Git Product home page Git Product logo

distributions's Introduction

Build Status Code Quality Latest Version

Distributions

Distributions provides low-level primitives for collapsed Gibbs sampling in Python and C++ including:

  • special numerical functions,
  • samplers and density functions from a variety of distributions,
  • conjugate component models (e.g., gamma-Poisson, normal-inverse-chi-squared),
  • clustering models (e.g., CRP, Pitman-Yor), and
  • efficient wrappers for mixture models.

Distributions powered a machine-learning-as-a-service for Prior Knowledge Inc., and now powers machine learning infrastructure at Salesforce.com.

Installation

For python-only support (no C++) you can install with pip:

pip install distributions

For help with other builds, see the installation documentation.

Documentation

The official documentation lives at http://distributions.readthedocs.org/.

Branch-specific documentation lives at

Authors (alphabetically)

License

Copyright (c) 2014 Salesforce.com, Inc. All rights reserved.

Licensed under the Revised BSD License. See LICENSE.txt for details.

distributions's People

Contributors

cap avatar dpfau avatar fritzo avatar jglidden-salesforce avatar svc-scm avatar

Stargazers

avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar

Watchers

avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar avatar

Forkers

ericmjonas karunya yarden dlovell stephentu jesserobertson datamicroscopes akkinenirajesh posterior zbxzc35 afcarl ckingdev isabella232

distributions's Issues

sample_from_likelihoods fails when sample_unif01 returns exactly 1

Really this is an issue with std::uniform_real_distribution, but until that's fixed we should make sure this code is robust to its weird behavior. Very rarely (say one in 10^7 samples) std::uniform_real_distribution (0.0, 1.0) returns exactly 1.0, meaning sample_unif01 returns exactly 1.0. If the vector of likelihoods passed to sample_from_likelihoods ends with zero, then sample_from_likelihoods will return the length of the likelihood array, rather than the last nonzero element of the likelihood array. This can be fixed by replacing if (DIST_UNLIKELY(t < 0)) { on line 325 of distributions/random.hpp with if (DIST_UNLIKELY(t <= 0)) {

Occasional failure of test_model_flavors.test_group for dpd datatype 

======================================================================
FAIL: distributions.tests.test_model_flavors.test_group('dpd',)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/ubuntu/.virtualenvs/sf/local/lib/python2.7/site-packages/nose/case.py", line 197, in runTest
    self.test(*self.arg)
  File "/home/ubuntu/sf/distributions/distributions/tests/test_model_flavors.py", line 86, in _test_group
    shared.add_value(value)
  File "/home/ubuntu/sf/distributions/distributions/dbg/models/dpd.py", line 138, in add_value
    assert value != OTHER, 'cannot add OTHER'
AssertionError: cannot add OTHER

test_random.test_sample_prob_from_scores occasionally fails 

======================================================================
FAIL: distributions.tests.test_random.test_sample_prob_from_scores
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/travis/virtualenv/python2.7_with_system_site_packages/local/lib/python2.7/site-packages/nose/case.py", line 197, in runTest
    self.test(*self.arg)
  File "/home/travis/build/forcedotcom/distributions/distributions/tests/test_random.py", line 210, in test_sample_prob_from_scores
    assert_samples_match_scores(sampler)
  File "/home/travis/build/forcedotcom/distributions/distributions/tests/util.py", line 217, in assert_samples_match_scores
    assert_counts_match_probs(counts, probs)
  File "/home/travis/build/forcedotcom/distributions/distributions/tests/util.py", line 203, in assert_counts_match_probs
    assert gof > tol, 'failed with goodness of fit {}'.format(gof)
AssertionError: failed with goodness of fit 0.000945166953359
--------------------  begin captured stdout ---------------------
EXPECT  ACTUAL  VALUE
10000.0 10000   0
goodness of fit = 1
EXPECT  ACTUAL  VALUE
5693.5  5747    0
4306.5  4253    1
goodness of fit = 0.126409842659
EXPECT  ACTUAL  VALUE
5779.1  5802    2
2534.8  2458    0
1686.1  1740    1
goodness of fit = 0.0671215589621
EXPECT  ACTUAL  VALUE
8314.4  8298    2
670.7   706 0
585.9   581 1
429.0   415 3
goodness of fit = 0.44019109939
EXPECT  ACTUAL  VALUE
4839.9  4842    3
1876.9  1901    2
1808.5  1814    1
1412.0  1379    0
62.6    64  4
goodness of fit = 0.856219051563
EXPECT  ACTUAL  VALUE
3088.0  3109    4
2174.6  2212    3
1713.8  1651    2
1430.5  1450    5
808.0   794 1
785.0   784 0
goodness of fit = 0.495742040274
EXPECT  ACTUAL  VALUE
2266.4  2226    0
2015.0  2095    5
1752.0  1722    2
1525.7  1621    3
904.2   846 4
872.3   807 6
664.4   683 1
goodness of fit = 0.000945166953359
---------------------  end captured stdout  ----------------------
----------------------------------------------------------------------
Ran 310 tests in 28.206s
FAILED (SKIP=10, failures=1)
make: *** [test_cy] Error 1
The command "make test$FLAVOR" exited with 2.
Done. Your build exited with 1.

