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