Merge branch 'release/0.8.0'

Author: wmayner

.gitignore

@@ -9,6 +9,7 @@ node_modules
 **/*.egg-info
 **/*.log
 **/*.BACKUP
+pyphi/data/hamming_matrices/10.npy
 
 __pyphi_cache__
 dist
@@ -17,5 +18,6 @@ docs/_build
 github-docs
 .coverage
 htmlcov
-profiling
 results
+html
+env

.travis.yml

@@ -1,7 +1,11 @@
 language: python
 
-# Provide a mongod process on port 27017 (default)
-services: mongodb
+sudo: false
+
+# Setup databases
+services:
+  - mongodb  # port 27017 (default)
+  - redis-server  # port 6379 (default)
 
 python:
   - "3.4"
@@ -16,8 +20,8 @@ before_install:
   - conda update --yes conda
   # The next couple lines fix a crash with multiprocessing on Travis and are
   # not specific to using Miniconda
-  - sudo rm -rf /dev/shm
-  - sudo ln -s /run/shm /dev/shm
+  # - sudo rm -rf /dev/shm
+  # - sudo ln -s /run/shm /dev/shm
 
 # Install packages
 install:
@@ -27,9 +31,14 @@ install:
   - pip freeze
 
 # Run tests
-script: coverage run --source=pyphi -m py.test
+script: coverage run --source=pyphi -m py.test --slow
 
 after_success: coveralls
 
 notifications:
-  - email: false
+  email: false
+  slack:
+    rooms:
+         secure: "C6PSvK/FEmyD+S71VVzS29NZpD9CFI/f3UMjc6i08xYHjFNMZUplfHyUqNLNqgVVBsKR1En4B8ryuriEnM8wwZzCVdxjh7HDsx8aNE2z1XG5tkqoRP4ppNfqNaYRoMcSZFW89DfVNbIVDM11wKXL7/YnRCmlJy0p1NtKzlncXCM="
+    on_success: never
+    on_failure: always

