POT : Python Optimal Transport

Python Optimal Transport

Last update: Jan 4, 2023

Related tags

Computer Vision python machine-learning pot wasserstein-barycenters numerical-optimization wasserstein emd optimal-transport ot-mapping-estimation wasserstein-barycenter ot-solver domain-adaptation wasserstein-discriminant-analysis gromov-wasserstein sinkhorn-divergences sinkhorn-knopp

Overview

POT: Python Optimal Transport

This open source Python library provide several solvers for optimization problems related to Optimal Transport for signal, image processing and machine learning.

Website and documentation: https://PythonOT.github.io/

Source Code (MIT): https://github.com/PythonOT/POT

POT provides the following generic OT solvers (links to examples):

OT Network Simplex solver for the linear program/ Earth Movers Distance [1] .
Conditional gradient [6] and Generalized conditional gradient for regularized OT [7].
Entropic regularization OT solver with Sinkhorn Knopp Algorithm [2] , stabilized version [9] [10], greedy Sinkhorn [22] and Screening Sinkhorn [26] with optional GPU implementation (requires cupy).
Bregman projections for Wasserstein barycenter [3], convolutional barycenter [21] and unmixing [4].
Sinkhorn divergence [23] and entropic regularization OT from empirical data.
Smooth optimal transport solvers (dual and semi-dual) for KL and squared L2 regularizations [17].
Non regularized Wasserstein barycenters [16] ) with LP solver (only small scale).
Gromov-Wasserstein distances and GW barycenters (exact [13] and regularized [12])
Fused-Gromov-Wasserstein distances solver and FGW barycenters [24]
Stochastic solver for Large-scale Optimal Transport (semi-dual problem [18] and dual problem [19])
Non regularized free support Wasserstein barycenters [20].
Unbalanced OT with KL relaxation and barycenter [10, 25].
Partial Wasserstein and Gromov-Wasserstein (exact [29] and entropic [3] formulations).
Sliced Wasserstein [31, 32].

POT provides the following Machine Learning related solvers:

Optimal transport for domain adaptation with group lasso regularization, Laplacian regularization [5] [30] and semi supervised setting.
Linear OT mapping [14] and Joint OT mapping estimation [8].
Wasserstein Discriminant Analysis [11] (requires autograd + pymanopt).
JCPOT algorithm for multi-source domain adaptation with target shift [27].

Some other examples are available in the documentation.

Using and citing the toolbox

If you use this toolbox in your research and find it useful, please cite POT using the following reference from our JMLR paper:

Rémi Flamary, Nicolas Courty, Alexandre Gramfort, Mokhtar Z. Alaya, Aurélie Boisbunon, Stanislas Chambon, Laetitia Chapel, Adrien Corenflos, Kilian Fatras, Nemo Fournier, Léo Gautheron, Nathalie T.H. Gayraud, Hicham Janati, Alain Rakotomamonjy, Ievgen Redko, Antoine Rolet, Antony Schutz, Vivien Seguy, Danica J. Sutherland, Romain Tavenard, Alexander Tong, Titouan Vayer,
POT Python Optimal Transport library,
Journal of Machine Learning Research, 22(78):1−8, 2021.
Website: https://pythonot.github.io/

In Bibtex format:

@article{flamary2021pot,
  author  = {R{\'e}mi Flamary and Nicolas Courty and Alexandre Gramfort and Mokhtar Z. Alaya and Aur{\'e}lie Boisbunon and Stanislas Chambon and Laetitia Chapel and Adrien Corenflos and Kilian Fatras and Nemo Fournier and L{\'e}o Gautheron and Nathalie T.H. Gayraud and Hicham Janati and Alain Rakotomamonjy and Ievgen Redko and Antoine Rolet and Antony Schutz and Vivien Seguy and Danica J. Sutherland and Romain Tavenard and Alexander Tong and Titouan Vayer},
  title   = {POT: Python Optimal Transport},
  journal = {Journal of Machine Learning Research},
  year    = {2021},
  volume  = {22},
  number  = {78},
  pages   = {1-8},
  url     = {http://jmlr.org/papers/v22/20-451.html}
}

