==================================== Release notes for DIPY version 1.3 ==================================== GitHub stats for 2020/09/09 - 2020/11/02 (tag: 1.2.0) Highlight: --------- - Gibbs Ringing correction 10X faster. - Spherical harmonics basis definitions updated. - Added SMT2 metrics from mean signal diffusion kurtosis. - New interface functions added to the registration module. - New linear transform added to the registration module. - New tutorials for DIPY command line interfaces. - Fixed compatibility issues with different dependencies. - Tqdm (multiplatform progress bar for data downloading) dependency added. - Large documentation update. - Bundle section highlight from BUAN added in Horizon. - Closed 134 issues and merged 49 pull requests. The following 14 authors contributed 284 commits. * Areesha Tariq * Ariel Rokem * Basile Pinsard * Bramsh Qamar * Charles Poirier * Eleftherios Garyfallidis * Eric Larson * Gregory Lee * Jaewon Chung * Jon Haitz Legarreta Gorroño * Philippe Karan * Rafael Neto Henriques * Serge Koudoro * Siddharth Kapoor We closed a total of 134 issues, 49 pull requests and 85 regular issues; this is the full list (generated with the script :file:`tools/github_stats.py`): Pull Requests (49): * :ghpull:`2181`: BUAN bundle highlight option in horizon * :ghpull:`2223`: [NF] new linear transforms for Rigid+IsoScaling and Rigid+Scaling * :ghpull:`2238`: [FIX] fix cython error from pre matrix * :ghpull:`2265`: Gibbs denoising: fft only along the axis of interest * :ghpull:`2206`: NF: Update definitions of SH bases and documentation * :ghpull:`2266`: STYLE: minor refactoring in TissueClassifierHMRF * :ghpull:`2255`: Modifying dti.design_matrix to take gtab.btens into account * :ghpull:`2271`: Increase Azure pipeline timeout * :ghpull:`2263`: [FIX] update multiple models due to cvxpy 1.1 (part2) * :ghpull:`2259`: [Fix] Allow read_bvals_bvecs to have 1 or 2 dwi volumes only * :ghpull:`2264`: BF: Fix `dipy_align_syn` default value assumptions * :ghpull:`2268`: BUG: Fix literal * :ghpull:`2267`: BUG: Fix string literal * :ghpull:`2262`: [FIX] update tests to respect numpy NEP 34 * :ghpull:`2244`: DOC : Denoising CLI * :ghpull:`2119`: RecoBundles updated to read and save .trk files from Old API * :ghpull:`2260`: [Fix] Better error handling in Diffeomorphic map `get_map` * :ghpull:`2258`: [FIX] update Azure OSX CI + remove Azure Linux CI's * :ghpull:`2257`: [Fix] warning if not the same number of points * :ghpull:`2261`: [DOC]:Removed tracking evaluation section * :ghpull:`1919`: [DOC] Add an overview of reconstruction models * :ghpull:`2256`: update BUAN citations * :ghpull:`2253`: Improve FFT efficiency in gibbs_removal * :ghpull:`2240`: [ENH] Deprecate interp parameter name in AffineMap * :ghpull:`2198`: Make single and multi tensor simulations compatible with btensors * :ghpull:`2025`: Adds an align.api module, which provides simplified API to registration functions * :ghpull:`2197`: Estimate smt2 metrics from mean signal kurtosis * :ghpull:`2227`: RF: Replaces our own custom progressbar with a tqdm progressbar. * :ghpull:`2250`: [ENH] Add parallelization to gibbs denoising * :ghpull:`2252`: BUG: Set tau factor to parameter value in local PCA * :ghpull:`2248`: [DOC] fetching dataset * :ghpull:`2249`: [fix] fix value_range in HORIZON * :ghpull:`2247`: BF: In LiFE, set nan signals to 0. * :ghpull:`2246`: [DOC] Replace simple backticks with double backticks * :ghpull:`2239`: [ENH] Add inplace kwarg to gibbs_removal * :ghpull:`2242`: maintenance of bundle_shape_similarity function * :ghpull:`2241`: STYLE: Exclude package information file from PEP8 checks * :ghpull:`2235`: DOC: Add tips to the documentation build section * :ghpull:`2234`: DOC: Improve some of the links in the `info.py` file * :ghpull:`2233`: Clarifying msmt response function docstrings * :ghpull:`2231`: DOC: Fix HTML tag in dataset list documentation table * :ghpull:`2221`: Robustify solve_qp for possible SolverError in one odd voxel * :ghpull:`2226`: STYLE: Conform to `reStructuredText` syntax in examples sections * :ghpull:`2225`: [CI] Replace Rackspace by https://anaconda.org/scipy-wheels-nightly * :ghpull:`2224`: Replace pytest.xfail by npt.assert_raises * :ghpull:`2220`: [DOC] move Denoising on its own section * :ghpull:`2218`: DOC : inconsistent save_seeds documentation * :ghpull:`2217`: Fixing numpy version rcond issue in numpy.linalg.lstsq * :ghpull:`2211`: [FIX] used numerical indices for references Issues (85): * :ghissue:`2181`: BUAN bundle highlight option in horizon * :ghissue:`2272`: DOC : Registration CLI * :ghissue:`2223`: [NF] new linear transforms for Rigid+IsoScaling and Rigid+Scaling * :ghissue:`2180`: [NF] add new linear transforms for Rigid+IsoScaling and Rigid+Scaling * :ghissue:`2238`: [FIX] fix cython error from pre matrix * :ghissue:`2265`: Gibbs denoising: fft only along the axis of interest * :ghissue:`2206`: NF: Update definitions of SH bases and documentation * :ghissue:`392`: mrtrix 0.3 default basis is different from mrtrix 0.2 * :ghissue:`2266`: STYLE: minor refactoring in TissueClassifierHMRF * :ghissue:`2255`: Modifying dti.design_matrix to take gtab.btens into account * :ghissue:`2271`: Increase Azure pipeline timeout * :ghissue:`2054`: Discrepancy between dipy.gibbs.gibbs_removal and reisert/unring/ * :ghissue:`2263`: [FIX] update multiple models due to cvxpy 1.1 (part2) * :ghissue:`2190`: Reconstruction with Multi-Shell Multi-Tissue CSD * :ghissue:`2259`: [Fix] Allow read_bvals_bvecs to have 1 or 2 dwi volumes only * :ghissue:`2046`: read_bvals_bvecs can't read a single volume dwi * :ghissue:`2264`: BF: Fix `dipy_align_syn` default value assumptions * :ghissue:`2268`: BUG: Fix literal * :ghissue:`2267`: BUG: Fix string literal * :ghissue:`2262`: [FIX] update tests to respect numpy NEP 34 * :ghissue:`2132`: Generating ndarrays with different shapes generates NumPy warning at testing * :ghissue:`1266`: test_mapmri_isotropic_static_scale_factor failure on OSX buildbot * :ghissue:`1264`: FBC measures test failure on PPC * :ghissue:`2244`: DOC : Denoising CLI * :ghissue:`2119`: RecoBundles updated to read and save .trk files from Old API * :ghissue:`2117`: RecoBundles workflow still using old API * :ghissue:`2260`: [Fix] Better error handling in Diffeomorphic map `get_map` * :ghissue:`2202`: Add error handling in Diffeomorphic map `get_map` * :ghissue:`2258`: [FIX] update Azure OSX CI + remove Azure Linux CI's * :ghissue:`2257`: [Fix] warning if not the same number of points * :ghissue:`342`: Missing a warning if not the same number of points * :ghissue:`2261`: [DOC]:Removed tracking evaluation section * :ghissue:`2115`: Independent section on Fiber tracking evaluation not necessary * :ghissue:`1744`: [WIP] [NF] Free Water Elimination for single-shell DTI (updated version) * :ghissue:`1919`: [DOC] Add an overview of reconstruction