CHANGELOG.md

@@ -1,6 +1,148 @@
 Changelog
 =========
 
+0.8.0
+------------------
+_2016-02-06_
+
+### API changes
+- Mechanisms and purviews are now passed to all functions and methods in node
+  index form (e.g. `(0, 1, 3)`). Previously, many functions took these
+  arguments as `Node` objects. Since nodes belong to a specific `Subsystem` it
+  was possible to pass nodes from one subsystem to another subsystem's methods,
+  leading to incorrect results.
+- `constellation_distance` no longer takes a `subsystem` argument because
+  concepts in a constellation already reference their subsystems.
+- Moved `utils.cut_mechanism_indices` and `utils.mechanism_split_by_cut` to
+  to `Cut.all_cut_mechanisms` and `Cut.splits_mechanism`, respectively;
+  moved `utils.cut_mice` to `Mice.damaged_by_cut`.
+- `Concept.__eq__`: when comparing concepts for equality, we no longer directly
+  check equality of their subsystems. Concept equality is now defined as
+  follows:
+    - Same φ
+    - Same mechanism node indices cause/effect purview node indices
+    - Same mechanism state
+    - Same cause/effect repertoires
+    - Same networks
+  This allows two concepts to be equal when _e.g._ the only difference between
+  them is that one's subsystem is a superset of the other's subsystem.
+- `Concept.__hash__`: the above notion of concept equality is also implemented
+  for concept hashing, so two concepts that differ only in that way will have
+  the same hash value.
+- Disabled concept caching; removed the `config.CACHE_CONCEPTS` option.
+
+### API Additions
+- Added `config.READABLE_REPRS` to control whether `__reprs__` of PyPhi models
+  default to using pretty string formatting.
+- Added a `Constellation` object.
+- Added `utils.submatrix` and `utils.relevant_connections` functions.
+- Added the `macro.effective_info` function.
+- Added the `utils.state_of` function.
+- Added the `Subsystem.proper_state` attribute. This is the state of the
+  subsystem's nodes, rather than the entire network state.
+- Added an optional Redis-backed cache for Mice objects. This is enabled with
+  `config.REDIS_CACHE` and configured with `config.REDIS_CONFIG`.
+- Enabled parallel concept evaluation with `config.PARALLEL_CONCEPT_EVALUATION`.
+
+### Fixes
+- `Concept.eq_repertoires` no longer fails when the concept has no cause or
+  effect.
+- Fixed the `Subsystem.proper_state` attribute.
+
+### Refactor
+- Subsystem Mice and cause/effect repertoire caches; Network purview caches.
+  Cache logic is now handled by decorators and custom cache objects.
+- Block reducibility tests and Mice connection computations.
+- Rich object comparisons on phi-objects.
+
+### Documentation
+- Updated documentation and examples to reflect node-to-index conversion.
+
+
+0.7.5
+------------------
+_2015-11-02_
+
+### API changes
+- Subsystem states are now validated rather than network states. Previously,
+  network states were validated, but in some cases there can be a
+  globally-impossible network state that is locally possible for a subsystem
+  (or vice versa) when considering the subsystem's TPM, which is conditioned
+  on the external nodes (i.e., background conditions). It is now impossible to
+  create a subsystem in an impossible state (a `StateUnreachableError` is
+  thrown), and accordingly no 𝚽 values are calculated for such subsystems; this
+  may change results from older versions, since in some cases the calculated
+  main complex was in fact in an impossible. This functionality is enabled by
+  default but can be disabled via the `VALIDATE_SUBSYSTEM_STATES` option.
+
+
+0.7.4
+------------------
+_2015-10-12_
+
+### Fixes
+- Fixed a caching bug where the subsystem's state was not included in its hash
+  value, leading to collisions.
+
+
+0.7.3
+------------------
+_2015-09-08_
+
+### API changes
+- Heavily refactored the `pyphi.json` module and renamed it to `pyphi.jsonify`.
+
+
+0.7.2
+------------------
+_2015-07-01_
+
+### API additions
+- Added `convert.nodes2state` function.
+- Added `constrained_nodes` keyword argument to `validate.state_reachable`.
+
+### API changes
+- Concept equality is now more permissive. For two concepts to be considered
+  equal, they must only have the same φ, the same mechanism and purviews (in
+  the same state), and the same repertoires.
+
+
+0.7.1
+------------------
+_2015-06-30_
+
+### API additions
+- Added `purviews`, `past_purviews`, `future_purviews` keyword arguments to
+  various concept-calculating methods. With these, the purviews that are
+  considered in the concept calculation can be restricted.
+
+### API changes
+- States are now associated with subsystems rather than networks. Functions in
+  the `compute` module that operate on networks now also take a state.
+
+### Fixes
+- Fixed a bug in `compute._constellation_distance_emd` where partitioned
+  concepts were unable to be moved to the null concept for the EMD calculation.
+  In some cases, the partitioned system has *greater* ∑φ than the unpartitioned
+  system; therefore it must be possible for the φ of partitioned-constellation
+  concepts to be moved to the null concept, not just vice versa.
+- Fixed a bug in `compute._constellation_distance_emd` where it was possible to
+  move concepts around within their own constellation; the distance matrix now
+  disallows any such intraconstellation paths. This is important because in
+  some cases paths from a concept in one constellation to a concept the other
+  can actually be shorter if a detour is taken through a different concept in
+  the same constellation.
+- Fixed a bug in `validate.state_reachable` where network states were
+  incorrectly validated.
+- `macro.emergence` now always returns a macro-network, even when 𝚽 = 0.
+- Fixed a bug in `repr(Network)` where the perturbation vector and connectivity
+  matrix were switched.
+
+### Documentation
+- Added example describing “magic cuts” that, counterintuitively, can create
+  more concepts.
+- Updated existing documentation to the new subsystem-state association.
+
 
 0.7.0
 ------------------
@@ -9,7 +151,7 @@ _2015-05-08_
 ### API additions
 - `pyphi.macro` provides several functions to analyze networks over different
   spatial scales.
-- `.convert.conditionally_independent(tpm)` checks if a TPM is conditionally
+- `convert.conditionally_independent(tpm)` checks if a TPM is conditionally
   independent.
 
 ### API changes

MANIFEST.in

@@ -1 +1 @@
-include README.rst LICENSE.md requirements.txt pyphi_config.yml
+include README.rst LICENSE.md requirements.txt pyphi_config.yml redis.conf

README.rst

@@ -1,18 +1,18 @@
 .. image:: https://zenodo.org/badge/4651/wmayner/pyphi.png
     :target: http://dx.doi.org/10.5281/zenodo.15787
-    :alt: Zenodo DOI
+    :alt: Zenodo DOI badge
 
 .. image:: https://travis-ci.org/wmayner/pyphi.svg?branch=master
     :target: https://travis-ci.org/wmayner/pyphi
-    :alt: Travis build
+    :alt: Travis build badge
 
-.. image:: https://coveralls.io/repos/wmayner/pyphi/badge.png?branch=master
+.. image:: https://coveralls.io/repos/wmayner/pyphi/badge.svg?branch=master&service=github
     :target: https://coveralls.io/r/wmayner/pyphi?branch=master
-    :alt: Coveralls.io
+    :alt: Coveralls.io badge
 
-*************************
-PyPhi: |phi| for Python 3
-*************************
+***********************
+PyPhi: |phi| for Python
+***********************
 
 PyPhi is a Python 3 library for computing integrated information (|phi|), and
 the associated quantities and objects.
@@ -62,12 +62,12 @@ Detailed installation guide for Mac OS X
 `See here <https://github.com/wmayner/pyphi/blob/develop/INSTALLATION.md>`_.
 
