Skip to content

Latest commit

 

History

History
663 lines (621 loc) · 25.9 KB

CHANGELOG.md

File metadata and controls

663 lines (621 loc) · 25.9 KB
Raw

2023-08-20 [ 1.12.0 ]:

  • bugfixes
    • Fixed prescrump, scrump, and scraamp miscalculation of AB-joins
    • Fixed bug in is False by converting to bool type
    • Fixed bug in snippet that caused loss in precision, add unit tests
    • Fixed RTD incompatibility with urllib3
    • Fixed loss of precision in distances computed for self-matches
    • Fixed libiomp5.dylib Github Actions location
    • Fixed incorrect post-processing in naive.mass_PI
    • Fixed incorrect number of contiguous windows in snippets
  • features
    • Improved matrix profile performance (15-20%) with uint64 indexing
    • Added top-k nearest neighbor feature
    • Added rolling_isconstant function
    • Added subseq_isconstant to API for transparent handling of constant time series subsequences
    • Added parallelized rolling_nanstd
    • Added abstraction layer for distributed client functions (e.g., Dask, Ray, etc)
    • Added initial support for numba function caching
    • Added Python 3.11 support
    • Added query_idx to improve distances computed for self-matches in motifs function
    • Added mmotifs for multi-dimensional motif discovery
    • Added function process_isconstant
  • tasks
    • Refactored mpdist
    • Added MERLIN notebook reproducer
    • Added negative index checks to mmotifs
    • Fixed mybinder badge and links
    • Renamed nave.mass to naive.mass_PI
    • Added row_wise parameter to naive.stump
    • Added input type (list) check to ostinato
    • Added ability to return fully filled bfs indices
    • Updated setup.py to enable Github dependency graph tracking
    • Improved stability of prescrimp
    • Split coverage tests for more verbosity
    • Added more features to custom test function
    • Removed explicit cancellation of dask futures
    • Removed max dask/distributed version requirement
    • Optimized stumpi and aampi class init
    • Improve aampi update behavior with constant sequences returning nan
    • Added pytest notebook link checking
    • Refactored match function
    • Added warnings to motifs and aamp_motifs
    • Added numba -s step in Github Actions workflow
    • Added explicit link to OpenMP for MacOS Github Actions workflow
    • Moved to Actions/Checkout V3 in Github Actions
    • Improved numba function signatures
    • Added '--editable' install mode to setup.sh
    • Updated Github Action codecov/codecov-action@v1 to v3
    • Maintained 100% code coverage
    • Used unittest.mock.patch to prevent overwriting of config variables during testing
    • Added Python 3.11 to test matrix
    • Added ability to detect missing parameter definitions in docstrings (docstring.py)
    • Updated pytest flags to report skips and added additional summary
    • Replaced logging.warning with warnings.warn
    • Improved multi-line warnings
    • Ensured bfs_indices are sent to correct GPU device
    • Refactored window size check in mass/mass_absolute functions
    • Added animated GIF to README
    • Updated minimum black version
    • Removed _parallel_rolling_func as it conflicted with numba caching
    • Removed mamba timeout
    • Improved warnings
    • Added missing p-norm param to idx_to_mp and floss functions
    • Added test failure when coverage is below 100%
    • Added boolean array test for rolling_isfinite
    • Added isort, resolved circular dependencies, updated examples
    • Only build HTML for RTD
    • Added mp param to stumpi to allow pre-computed matrix profile as input
    • Added various unit tests
    • Removed codecov as dependency
    • Added ability to test the execution of tutorial notebooks
    • Added check for negative matrix profile indices
    • Added test_precision.py for all issues related to loss-of-precision
    • Fixed tls deprecation warning
    • Replaced np.int with np.int64
    • Specified fastmath flags to include nan/inf values in inputs/outputs
    • Replaced bool dtype with np.bool_
    • Replaced np.newaxis with np.expand_dims
    • Added check for docstring and parameter mismatch
    • Update URLs for minimum version references
    • Added explicit shell declaration in Github Actions workflow
    • Show OpenMP libraries in Github Actions workflow
    • Added pip.sh script for setting up dev environment using pip
    • Removed parallel=True in core._compute_multi_PI
    • Updated coverage testing to include all modules
    • Improved code consistency for T_A and T_A definitions
    • Refactored test.sh and include missing test files in unit tests
    • Added minimum dependency compatibility script (min.py)
    • Updated minimum dependency bumping instructions
    • Bumped minimum Python version to 3.8
    • Update PyPI downloads badge
  • documentation
    • Improved/updated various docstrings
    • Added shapelet discovery tutorial
    • Clarified unanchored chain description
    • Fixed typos
    • Improved core._get_QT docstring
    • Fixed imbalanced tree representation in docstring
    • Improved scrump documentation
    • Made light mode default and remove theme switcher from header nav bar
    • Improved dataframe layout display in tutorials
    • Added multi-dimensional motif and match tutorial = Added T_subseq_isfinite to docstring
    • Added tutorial for "Discovering motifs under uniform scaling"
    • Updated docs for using a dask client
    • Fixed malformed link in floss docstring
    • Added missing parameter section in various docstrings
    • Added Minkowski docstring for Euclidean distance
    • Added missing parameters for GPU functions in docstrings
    • Improved documentation for P in motifs function