models * :ghissue:`1489`: Documentation: how to know which models support multi-shell? * :ghissue:`2256`: update BUAN citations * :ghissue:`2253`: Improve FFT efficiency in gibbs_removal * :ghissue:`2240`: [ENH] Deprecate interp parameter name in AffineMap * :ghissue:`2192`: Bringing AffineMap and DiffeomorphicMap a little closer together * :ghissue:`2198`: Make single and multi tensor simulations compatible with btensors * :ghissue:`2025`: Adds an align.api module, which provides simplified API to registration functions * :ghissue:`2201`: Gradient table message error * :ghissue:`2232`: This should be len(np.unique(gtab.bvals)) - 1 or somesuch * :ghissue:`2197`: Estimate smt2 metrics from mean signal kurtosis * :ghissue:`2227`: RF: Replaces our own custom progressbar with a tqdm progressbar. * :ghissue:`2219`: Replace fetcher progress bar with tqdm * :ghissue:`2250`: [ENH] Add parallelization to gibbs denoising * :ghissue:`2236`: Parallelize gibbs_removal * :ghissue:`2254`: Trackvis header saved with Dipy (nibabel) is not read by Matlab or other tools * :ghissue:`2252`: BUG: Set tau factor to parameter value in local PCA * :ghissue:`2251`: localpca tau_factor is hard coded to 2.3 * :ghissue:`2248`: [DOC] fetching dataset * :ghissue:`2249`: [fix] fix value_range in HORIZON * :ghissue:`2243`: Unable to visualize data through dipy_horizon * :ghissue:`2247`: BF: In LiFE, set nan signals to 0. * :ghissue:`2246`: [DOC] Replace simple backticks with double backticks * :ghissue:`2239`: [ENH] Add inplace kwarg to gibbs_removal * :ghissue:`2237`: gibbs_removal overwrites input data when inputting 3-d or 4-d data. * :ghissue:`2245`: DOC: Fix Sphinx verbatim syntax in coding style guide * :ghissue:`2242`: maintenance of bundle_shape_similarity function * :ghissue:`2241`: STYLE: Exclude package information file from PEP8 checks * :ghissue:`2235`: DOC: Add tips to the documentation build section * :ghissue:`2234`: DOC: Improve some of the links in the `info.py` file * :ghissue:`2222`: How can I track different streamlines in DIPY? * :ghissue:`2233`: Clarifying msmt response function docstrings * :ghissue:`2231`: DOC: Fix HTML tag in dataset list documentation table * :ghissue:`2230`: TST: Assert the shape of the output based on the docstring. * :ghissue:`2228`: Best practices for saving a tissue classifier object? * :ghissue:`2221`: Robustify solve_qp for possible SolverError in one odd voxel * :ghissue:`2109`: DIPY lab meetings, Spring 2020 * :ghissue:`2226`: STYLE: Conform to `reStructuredText` syntax in examples sections * :ghissue:`2225`: [CI] Replace Rackspace by https://anaconda.org/scipy-wheels-nightly * :ghissue:`2214`: Rackspace does not exist anymore -> Update PRE-matrix on Travis and Azure required * :ghissue:`2224`: Replace pytest.xfail by npt.assert_raises * :ghissue:`2220`: [DOC] move Denoising on its own section * :ghissue:`2218`: DOC : inconsistent save_seeds documentation * :ghissue:`2217`: Fixing numpy version rcond issue in numpy.linalg.lstsq * :ghissue:`2216`: test_multi_shell_fiber_response failed with Numpy 1.13.3 * :ghissue:`2211`: [FIX] used numerical indices for references * :ghissue:`2185`: Inconsistency in stating references in dti.py * :ghissue:`2215`: problem with fetching stanford data * :ghissue:`1762`: Font on instructions is small on mac * :ghissue:`1354`: strange tracks * :ghissue:`325`: streamline extraction from eudx is failing - but perhaps eudx is failing