Pip install fails

pip fails on OSX 10.10.5 with the following error

    clang++ -bundle -undefined dynamic_lookup build/temp.macosx-10.10-x86_64-2.7/distributions/lp/special.o -L/usr/local/lib -L/usr/local/opt/openssl/lib -L/usr/local/opt/sqlite/lib -ldistributions_shared -lm -o build/lib.macosx-10.10-x86_64-2.7/distributions/lp/special.so
ld: library not found for -ldistributions_shared
clang: error: linker command failed with exit code 1 (use -v to see invocation)
error: command 'clang++' failed with exit status 1

pip install distributions fails

I'm trying to install distributions via pip install distributions on OS X, and I get the following error (after a lot of spew):

distributions/lp/models/_bb.cpp:6342:42: error: expected the class name after '~' to name a destructor

  p->scores.distributions::VectorFloat::~VectorFloat();

                                         ^

5 warnings and 1 error generated.

error: command 'gcc' failed with exit status 1

My environment:

(x)stephentu@ibanez:~$ python --version
Python 2.7.7 :: Anaconda 2.0.1 (x86_64)
(x)stephentu@ibanez:~$ uname -a
Darwin ibanez 13.3.0 Darwin Kernel Version 13.3.0: Tue Jun  3 21:27:35 PDT 2014; root:xnu-2422.110.17~1/RELEASE_X86_64 x86_64
(x)stephentu@ibanez:~$ python -c 'import Cython; print Cython.__version__'
0.20.2

This error I believe is related to cython generator invalid C++ code for clang (gcc is OK with it). A simple fix I believe is to require cython>=0.20.2b1 when compiling with clang

officially support conda install distributions?

do we want this? data-microscopes is using my unofficial version of it already, and it seems to work OK. if so, I think we'll want to do two things:

(A) make an official distributions channel on binstar
(B) get a travis CI hook to build new recipes when tags are made (or this can be manual, but it has to be done on both an OS X machine and Linux machine separately)

right now the pip install route is (a) not kept up to date and (b) doesn't quite work out of the box (you still need to compile libdistributions by yourself). I think the conda install can potentially be more portable since it, for example, packages libprotobuf so you don't have to worry about protobuf library version discrepancies. conda also gets the rpath stuff right so you don't have to keep exporting (DY)LD_LIBRARY_PATH

Sampler doesn't work in python 

> from distributions.lp.models import nich
> shared = nich.Shared.from_dict(nich.EXAMPLES[0]['shared'])
> group = nich.Group.from_values(shared, nich.sample_group(shared, 10))
> sampler = nich.Sampler(shared, group)
> sampler.eval(shared)
0.0

and

> nosetests distributions/tests/test_models.py:test_joint
FSegmentation fault (core dumped)

core dump is also not caught by running tests via make.

Installation and compilation errors 

pip install distributions

Downloading/unpacking distributions
  Downloading distributions-2.0.0.tar.gz (67kB): 67kB downloaded
  Running setup.py egg_info for package distributions
    building cython extensions
    Traceback (most recent call last):
      File "<string>", line 16, in <module>
      File "/home/wiecki/envs/pymc3/build/distributions/setup.py", line 48, in <module>
        with open('README.md') as f:
    IOError: [Errno 2] No such file or directory: 'README.md'
    Complete output from command python setup.py egg_info:
    building cython extensions

Traceback (most recent call last):

  File "<string>", line 16, in <module>

  File "/home/wiecki/envs/pymc3/build/distributions/setup.py", line 48, in <module>

    with open('README.md') as f:

IOError: [Errno 2] No such file or directory: 'README.md'