2022-03-31 [ 1.11.1 ]:

  • bugfixes
    • Fixed #582 Allow 1D mean/stddev inputs for stumpy.match
  • features
    • N/A
  • tasks
    • Added mmotifs and aamp_mmotifs to init.py
  • documentation
    • Added mmotifs docstring in RTD API

2022-03-21 [ 1.11.0 ]:

  • bugfixes
    • Fixed #576 Incorrect stimp, stimped, gpu_stimp normalize rerouting
    • Fixed bad index in naive.stimp normalization method
    • Fixed unit tests
    • Fixed cutoff=np.inf edge case in _motifs function
  • features
    • Added mmotifs and aamp_mmotifs functions for multi-dimensional motif discovery and unit tests
    • Added all AAMP/p-norm stimp implementations
    • Added atol parameter to motifs function
    • Added parallelized prescraamp
    • Added parallelized prescrump
    • Added _get_ranges function
    • Added shoelace formula for computing total diagonal ndists
    • Added p_norm support
    • Added _preprocess function
    • Added minimum description length function, mdl
    • Added Python 3.10 support
    • Added AAMP/p-norm support for pan matrix profiles
  • tasks
    • Fixed typos
    • Added gpu_aamp_stimp driver error
    • Increased minimum dependencies
    • Updated coverage testings to include all unit tests and additional modules
    • Added pip.sh script for setting dev environment
    • Replaced argsort with argmin/argmax
    • Refactored redundant preprocessing steps
    • Replaced deprecated scipy.ndimage.filters module
    • Updated conda environment installation steps in conda.sh script
    • Updated subspace vs subspaces definition
    • Replaced py.test with pytest
    • Updated minimum black version to 22.1.0
    • Replaced elbow method with mdl in multi-dimensional motif tutorial
    • Converted float/int type to np.float64/np.int64
    • Added dtype check and fill value to apply_excl_zone function
    • Added note on anti-correlated subsequences
  • documentation
    • Updated README and various docstrings
    • Added MPdist tutorial draft
    • Added annotation vector tutorial
    • Added geometric time series chains tutorial draft
    • Added top-K motif section to motif discovery tutorial

2021-12-15 [ 1.10.2 ]:

  • bugfixes
    • Fixed #501 Allow max_distance = np.inf in match function
  • features
    • Added atol=1e-8 parameter to match and motifs functions
  • tasks
    • Fixed typos
    • Removed conda download badge, updated PyPI download badge
    • Added removal pycache in test clean up phase
  • documentation
    • Updated pan matrix profile tutorial
    • Updated docstring for match and motifs functions

2021-12-15 [ 1.10.1 ]:

  • bugfixes
    • Fixed #495 Reduce import time by removing Numba NJIT signatures
  • features
    • Added multi_distance_profile function
  • tasks
    • Refactored _query_mstump_profile (see multi_distance_profile)
    • Added SVG STUMPY logo
  • documentation
    • Refactored tutorial

2021-11-02 [ 1.10.0 ]:

  • bugfixes
    • Raise TypeError for non np.float64 input arrays
    • Fixed NumbaPerformanceWarning backwards compatibility
  • features
    • Converted all dtypes to np.float64/np.int64
    • Added explicit NJIT function signatures
    • Added Python 3.9 support
    • Added core._idx_to_mp convenience function
    • Added Nan/Inf support for motifs and matches functions
  • tasks
    • Removed "in-tree" build flag for pip
    • Added pypi stats query for bigquery
    • Set explicit sphinx version for RTD
    • Fixed typos
    • Replaced direct installation via setup.py
    • Added numpydoc
    • Removed ipywidgets from requirements
    • Removed Azure Pipelines from CI
    • Used official OSI name in the license metadata, added license name in classifiers section
    • Added Github Citation BIB format
    • Fixed broken tutorial links
  • documentation
    • Fixed multi-dimensional matrix profile description
    • Added clear warning in discord section of MSTUMP tutorial
    • Added API examples to all docstrings
    • Updated docstrings to use numpy.ndarray
    • Added interactive threshold example for STIMP
    • Updated matplotlib style sheet to use URL for all tutorials

2021-07-28 [ 1.9.2 ]:

  • bugfixes
    • Fixed cutoff parameter not being used in motifs.py
  • features
    • Unified motif discovery and pattern matching tools
  • tasks
    • Added binder link to tutorial
  • documentation
    • Updated pattern matching tutorial
    • Added Pearson correlation notebook
    • Fixed missing sphinx docstring for Python class

2021-07-20 [ 1.9.1 ]:

  • tasks Bumped version

2021-07-20 [ 1.9.0 ]:

  • bugfixes
    • Fixed scenarios where n_chunks == 0 in _get_array_ranges and fixed empty input array edge case
  • features
    • Added normalize to core.mass
    • Added motif discovery (stumpy.motifs and stumpy.match)
    • Added snippets for identifying regimes (stumpy.snippets)
    • Added pan matrix profile (stumpy.stimp, stumpy.stimped, stumpy.gpu_stimp)
    • Added excl_zone parameter to config.py (config.STUMPY_EXCL_ZONE_DENOM)
  • tasks
    • Aggregate or Refactor Dask Unit Tests
    • Added script for testing latest Numba release candidate
    • Converted bash scripts to [[ ... ]] construct
    • Updated Python class declaration
    • Updated to RTD to PyData Sphinx Theme
    • Updated conda installation environment
    • Refactored test files and added check to ensure that naive implementations always come ahead of tests
  • documentation
    • Corrected binder badges to point to "main"
    • Updated tutorial to discover motif/discord indices
    • Added missing docstrings and fixed minor typos
    • Added missing logo file and favicon to _static directory
    • Updated source installation instructions
    • Added instructions for Apple M1 chip
    • Updated Contributor guide
    • Added include/discords tutorial example to subspace
    • Added bonus content on interpreting the columns of a matrix profile
    • Added syntax highlighting to tutorials
    • Replaced matplotlib params with style file
    • Added Annotation Vectors Tutorial
    • Added Binder links to top of tutorials
    • Added introduction to Snippets Tutorial

2021-02-04 [ 1.8.0 ]:

  • bugfixes
    • Fixed chunk size for scrump and scraamp when time series are short
  • features
    • Added maamp and maamped functions
    • Added scraamp function
    • Added a new core.non_normalized decorator that re-routes normalized functions to non-normalized functions
    • All z-normalized functions now accept a normalize parameter
  • tasks
    • Renamed main branch
    • Removed Azure pipelines badge
    • Refactored subspace
    • Refactored non-normalized functions
    • Added non-normalized support to floss
  • documentation
    • Updated README with if __name__ == "__main__" for Dask and Jupyter notebooks
    • Removed all aamp references as normalize=False should be used instead
    • Fixed function docstrings and typos in API docs

2021-01-20 [ 1.7.2 ]:

  • bugfixes
    • None
  • features
    • Added the NEP 29 policy
  • tasks
    • Added CI for minimum version dependencies
  • documentation
    • Updated README to conver NEP 29

2021-01-19 [ 1.7.1 ]:

  • bugfixes
    • None
  • features
    • None
  • tasks
    • Bumped minimum NumPy version to use np.array_equal
  • documentation
    • None

2021-01-17 [ 1.7.0 ]:

  • bugfixes
    • None
  • features
    • Added maximum window size check
    • Added window size checking to preprocessing
  • tasks
    • Replaced array comparison in core.are_arrays_equal with np.array_equal
    • Added Github Actions and badge
    • Improved stability for integration with other packages
  • documentation
    • Fixed typo
    • Replaced NABDConf motivation video with PyData Global video