 
-Optional: caching with a database
+Optional: caching with MongoDb
 `````````````````````````````````
 
 PyPhi stores the results of |Phi| calculations as they're computed in order to
 avoid expensive re-computation. These results can be stored locally on the
-filesystem (the default setting), or in a full-fledged database. 
+filesystem (the default setting), or in a full-fledged database.
 
 Using the default caching system is easier and works out of the box, but using
 a database is more robust.
@@ -87,6 +87,27 @@ You can also check out MongoDB's `Getting Started guide
 `manual <http://docs.mongodb.org/manual/>`_.
 
 
+Optional: caching with Redis
+`````````````````````````````
+
+PyPhi can also use Redis as a fast in-memory global LRU cache to store Mice
+objects, reducing the memory load on PyPhi processes.
+
+`Install Redis <http://redis.io/download>`_. The `redis.conf` file provided with
+PyPhi includes the minimum settings needed to run Redis as an LRU cache:
+
+.. code:: bash
+
+    redis-server /path/to/pyphi/redis.conf
+
+Once the server is running you can enable Redis caching by setting
+``REDIS_CACHE: true`` in your ``pyphi_config.yml``.
+
+**Note:** PyPhi currently flushes the connected Redis database at the start of 
+every execution. If you are running Redis for another application be sure PyPhi
+connects to its own Redis server.
+
+
 Contributing
 ~~~~~~~~~~~~
 

benchmarks/asv.conf.json

@@ -0,0 +1,72 @@
+{
+    // The version of the config file format.  Do not change, unless
+    // you know what you are doing.
+    "version": 1,
+
+    // The name of the project being benchmarked
+    "project": "PyPhi",
+
+    // The project's homepage
+    "project_url": "https://github.com/wmayner/pyphi",
+
+    // The URL or local path of the source code repository for the
+    // project being benchmarked
+    "repo": "..",
+
+    // List of branches to benchmark. If not provided, defaults to "master"
+    // (for git) or "tip" (for mercurial).
+    "branches": ["develop"], // for git
+
+    // The DVCS being used.  If not set, it will be automatically
+    // determined from "repo" by looking at the protocol in the URL
+    // (if remote), or by looking for special directories, such as
+    // ".git" (if local).
+    "dvcs": "git",
+
+    // The tool to use to create environments.  May be "conda",
+    // "virtualenv" or other value depending on the plugins in use.
+    // If missing or the empty string, the tool will be automatically
+    // determined by looking for tools on the PATH environment
+    // variable.
+    "environment_type": "virtualenv",
+
+    // the base URL to show a commit for the project.
+    "show_commit_url": "http://github.com/wmayner/pyphi/",
+
+    // The Pythons you'd like to test against.  If not provided, defaults
+    // to the current version of Python used to run `asv`.
+    // "pythons": ["2.7", "3.3"],
+
+    // The matrix of dependencies to test.  Each key is the name of a
+    // package (in PyPI) and the values are version numbers.  An empty
+    // list indicates to just test against the default (latest)
+    // version.
+    // "matrix": {
+    //     "numpy": ["1.6", "1.7"]
+    // },
+
+    // The directory (relative to the current directory) that benchmarks are
+    // stored in.  If not provided, defaults to "benchmarks"
+    "benchmark_dir": "benchmarks",
+
+    // The directory (relative to the current directory) to cache the Python
+    // environments in.  If not provided, defaults to "env"
+    "env_dir": "env",
+
+
+    // The directory (relative to the current directory) that raw benchmark
+    // results are stored in.  If not provided, defaults to "results".
+    "results_dir": "results",
+
+    // The directory (relative to the current directory) that the html tree
+    // should be written to.  If not provided, defaults to "html".
+    //"html_dir": "html",
+
+    // The number of characters to retain in the commit hashes.
+    // "hash_length": 8,
+
+    // `asv` will cache wheels of the recent builds in each
+    // environment, making them faster to install next time.  This is
+    // number of builds to keep, per environment.
+    // "wheel_cache_size": 0
+}