I then tried to pip install directly from the github repo but got:

building 'distributions.hp.random' extension

x86_64-linux-gnu-gcc -pthread -fno-strict-aliasing -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fPIC -Iinclude -Idistributions -I/home/wiecki/envs/pymc3/local/lib/python2.7/site-packages/numpy/core/include -I/usr/include/python2.7 -c distributions/hp/random.cpp -o build/temp.linux-x86_64-2.7/distributions/hp/random.o -DDIST_DEBUG_LEVEL=3 -DDIST_THROW_ON_ERROR=1 -std=c++0x -Wall -Werror -Wno-unused-function -Wno-sign-compare -Wno-strict-aliasing -O3 -ffast-math -funsafe-math-optimizations -mfpmath=sse -msse4.1 -DNPY_NO_DEPRECATED_API=NPY_1_7_API_VERSION

cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++ [enabled by default]

distributions/hp/random.cpp: In function ‘int __pyx_pf_5numpy_7ndarray___getbuffer__(PyArrayObject*, Py_buffer*, int)’:

distributions/hp/random.cpp:2044:52: error: ‘NPY_C_CONTIGUOUS’ was not declared in this scope

     __pyx_t_2 = ((!(PyArray_CHKFLAGS(__pyx_v_self, NPY_C_CONTIGUOUS) != 0)) != 0);

                                                    ^

distributions/hp/random.cpp:2082:52: error: ‘NPY_F_CONTIGUOUS’ was not declared in this scope

     __pyx_t_1 = ((!(PyArray_CHKFLAGS(__pyx_v_self, NPY_F_CONTIGUOUS) != 0)) != 0);

                                                    ^

distributions/hp/random.cpp:2245:42: error: ‘PyArrayObject’ has no member named ‘descr’

   __pyx_t_4 = ((PyObject *)__pyx_v_self->descr);

                                          ^

In file included from /usr/include/python2.7/Python.h:80:0,

                 from distributions/hp/random.cpp:16:

distributions/hp/random.cpp: In function ‘void __pyx_f_5numpy_set_array_base(PyArrayObject*, PyObject*)’:

distributions/hp/random.cpp:3797:27: error: ‘PyArrayObject’ has no member named ‘base’

   Py_XDECREF(__pyx_v_arr->base);

                           ^

/usr/include/python2.7/object.h:823:34: note: in definition of macro ‘Py_XDECREF’

 #define Py_XDECREF(op) do { if ((op) == NULL) ; else Py_DECREF(op); } while (0)

                                  ^

distributions/hp/random.cpp:3797:27: error: ‘PyArrayObject’ has no member named ‘base’

   Py_XDECREF(__pyx_v_arr->base);

                           ^

/usr/include/python2.7/object.h:772:24: note: in definition of macro ‘Py_DECREF’

         --((PyObject*)(op))->ob_refcnt != 0)            \

                        ^

distributions/hp/random.cpp:3797:3: note: in expansion of macro ‘Py_XDECREF’

   Py_XDECREF(__pyx_v_arr->base);

   ^

distributions/hp/random.cpp:3797:27: error: ‘PyArrayObject’ has no member named ‘base’

   Py_XDECREF(__pyx_v_arr->base);

                           ^

/usr/include/python2.7/object.h:115:47: note: in definition of macro ‘Py_TYPE’

 #define Py_TYPE(ob)             (((PyObject*)(ob))->ob_type)

                                               ^

/usr/include/python2.7/object.h:775:9: note: in expansion of macro ‘_Py_Dealloc’

         _Py_Dealloc((PyObject *)(op));                  \

         ^

/usr/include/python2.7/object.h:823:54: note: in expansion of macro ‘Py_DECREF’

 #define Py_XDECREF(op) do { if ((op) == NULL) ; else Py_DECREF(op); } while (0)

                                                      ^

distributions/hp/random.cpp:3797:3: note: in expansion of macro ‘Py_XDECREF’

   Py_XDECREF(__pyx_v_arr->base);

   ^

distributions/hp/random.cpp:3797:27: error: ‘PyArrayObject’ has no member named ‘base’

   Py_XDECREF(__pyx_v_arr->base);

                           ^

/usr/include/python2.7/object.h:762:45: note: in definition of macro ‘_Py_Dealloc’

     (*Py_TYPE(op)->tp_dealloc)((PyObject *)(op)))

                                             ^

/usr/include/python2.7/object.h:823:54: note: in expansion of macro ‘Py_DECREF’

