 | Kshitij Chawla | 04 July 2019, 15:21:04 UTC | Merge branch 'master' into unvendor-sphinx-gallery | 04 July 2019, 15:21:04 UTC |
| a30597a | Kshitij Chawla | 04 July 2019, 14:39:21 UTC | Fixed conda env creation for master | 04 July 2019, 14:39:21 UTC |
| ce825eb | Kshitij Chawla | 04 July 2019, 12:53:52 UTC | PR #364 Restore CircleCI config to before April 2019 sprint Restore CircleCI config to before April 2019 sprint | 04 July 2019, 12:53:52 UTC |
| 15d7c61 | Ana Luisa Pinho | 04 July 2019, 12:33:41 UTC | FIXUP: whats_new.rst | 04 July 2019, 12:33:41 UTC |
| 368eed4 | Kshitij Chawla | 04 July 2019, 11:29:05 UTC | Merge branch 'master' into circleci-presprint | 04 July 2019, 11:29:05 UTC |
| 6759ced | Kshitij Chawla | 04 July 2019, 10:52:45 UTC | PR #366 Pinned Conda to version 4.6 due to incompatibilities & errors in 4.7.5 Reassessment in 1 month. | 04 July 2019, 10:52:45 UTC |
| c998833 | Kshitij Chawla (kchawla-pi) | 03 July 2019, 12:30:01 UTC | Replaced Sphinx's autodoc_default_flags list with autodoc_default_options dict - Stop a warning possibly causing error during make html-strict. | 03 July 2019, 12:30:01 UTC |
| 389ecb5 | Kshitij Chawla (kchawla-pi) | 03 July 2019, 12:03:03 UTC | Pinned Conda to version 4.6 due to incompatibilities & errors in 4.7.5 - Reassessment in 1 month. | 03 July 2019, 12:03:03 UTC |
| 7b40ba4 | Kshitij Chawla (kchawla-pi) | 02 July 2019, 12:50:43 UTC | Restored Matplotlib==3.0.3 fixed version | 02 July 2019, 12:50:43 UTC |
| aacc669 | Kshitij Chawla (kchawla-pi) | 02 July 2019, 12:11:53 UTC | Resored CircleCI config to before April 2019 sprint | 02 July 2019, 12:11:53 UTC |
| 57c509e | Jerome Dockes | 12 June 2019, 13:50:42 UTC | Merge remote-tracking branch 'upstream/master' into fix_effect_sizes_are_summed | 12 June 2019, 13:50:42 UTC |
| 7b3f94c | Kshitij Chawla | 12 June 2019, 12:13:08 UTC | PR #360 Pin MPL==3.0.3 to fix CircleCI failure due to breaking change in MPL3.1 | 12 June 2019, 12:13:08 UTC |
| 8e3a678 | Kshitij Chawla (kchawla-pi) | 12 June 2019, 12:03:30 UTC | CircleCI: MPL==3.0.3; circumvents Nilearn 0.5.2 breaking change in MPL3.1 | 12 June 2019, 12:03:30 UTC |
| c303adf | Kshitij Chawla | 12 June 2019, 00:02:15 UTC | PR #359 Replace Nilearn's deprecated fetch_surf_fsaverage5 call with fetch_surf_fsaverage fetch_surf_fsaverage5 -> fetch_surf_fsaverage | 12 June 2019, 00:02:15 UTC |
| ba7b0d9 | Jerome Dockes | 11 June 2019, 22:34:44 UTC | fetch_surf_fsaverage5 -> fetch_surf_fsaverage | 11 June 2019, 22:34:44 UTC |
| b80487a | Ana Luisa Pinho | 11 June 2019, 16:51:42 UTC | FIXUP: fixes in the documentation | 11 June 2019, 16:51:42 UTC |
| bb08a71 | Jerome Dockes | 11 June 2019, 14:56:54 UTC | average rather than sum effect sizes in fixed effect contrasts | 11 June 2019, 14:56:54 UTC |
| 77021af | Ana Luisa Pinho | 02 June 2019, 21:48:37 UTC | FIXUP: padding contrast vectors with 5 more zeros | 02 June 2019, 21:48:37 UTC |
| e0855de | Ana Luisa Pinho | 23 May 2019, 22:09:14 UTC | ENH: improve comment about drift model | 23 May 2019, 22:09:14 UTC |
| 308dce5 | Ana Luisa Pinho | 23 May 2019, 11:52:26 UTC | DOC: add txt label warning for fmri_glm | 23 May 2019, 11:52:26 UTC |
| 4839b70 | Ana Luisa Pinho | 22 May 2019, 23:09:31 UTC | MISC: change high_pass to default value in plot_single_subject_single_run.py | 22 May 2019, 23:09:31 UTC |
| 632f78a | Ana Luisa Pinho | 22 May 2019, 22:52:15 UTC | FIXUP: plot_first_level_model_details.py | 22 May 2019, 22:52:15 UTC |
| 8635590 | Ana Luisa Pinho | 22 May 2019, 22:29:42 UTC | CLEANUP: remove unnecessary files | 22 May 2019, 22:29:42 UTC |