2020-12-30 [ 1.6.1 ]:

  • bugfixes
    • Fixed inconsistent mstump/mstumped output to match description in published work
  • features
    • Added subspace function for compute multi-dimensional matrix profile subspace
    • Added include and discords handling to subspace
    • Added mdl building blocks
    • Added ostinato Dask distributed, GPU, and aamp variants
    • Added fast rolling min/max functions
  • tasks
    • Updated Azure Pipelines CI coverage tests to use latest images
    • Added mstump tutorial to Binder
    • Fixed bad reference to aamp_gpu_ostinato
    • Fixed aamp_ostinato unit test coverage
    • Converted conda.sh to bash
    • Moved all private functions out of __init__.py
    • Added commands to remove build/, dist/, and stumpy.egg/ after PyPI upload
    • Added twine to environment setup script
    • Fixed incorrect date in CHANGELOG.md
  • documentation
    • Added mstump tutorial to RTD
    • Updated various function docstrings
    • Added Github Discussions to README and RTD
    • Updated API list on RTD

2020-12-10 [ 1.6.0 ]:

  • bugfixes
    • Fixed incorrect cancelling of Dask data that was being scattered with hash=False
    • Fixed floating point imprecision in computing distances with mass_absolute
  • features
    • Added ostinato function for computing consensus motifs
    • Added new approach for retrieving the most central consensus motif from ostinato
    • Added mpdist function for computing the MPdist distance measure
    • Added mpdisted function for computing the MPdist distance measure
    • Added gpu_mpdist function for computing the MPdist distance measure
    • Added aampdist function for computing the MPdist distance measure
    • Added aampdisted function for computing the MPdist distance measure
    • Added gpu_aampdist function for computing the MPdist distance measure
    • Changed np.convolve to the faster scipy.signal.convolve for FFT
    • Added matrix profile subspace for multi-dimensional motif discovery
    • Replaced existing rolling functions with fast Welford nanstd and nanvar functions
  • tasks
    • Updated Azure Pipelines CI to use latest images
    • Fixed tutorial typos
    • Added ostinato paper to README References
    • Added mamba to speed up Python environment installation
    • Added ostinato tutorial
    • Added a series of new unit tests (maintained at 100% test coverage)
    • Updated all tutorials to use Zenodo links for data retrieval
    • Removed tutorial plotting function that set default plotting conditions
    • Added AB-join tutorial
    • Replaced rolling window isfinite with a much faster function
    • Updated Github PR template to use conda-forge channel
    • Updated Welford corrected-sum-of-squares derivation
    • Updated Binder environment to use STUMPY release in Binder postBuild
  • documentation
    • Fixed GPU function signatures that were being displayed on RTD
    • Fixed incorrect docstring indentation
    • Added STUMPY docs and Github code repo to Resources section of tutorials
    • Added default values to docstrings

2020-10-19 [ 1.5.1 ]:

  • bugfixes
    • Fixed AB-join so that it now matches the published definition (previously, BA-join)
  • features
    • Added nan/inf support to FLOSS/FLUSS
  • tasks
    • Removed Pandas series in GPU tests to improve unit test speed in CI
    • Identify operating system prior to installing cuda toolkit
    • Changed left/right keywords in all unit tests to ref/comp
    • Split up unit tests and coverage testing
    • Updated displayNames in Azure Pipelines
    • test.sh now accepts unit, custom, and coverage keywords
    • Fixed typos
    • Added pattern searching (MASS) tutorial
    • Added Contribute notebook to RTD table of contents for first time contributors
    • Refactored _compute_diagonal for speed improvements
    • Replaced np.roll with slice indexing in stumpy.floss
    • Refactored and improved aampi and stumpi update performance
    • Added lxml to environment.yml
  • documentation
    • Added plt.show() to code figures in tutorials
    • Updated stumpi tutorial with egress=False

2020-08-31 [ 1.5.0 ]:

  • bugfixes
    • Fixed warning and check when time series has inappropriate dtype
    • Fixed scenarios where identical subsequences produce non-zero distances
  • features
    • For interactive data science work, matrix profile calcs are 10-15x faster
    • Added aamp with non-normalized Euclidean distance (i.e., no z-normalization)
    • Added aamped
    • Added aampi
    • Added gpu_aamp
    • Added egress for stumpi and egress is now the default behavior
    • Added a mass_absolute function for non-normalized distance calculation with FFT convolution
    • Added diagonal pre-processing function to core.py
    • Added centered-sum-of-products and Pearson correlation in place of sliding dot products
    • Added left and right matrix profile indices to scrump and converted to Pearson correlation
  • tasks
    • Removed Pandas series in GPU tests to improve unit test speed in CI
    • Updated to latest version of black for better formatting
    • Refactored redundant test section
    • Added unit test for inappropriate dtype inputs
    • Corrected absolute stumpy import to be relative import
    • Replaced ._illegal attribute with a more obvious ._T_isfinite attribute
    • Moved common diagonal functions to core.py
    • Replaced order variable with the more obvious diag name
    • Added environment.yml for easier installation of dependencies
    • Removed random print statement in code
    • Moved STUMPY thresholds to global parameters in config.py
    • Refactored left/right matrix profile indices
    • Refactored NaN checking
    • Check for Linux OS and add TBB dynamically especially for CI
  • documentation
    • Added aamp reference to README
    • Update docstrings to be less verbose for API documentation
    • Fixed some typos
    • Replaced sep="\s+" with sep="\\s+" in tutorials
    • Added notes and derivations for Pearson correlation and centered-sum-of-products
    • Renamed tutorials with underscores for consistency
    • Added all aamp-like functions to API reference
    • Replaced MS Word docs with LaTeX notebooks

2020-06-15 [ 1.4.0 ]:

  • bugfixes
    • Fixed bad chunking in compute_mean_std
  • features
    • Added parallelized scrump
    • Added NaN/inf support to scrump
    • Added prescrump (and, consequently, scrump++)
    • Added AB-join to scrump
    • Added unit test for scrump
    • Added constrained inclusion motif search for mstump/mstumped
    • Added discord support for mstump/mstumped
    • Changed sorting to pure numpy in mstump/mstumped for better performance
    • Added NaN/inf support to gpu_stump
    • Added new core.preprocess function
    • Added NaN/inf support to core.mass
    • Added core.apply_exclusion_zone for consistent exclusion zone across functions
    • Added stumpi for incrementally updating matrix profiles with streaming data
    • Added stumpi unit tests
    • Added NaN/inf support for stumpi
    • Converted floss.floss generator to class
  • tasks
    • Added Python 3.8 to Azure Pipelines for unit testing
    • Moved stomp to _stomp to prevent public usage
    • Fixed numerous typos
    • Added several np.asarray calls to input arrays
    • Split some unit tests out into separate files
    • Updated distributed teststo use context manager
    • Refactored _calculate_squared_distance
    • Added core.py to JIT-compiled unit test section
    • Remove mypy config file
    • Corrected cuda.jit signature for gpu_stump
    • Shortened time series length for faster GPU tests
    • Removed link to discourse from documentation
    • Added global variables for controlling chunking in compute_mean_std
    • Replaced name of naive_mass with naive_stamp
    • Removed redundant "caption" in RTD ToC
    • Renamed stamp.mass to private stamp._mass_PI
    • Added flake8-docstrings checking
    • Renamed utils.py to naive.py and updated corresponding function calls
    • Added stumpi and scrump to STUMPY API (RTD)
  • documentation
    • Initialized shapelet discovery tutorial (WIP)
    • Updated check_window_size docstring
    • Added gpu_stump to tutorial
    • Added scrump tutorial for Fast Approximate Matrix Profiles
    • Updated string formatting to conform to flake8
    • Added stumpi tutorial
    • Improved mstump tutorial (WIP)
    • Added additional references to original matrix profile papers

2020-03-27 [ 1.3.1 ]:

  • bugfixes
    • Fixed MSTUMP/MSTUMPED input dimensions check
    • Fixed inconsistent MSTUMP/MSTUMPED output
  • features
    • Added support for constant subsequences and added unit tests
    • Improved GPU memory consumption for self-join
    • Added ability to handle NaN and inf values in all matrix profile algorithms (except gpu_stump)
  • tasks
    • Updated performance table with new performance results, better color scheme, intuitive hardware grouping
    • Re-organized ndarray input verification steps
    • Added more unit tests and rearranged test order
    • Removed Python type hints or type annotations
    • Split failing dask unit tests into multiple test files
    • Added PR template
    • Updated Mac OS X image for Azure Pipelines
    • Replaced stddev computation with a memory efficient rolling chunked stddev
    • Modified exclusion zone to be symmetrical
    • Refactored multi-dimensional mass
    • Fixed scenarios where subsequence contains zero mean
    • Added explicit PR trigger to Azure Pipelines
    • Updated installation instructions to use conda-forge channel
    • Fixed time series chains all_c test to handle differences in Python set order
  • documentation
    • Fixed various typos
    • Refactored tutorials for clarity

2019-12-30 [ 1.3.0 ]:

  • bugfixes
    • Fixed MSTUMP/MSTUMPED input dimensions check
    • Fixed inconsistent MSTUMP/MSTUMPED output
  • features
    • Added parallel GPU-STUMP (i.e., multi-GPU support) using Python multiprocessing and file I/O
    • Added Python type hints/type annotations
  • tasks
    • Updated performance table and plots with STUMPY.2, GPU-STUMP.1, GPU-STUMP.2, GPU-STUMP.DGX1, and GPU-STUMP.DGX2 results
    • Fixed test function names
    • Added Python script for easier performance timing conversion
  • documentation
    • Added window size (m = 50) for performance calculations
    • Fixed various typos
    • Added missing and improved docstrings
    • Replaced Bokeh with Matplotlib
    • Updated GPU-STUMP example with multi-GPU support

2019-12-03 [ 1.2.4 ]:

  • features
    • Added ability to select GPU device in gpu_stump
  • tasks
    • Changed all README hyperlinks to double underscores
    • Added API links to README
    • Added STUMPY circle image (logo)
    • Suppressed pytest junit_family deprecation warning
    • Replaced python install with python -m pip install .

2019-11-26 [ 1.2.3 ]:

  • bugfixes
    • Fixed incorrect GPU output for self joins
  • features
    • Added array check for NaNs
    • Improved test script for CI
  • tasks
    • Added discourse group at stumpy.discourse.group
    • Updated formatting and added newline before logo
    • Made logo a hyperlink to Github repo
    • Added custom CSS to sphinx theme

2019-11-03 [ 1.2.2 ]:

  • bugfixes
    • Fixed Python AST utf8 file reading bug on Windows

2019-11-03 [ 1.2.1 ]:

  • bugfixes
    • Fixed driver not found function when no GPU present
    • Fixed gpu_stump docstring in RTD

2019-11-02 [ 1.2.0 ]:

  • bugfixes
    • Fixed transposed Pandas DataFrame issue #66
  • features
    • Added GPU support (NVIDIA Only)
  • tasks
    • Added CHANGELOG
    • Added GPU tests for GPU-STUMP
    • Added CPU tests (via CUDA simulator) for GPU-STUMP
    • Added additional STUMPY logos
    • Disabled bidirectional and left indices for FLOSS
    • Added STUMPY to Python 3 Statement
  • documentation
    • Added reference to semantic segmentation
    • Added gpu_stump API to RTF
    • Added gpu_stump example to README
    • Improved docstring for mstump(ed)
    • Improved time series chains tutorial
    • Added FLUSS to README
    • Added missing docstring for FLOSS
    • Added FLOSS and FLUSS to RTD API
    • Updated docstring DOIs with URLs to primary references
    • Added motif discovery to exisiting tutorial

2019-08-03 [ 1.1.0 ]:

  • bugfixes
    • Removed incorrect compatibility with Py35
    • Split Numba JIT tests tests to ensure proper Dask PyTest teardown
  • features
    • Added FLUSS and FLOSS
    • Added Pandas Series/DataFrame support
  • tasks
    • Added conda-forge support
    • Set up CI with Azure Pipelines
    • Added black and flake8 checks
    • 100% test coverage
    • Added badges
    • Added version attr
    • Added performance graph
    • Published in JOSS
    • Added STUMPY logo suite
    • Added CHANGELOG
  • documentation
    • Added RTD documentation
    • Added Tutorials
    • Added Binder support
    • Added NABDConf presentation

2019-05-12 [ 1.0.0 ]:

  • Initial release