 #define Py_XDECREF(op) do { if ((op) == NULL) ; else Py_DECREF(op); } while (0)

                                                      ^

distributions/hp/random.cpp:3797:3: note: in expansion of macro ‘Py_XDECREF’

   Py_XDECREF(__pyx_v_arr->base);

   ^

distributions/hp/random.cpp:3806:16: error: ‘PyArrayObject’ has no member named ‘base’

   __pyx_v_arr->base = __pyx_v_baseptr;

                ^

distributions/hp/random.cpp: In function ‘PyObject* __pyx_f_5numpy_get_array_base(PyArrayObject*)’:

distributions/hp/random.cpp:3841:30: error: ‘PyArrayObject’ has no member named ‘base’

   __pyx_t_1 = ((__pyx_v_arr->base == NULL) != 0);

                              ^

In file included from /usr/include/python2.7/Python.h:80:0,

                 from distributions/hp/random.cpp:16:

distributions/hp/random.cpp:3864:44: error: ‘PyArrayObject’ has no member named ‘base’

     __Pyx_INCREF(((PyObject *)__pyx_v_arr->base));

                                            ^

/usr/include/python2.7/object.h:767:18: note: in definition of macro ‘Py_INCREF’

     ((PyObject*)(op))->ob_refcnt++)

                  ^

distributions/hp/random.cpp:3864:5: note: in expansion of macro ‘__Pyx_INCREF’

     __Pyx_INCREF(((PyObject *)__pyx_v_arr->base));

     ^

distributions/hp/random.cpp:3865:41: error: ‘PyArrayObject’ has no member named ‘base’

     __pyx_r = ((PyObject *)__pyx_v_arr->base);

                                         ^

error: command 'x86_64-linux-gnu-gcc' failed with exit status 1

This is with most recent cython and numpy 1.8.1.

Tests fail on OSX

Running CC=clang PYDISTRIBUTIONS_USE_LIB=1 make test on OSX, I see these errors:

======================================================================
FAIL: distributions.tests.test_clustering.test_models(<type 'distributions.lp.clustering.PitmanYor'>, 5)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/Users/cpetschulat/.virtualenvs/pk/lib/python2.7/site-packages/nose/case.py", line 197, in runTest
    self.test(*self.arg)
  File "/Users/cpetschulat/src/distributions/distributions/tests/test_clustering.py", line 97, in _test_models
    'not normalized: {}'.format(total))
AssertionError: not normalized: 0.989162792478
-------------------- >> begin captured stdout << ---------------------
Example 0
    Prob    Count
   0.200      190 ------------------------------------------------------------
   0.050       63 --------------------
   0.050       53 -----------------
   0.050       52 ----------------
   0.050       42 -------------
   0.050       40 -------------
   0.017       22 -------
   0.017       21 -------
PitmanYor gof = 0.214
Example 1
    Prob    Count
   0.161      151 ------------------------------------------------------------
   0.045       61 ------------------------
   0.045       52 ---------------------
   0.045       47 -------------------
   0.045       45 ------------------
   0.045       49 -------------------
   0.020       20 --------
   0.019       22 ---------
PitmanYor gof = 0.412
Example 2
    Prob    Count
   0.755      763 ------------------------------------------------------------
   0.016       18 -
   0.016       18 -
   0.016       15 -
   0.016       11 -
   0.016       15 -
   0.016       15 -
   0.016       14 -
PitmanYor gof = 0.919
Example 3

--------------------- >> end captured stdout << ----------------------

======================================================================
FAIL: distributions.tests.test_models.test_sample_group('lp.models.dpd',)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/Users/cpetschulat/.virtualenvs/pk/lib/python2.7/site-packages/nose/case.py", line 197, in runTest
    self.test(*self.arg)
  File "/Users/cpetschulat/src/distributions/distributions/tests/test_models.py", line 95, in test_one_model
    test_fun(Model, EXAMPLE)
  File "/Users/cpetschulat/src/distributions/distributions/tests/test_models.py", line 292, in test_sample_group
    assert_greater(gof, MIN_GOODNESS_OF_FIT)
AssertionError: 1.4720371676952502e-06 not greater than 0.001
-------------------- >> begin captured stdout << ---------------------
example 1/1
    Prob    Count
   0.417      368 ------------------------------------------------------------
   0.188      195 --------------------------------
   0.188      187 ------------------------------
   0.042       65 -----------
   0.042       56 ---------
   0.042       48 --------
   0.042       30 -----
   0.021       34 ------
DirichletProcessDiscrete gof = 1.47e-06

