0.5.6

• Since manylinux nupic.bindings wheel 0.4.10 has now been released to PyPi, we no longer need to install nupic.bindings from S3.
• fix logic in _getColumnNeighborhood
• Bugfix in flatIdx reuse after a segment is destroyed
• Change private _burstColumn class method signature to accept a cellsForColumn argument in lieu of a cellsPerColumn argument. Move the calculation that otherwise depends on cellsPerColumn into the instance method.
• TM: Support extensibility by using traditional methods
• Update expected error for topology changes
• Update expected hotgym result for topology changes
• Adds RELEASE.md with documentation for releasing NuPIC.
• Match nupic.core's SP neighborhood ordering.
• Update inhibition comments and docstrings.
• Introduce mechanism by which already-installed pre-release versions of nupic.bindings are ignored during installation
• Assign self.connections value from self.connectionsFactory() rather than direct usage of Connections constructor. Allows better extensibility should the user want to change some aspect of the creation of the connections instance in a subclass
• Removed obsolete directory src/nupic/bindings/
• Remove the notion of "destroyed" Segments / Synapses
• Enable proper subclassing by converting staticmethods that referenced TemporalMemory to classmethods that reference their class.
• Fixup TemporalMemory.write() to handle columnDimensions as tuples.
• Initialize columnDimensions as a tuple in test to reflect common convention. This forces the TemporalMemoryTest.testWriteRead test to fail in its current state.
• Store "numActivePotentialSynapses". No more "SegmentOverlap".
• Add a lot more scenarios to the TM perf benchmark
• Moved audiostream example to htm-community
• Safer "addToWinners" value. Play nicely with surgical boosting.
• Bugfix: With no stimulus threshold, still break ties when overlaps=0
• Clean up trailing whitespace and tabs
• Properly apply the stimulus threshold
• Add test for new "learn on predicted segments" behavior
• Split compute into activateCells and activateDendrites
• Grow synapses in predicted columns, not just bursting columns
• Removed bundled get-pip.py and instead fetch version copy from S3
• Removed .nupic_modules and now rely on versioned release of nupic.bindings on PyPI
• averagingWindow size updated to improve HTM scores for RES-296
• Build system updates for Bamboo (Linux), Travis (OS X), and AppVeyor (Windows)
• Added nyc taxi example for anomaly detection

0.5.5

• Renamed a misclassed class name from ConnectionsTest to GroupByTest
• not _ is => is not and fixes groupby comment and passes integration tests
• overhaul to groupby, now 10% faster than current implementation
• NUP-2299 Install specific versions of pip, setuptools, and wheel.
• NUP-2299 Added platform-conditional dependency on pycapnp==0.5.8 using PEP-508.
• lazy group_by and changes to GroupByGenerator
• perf improvement to segment comparison in compute activity
• 100 % increase in spped
• small perf changes
• demonstrate that compatability test works with predictedSegmentDec not 0.0
• fixes subtle bug in numSegments that caused integration tests to fail
• fixes bug where minIdx could be passed as a float rather than an int
• skip serialization test if capnp is not installed
• lints and updates comments in group_by.py and group_by_tests.py
• gets same results as c++ temporal memory after group_by changes
• ports group_by tests and they pass
• adds groupByN utility function for use in TM
• all connections tests written and passing, moved some stuff around and added missing function to connections
• started porting new connections tests and minor changes to connections.py
• improves permanence >= testing in computeActivity
• confirmed python implementation is same as cpp version. Needs better perf now
• adds back AnomalyRegion and Anomaly class in anomaly.py and related tests
• fixes bug in growSynapses, almost exactly the same
• Updated core SHA and default SDR classifier implementation
• Updated SDRClassifier factory and region to handle cpp
• changed input name from value to metricValue
• updates variables names in anomaly_likelihood.py and AnomalyLikelihoodRegion
• adds new connections methods
• create new methods for creating/destroying synapses/segments
• continues change of connections datastructures
• move raw anomaly calculation back to nupic.algorithms.anomaly
• Finished swarming/hypersearch separation
• Moved base hypersearch classes to hypersearch
• Moved experimentutils to nupic.swarming
• Updated SDR classifier internals
• calculate raw anomly score in KNNAnomalyClassifier
• removes anomaly.py dependency in network_api_demo.py
• changes how TPRegion computes prevPredictdColumns and updates clamodel
• Install pip from local copy, other simplifications
• Fixup PYTHONPATH to properly include previously-defined PYTHONPATH
• adds pseudocode to core functions
• continues implementation of AnomalyLikelihoodRegion
• Limit tests to unit after ovverriding pytest args on cli
• DEVOPS-85 OS X build infrastructure for Bamboo CI environment
• replaces segmentCMP with lambda and updates docstrings
• uses arrays instead of dicts in computeActivity
• Corrections to examples in tm_high_order.py
• incorporates binary search into the algorithm where applicable
• remove outdated nab unit tests
• use Q function
• Corrections to examples in tm_high_order.py
• change to column generator
• Added tm_high_order.py to show examples of the temporal memory.
• Fixed conversion bug in SDRClassifier serialization
• Fixed patternNZ proto writing.
• Slight fix for pattern history handling in sdr classifier
• Small fix on SDR classifier
• Better fix for #3172, using the initialize() function and checking if _sdrClassifier is set
• Updated learning rate for SDR classifier + slight changes to the error ranges in OPF test
• Updated hotgym test with actual value and implemented first fix for OPF test
• Updated tests and examples with SDR classifier
• Finished updating examples with SDR classifier.
• Updated hotgym and general anomaly examples with SDR classifier.
• Updates pycapnp to 0.5.8
• test_db-fixes avoids printing user password in plaintext
• test_db-fixes updates database and table name
• Corrections made to the spatial pooler tutorial.
• changes maxBoost default value to 1.0
• fixes connection tests and prints config file used in test_db.py
• Moved back overlap accesors test for spatial_pooler from API tests to unit tests.
• Added tutorial script for the spatial pooler. Modified README file accordingly.
• Moved the unit test for SP overlap accesors to API tests.

