About: Universal Python-written numerical optimization toolbox. Problems: NLP, LP, QP, NSP, MILP, LSP, LLSP, MMP, GLP, SLE etc; automatic differentiation is available

http://openopt.org/Changelog

About: Python module to ease pattern classification analyses of large datasets. It provides high-level abstraction of typical processing steps (e.g. data preparation, classification, feature selection, [...]

0.4.4 (Mon, Feb 2 2010) (Total: 144 commits)

Primarily a bugfix release, probably the last in 0.4 series since development for 0.5 release is leaping forward.

• New functionality (19 NF commits):

o GNB implements Gaussian Naïve Bayes Classifier.

o read_fsl_design() to read FSL FEAT design.fsf files (Contributed by Russell A. Poldrack).

o SequenceStats to provide basic statistics on labels sequence (counter-balancing, autocorrelation).

o New exceptions DegenerateInputError and FailedToTrainError to be thrown by classifiers primarily during training/testing.

o Debug target STATMC to report on progress of Monte-Carlo sampling (during permutation testing).

• Refactored (15 RF commits):

o To get users prepared to 0.5 release, internally and in some examples/documentation, access to states and parameters is done via corresponding collections, not from the top level object (e.g. clf.states.predictions instead of soon-to-be-deprecated clf.predictions). That should lead also to improved performance.

o Adopted copy.py from python2.6 (support Ellipsis as well). ed (38 BF commits):

o GLM output does not depend on the enabled states any more.

o Variety of docstrings fixed and/or improved.

o Do not derive NaN scaling for SVM’s C whenever data is degenerate (lead to never finishing SVM training).

o sg : + KRR is optional now – avoids crashing if KRR is not available.

• tolerance to absent set_precompute_matrix in svmlight in recent shogun versions.

• support for recent (present in 0.9.1) API change in exposing debug levels.

o Python 2.4 compatibility issues: kNN and IFS

About: The SHOGUN machine learning toolbox's focus is on large scale learning methods with focus on Support Vector Machines (SVM), providing interfaces to python, octave, matlab, r and the command line.

This release contains several enhancements, cleanups and bugfixes:

Features

• Integrate LaRank.
• Memory Mapped Features (for data sets that don't fit into memory).
• Compressor module with compression and decompression support for lzo, gzip, bzip2 and lzma.
• Compressed String Features with on-the-fly decompression (CDecompressString preproc).
• Parallel computation of get_kernel_matrix().
• One may now prefix all shogun print/outputs with file name and line number (obj.io.enable_file_and_line())
• Chinese Documentation thanks Elpmis Lee.

Bugfixes

• Fix One class MKL testing in static interfaces.
• Configure fixes: Let octave not write history on configure; fail when cplex is forcefully enabled but not found; add cplex 12 support.
• Fix a problem with regression and CombinedKernels employing only Custom kernels.

Cleanup and API Changes

• String Features now (like SimpleFeatures) upon get_feature_vector require an additional do_free argument and need to be freed using free_feature_vector.

About: Elefant is an open source software platform for the Machine Learning community licensed under the Mozilla Public License (MPL) and developed using Python, C, and C++. We aim to make it the platform [...]

This release contains the Stream module as a first step in the direction of providing C++ library support. Stream aims to be a software framework for the implementation of large scale online learning algorithms. Large scale, in this context, should be understood as something that does not fit in the memory of a standard desktop computer.

Added Bundle Methods for Regularized Risk Minimization (BMRM) allowing to choose from a list of loss functions and solvers (linear and quadratic).

Added the following loss classes: BinaryClassificationLoss, HingeLoss, SquaredHingeLoss, ExponentialLoss, LogisticLoss, NoveltyLoss, LeastMeanSquareLoss, LeastAbsoluteDeviationLoss, QuantileRegressionLoss, EpsilonInsensitiveLoss, HuberRobustLoss, PoissonRegressionLoss, MultiClassLoss, WinnerTakesAllMultiClassLoss, ScaledSoftMarginMultiClassLoss, SoftmaxMultiClassLoss, MultivariateRegressionLoss

Graphical User Interface provides now extensive documentation for each component explaining state variables and port descriptions.

Changed saving and loading of experiments to XML (thereby avoiding storage of large input data structures).

Unified automatic input checking via new static typing extending Python properties.

Full support for recursive composition of larger components containing arbitrary statically typed state variables.

About: BCPy2000 provides a platform for rapid, flexible development of experimental Brain-Computer Interface systems based on the BCI2000.org project. From the developer's point of view, the implementation [...]

Minor fixes since the last release, ready for the tutorial at the BCI2000 workshop http://bci2000.org/BCI2000/Workshop.html

About: This software implements the Dirichlet Forest (DF) Prior within the Latent Dirichlet Allocation (LDA) model. When combined with LDA, the Dirichlet Forest Prior allows the user to encode domain knowledge (must-links and cannot-links between words) into the prior on topic-word multinomials.

Initial Announcement on mloss.org.

About: This software implements the DeltaLDA model, which is a modification of the Latent Dirichlet Allocation (LDA) model. DeltaLDA can use multiple topic mixing weight priors to jointly model multiple [...]

-fixed some npy_intp[] memory leaks -fixed phi normalization bug

About: Monte (python) is a small machine learning library written in pure Python. The focus is on gradient based learning, in particular on the construction of complex models from many smaller components.

Initial Announcement on mloss.org.