--------------------- >> end captured stdout << ----------------------

APPCRASH when trying to install within Anaconda3, Windows 7

Problem Event Name: APPCRASH
Application Name: pythonw.exe
Application Version: 0.0.0.0
Application Timestamp: 5398d7c1
Fault Module Name: python34.dll
Fault Module Version: 3.4.1150.1013
Fault Module Timestamp: 5398d7c0
Exception Code: c0000005
Exception Offset: 00000000000d7881
OS Version: 6.1.7601.2.1.0.256.4
Locale ID: 1033

This happens whenever I try to install distributions-2.0.26 under Anaconda3 (Python 3.4), in Windows 7, on 64-bit. Happens if pip is used or install is attempted from source. I have tried installing from an Administrator Command Prompt as well, and from the QT window. Also have gone back to distributions-2.0.10 and tried that. This happens trying to install dpmix as well.

Anaconda3 was just installed today, and I ran updates of numpy, scipy, and anaconda itself.

Thoughts?

Distributions not working with loom

Am I missing some undocumented environment variable?

python -m loom.datasets init
Traceback (most recent call last):
  File "/usr/lib/python2.7/runpy.py", line 162, in _run_module_as_main
    "__main__", fname, loader, pkg_name)
  File "/usr/lib/python2.7/runpy.py", line 72, in _run_code
    exec code in run_globals
  File "/home/ubuntu/sf/loom/loom/datasets.py", line 32, in <module>
    import loom.format
  File "loom/format.py", line 39, in <module>
    import loom.cFormat
ImportError: libdistributions_shared.so: cannot open shared object file: No such file or directory

merge two groups?

Does it make sense to add a plus_group method to group to combine suff stats? This is a common pattern in inference for HDP models and, I believe, split-merge kernels.

shared = model.Shared()
group1 = model.Group()
group1.init(shared)
group2 = model.Group()
group2.init(shared)
group3 = group1.plus_group(group2)

sparse dirichlet discrete

I think Fritz and I had a conversation somewhat related to this topic, but I sort of forgot the outcome. If I have a DD where the dim is very large, and I expect the non-zero entries of the suffstats (e.g. the counts) to be very sparse, what's the right way to do this in distributions?

Essentially I want a DD where the counts[] is a Sparse<> instead of float[]. Would it be worth created a separate model which is SparseDD?

test_sample_matches_score_counts fails for lp.LowEntropy

I am getting persistent failure of one of the tests:

======================================================================
FAIL: distributions.tests.test_clustering.test_sample_matches_score_counts('lp.LowEntropy',)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/Users/stephentu/anaconda/envs/distributions_clean/lib/python2.7/site-packages/nose/case.py", line 197, in runTest
    self.test(*self.arg)
  File "/Users/stephentu/distributions_clean/distributions/tests/test_clustering.py", line 92, in test_one_model
    test_fun(Model, EXAMPLE, sample_count)
  File "/Users/stephentu/distributions_clean/distributions/tests/test_clustering.py", line 164, in test_sample_matches_score_counts
    assert_greater(gof, MIN_GOODNESS_OF_FIT)
AssertionError: 0.00042565398058435689 not greater than 0.001
-------------------- >> begin captured stdout << ---------------------
example 1/4
sample_size = 2
    Prob    Count
   0.738     1428 ------------------------------------------------------------
   0.262      572 ------------------------
LowEntropy gof = 0.000426

--------------------- >> end captured stdout << ----------------------

Here's how I'm building/running the tests:

(distributions_clean)stephentu@ibanez:~/distributions_clean(master)$ git log -n1
commit 8867b7823b3ab1dc0fa041f1fe25089d8393ca89
Merge: e1c80d0 3c82590
Author: jglidden-salesforce <[email protected]>
Date:   Fri Aug 1 16:26:40 2014 -0700

    Merge pull request #72 from forcedotcom/rng-args

    Rng args
(distributions_clean)stephentu@ibanez:~/distributions_clean(master)$ git status
On branch master
Your branch is up-to-date with 'origin/master'.