0.5.4

• Added overlap accessors to spatial_pooler.py plus unit tests. (Code style corrected)
• Updated VERSION in Spatial Pooler and added backward compatibility in setstate()
• Added members overlaps and boostedOverlaps to SpatialPooler class.
• Addition of overlaps and boostedOverlaps members to SpatialPooler class plus unit tests.
• Added docs for return type in RDSE internal func.
• tm_cpp with tuned parameters
• RES-215 Changes to add params for new TM subclass for NAB
• Remove main function from SDRClassifierRegion
• remove unused methods from SDRClassifierRegion
• Add simple end-to-end integration test for SDRClassifierRegion
• use string split instead of eval to parse strings
• correct inconsistent error msg in sdr_classifier_factory.py
• Fix readWrite test of SDR classifier
• Add SDRClassifier Region to pyRegions
• Initial implementation of SDRClassifier Region
• implement SDR classifier factory
• Add capnp proto for SDR classifier region
• Add default value for SDR classifier implementation in nupic-default.xml

0.5.3

• Default DATETIME columns to NULL in ClientJobsDAO for compatibility across mysql versions. As of mysql 5.7.8, values of 0 are not allowed for DATETIME columns, and CURRENT_TIMESTAMP is semantically inappropriate for those columns.
• Suppress this optional dependency on matplotlib without logging, because python logging implicitly adds the StreamHandler to root logger when calling logging.debug, etc., which may undermine an application's logging configuration
• Bugfix: Write the 'actualValues' to the output, don't reassign the output
• Fixed Username Regex in ClientJobsDAO
• cleaned up region a bit to make it compliant with numenta's coding guidelines.

0.5.2

• Fixe to GCE to return the right number of scalars when altitude is missing.

0.5.1

• Improves SDR classifier and tests
• Modify the continuous online learning test
• Add 3 tests on multiple item prediction
• Fix test_pFormatArray
• Implement SDR classifier in NuPIC
• Make the 'arrayTypes' list more informative
• Add getParameter/setParameter support for Bool and BoolArray
• Improved anomaly params (from NAB)
• Added minSparsity option
• Get the encoder's outputWidth via parameter
• Use nupic.core encoders from nupic via the Network API
• Fix bugs and inconsistencies in the custom region demo
• Adds BINDINGS_VERSION envvar to wheel filename (for iterative builds)

0.5.0

• Removes references to FastTemporalMemory.
• Lower TM epsilon threshold for compatibility.
• Add documentation for the Monitor Mixins
• Removed FastTemporalMemory from nupic
• Update temporal memory compatibility test to use C++ TM.
• Sort segments before iterating for compatibility with C++
• Sort unpredictedActiveColumns before iterating for compatibility with C++

0.4.5

• This release is just to sync with nupic.bindings 0.3.1.

0.4.3

• Updating to proper core sha

0.4.2

• Using official release version of bindings for nupic release.