nothing to commit, working directory clean
(distributions_clean)stephentu@ibanez:~/distributions_clean(master)$ mkdir build
(distributions_clean)stephentu@ibanez:~/distributions_clean(master)$ cd build
(distributions_clean)stephentu@ibanez:~/distributions_clean/build(master)$  cmake -DCMAKE_INSTALL_PREFIX=/Users/stephentu/anaconda/envs/distributions_clean -DEXTRA_INCLUDE_PATH=/Users/stephentu/anaconda/envs/distributions_clean/include -DEXTRA_LIBRARY_PATH=/Users/stephentu/anaconda/envs/distributions_clean/lib ..
-- The C compiler identification is AppleClang 5.1.0.5030040
-- The CXX compiler identification is AppleClang 5.1.0.5030040
-- Check for working C compiler: /usr/bin/cc
-- Check for working C compiler: /usr/bin/cc -- works
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working CXX compiler: /usr/bin/c++
-- Check for working CXX compiler: /usr/bin/c++ -- works
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
CXX_FLAGS = -stdlib=libc++ -mmacosx-version-min=10.7 -Wno-deprecated-register -fPIC -g -std=c++0x -Wall -Wextra -Werror -Wno-unused-parameter -Wno-strict-aliasing -O3 -mfpmath=sse -msse4.1 -ffast-math -funsafe-math-optimizations
[... snip ...]
(distributions_clean)stephentu@ibanez:~/distributions_clean/build(master)$  make && make install && cd ..
[... snip ...]
(distributions_clean)stephentu@ibanez:~/distributions_clean(master)$ PYDISTRIBUTIONS_USE_LIB=1 LIBRARY_PATH=/Users/stephentu/anaconda/envs/distributions_clean/lib EXTRA_INCLUDE_PATH=/Users/stephentu/anaconda/envs/distributions_clean/include pip install -e .
(distributions_clean)stephentu@ibanez:~/distributions_clean(master)$ export DYLD_FALLBACK_LIBRARY_PATH=~/anaconda/envs/distributions_clean/lib
(distributions_clean)stephentu@ibanez:~/distributions_clean(master)$ nosetests -v distributions

DOC: Add hello_world example and better description

I think for a newcomer it would be really useful to have a much simpler example. E.g. the classic: samples from a mixture of gaussians that's then estimated parametrically and non-parametrically.

I also found the compression example a little hard to follow. Could certainly use more comments at each step and what the model is actually inferring.

libdistributions_shared.so undefined reference

make test now fails for me with:

...
Linking CXX executable foo
...
/home/ubuntu/sf/distributions/build/src/libdistributions_shared.so: undefined reference to `google::protobuf::internal::WireFormat::SerializeUnknownFields(google::protobuf::UnknownFieldSet const&, google::protobuf::io::CodedOutputStream*)'

Am I missing an undocumented environment variable?

$ env | grep DIST
DISTRIBUTIONS_PATH=/home/ubuntu/sf/distributions
DISTRIBUTIONS_USE_PROTOBUF=1
$ env | grep PATH
LD_LIBRARY_PATH=:/home/ubuntu/johann/lib:/home/ubuntu/pomagma/lib:/home/ubuntu/johann/lib:/home/ubuntu/pomagma/lib:/home/ubuntu/johann/lib:/home/ubuntu/pomagma/lib:/home/ubuntu/johann/lib:/home/ubuntu/pomagma/lib:/home/ubuntu/johann/lib:/home/ubuntu/pomagma/lib:/home/ubuntu/johann/lib:/home/ubuntu/pomagma/lib:/home/ubuntu/johann/lib:/home/ubuntu/pomagma/lib:/home/ubuntu/johann/lib:/home/ubuntu/pomagma/lib:/home/ubuntu/johann/lib:/home/ubuntu/pomagma/lib:/home/ubuntu/johann/lib:/home/ubuntu/pomagma/lib:/home/ubuntu/johann/lib:/home/ubuntu/pomagma/lib:/home/ubuntu/johann/lib:/home/ubuntu/pomagma/lib
PATH=/home/ubuntu/.virtualenvs/sf/bin:/usr/local/heroku/bin:/home/ubuntu/.virtualenvs/sf/bin:/usr/local/heroku/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/home/ubuntu/bin:/home/ubuntu/tool_config/bin:/home/ubuntu/johann/bin:/home/ubuntu/kazoo/bin:/home/ubuntu/jtext/bin:/home/ubuntu/bin:/home/ubuntu/tool_config/bin:/home/ubuntu/johann/bin:/home/ubuntu/kazoo/bin:/home/ubuntu/jtext/bin
DISTRIBUTIONS_PATH=/home/ubuntu/sf/distributions

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.