Installation

The library has been tested on Linux, MacOSX and Windows. It requires a C++ compiler for building/installing the EMD solver and relies on the following Python modules:

Numpy (>=1.16)
Scipy (>=1.0)
Cython (>=0.23)
Matplotlib (>=1.5)

Pip installation

Note that due to a limitation of pip, cython and numpy need to be installed prior to installing POT. This can be done easily with

pip install numpy cython

You can install the toolbox through PyPI with:

pip install POT

or get the very latest version by running:

pip install -U https://github.com/PythonOT/POT/archive/master.zip # with --user for user install (no root)

Anaconda installation with conda-forge

If you use the Anaconda python distribution, POT is available in conda-forge. To install it and the required dependencies:

conda install -c conda-forge pot

Post installation check

After a correct installation, you should be able to import the module without errors:

import ot

Note that for easier access the module is name ot instead of pot.

Dependencies

Some sub-modules require additional dependences which are discussed below

ot.dr (Wasserstein dimensionality reduction) depends on autograd and pymanopt that can be installed with:

pip install pymanopt autograd

ot.gpu (GPU accelerated OT) depends on cupy that have to be installed following instructions on this page. Obviously you will need CUDA installed and a compatible GPU.

Examples

Short examples

Import the toolbox

import ot

Compute Wasserstein distances

# a,b are 1D histograms (sum to 1 and positive)
# M is the ground cost matrix
Wd = ot.emd2(a, b, M) # exact linear program
Wd_reg = ot.sinkhorn2(a, b, M, reg) # entropic regularized OT
# if b is a matrix compute all distances to a and return a vector

Compute OT matrix

# a,b are 1D histograms (sum to 1 and positive)
# M is the ground cost matrix
T = ot.emd(a, b, M) # exact linear program
T_reg = ot.sinkhorn(a, b, M, reg) # entropic regularized OT

Compute Wasserstein barycenter

# A is a n*d matrix containing d  1D histograms
# M is the ground cost matrix
ba = ot.barycenter(A, M, reg) # reg is regularization parameter

Examples and Notebooks

The examples folder contain several examples and use case for the library. The full documentation with examples and output is available on https://PythonOT.github.io/.

Acknowledgements

This toolbox has been created and is maintained by

The contributors to this library are

Alexandre Gramfort (CI, documentation)
Laetitia Chapel (Partial OT)
Michael Perrot (Mapping estimation)
Léo Gautheron (GPU implementation)
Nathalie Gayraud (DA classes)
Stanislas Chambon (DA classes)
Antoine Rolet (EMD solver debug)
Erwan Vautier (Gromov-Wasserstein)
Kilian Fatras (Stochastic solvers)
Alain Rakotomamonjy
Vayer Titouan (Gromov-Wasserstein -, Fused-Gromov-Wasserstein)
Hicham Janati (Unbalanced OT)
Romain Tavenard (1d Wasserstein)
Mokhtar Z. Alaya (Screenkhorn)
Ievgen Redko (Laplacian DA, JCPOT)
Adrien Corenflos (Sliced Wasserstein Distance)

This toolbox benefit a lot from open source research and we would like to thank the following persons for providing some code (in various languages):

Gabriel Peyré (Wasserstein Barycenters in Matlab)
Mathieu Blondel (original implementation smooth OT)
Nicolas Bonneel ( C++ code for EMD)
Marco Cuturi (Sinkhorn Knopp in Matlab/Cuda)

Contributions and code of conduct

Every contribution is welcome and should respect the contribution guidelines. Each member of the project is expected to follow the code of conduct.

Support

You can ask questions and join the development discussion:

On the POT slack channel
On the POT gitter channel
On the POT mailing list

You can also post bug reports and feature requests in Github issues. Make sure to read our guidelines first.

References

[1] Bonneel, N., Van De Panne, M., Paris, S., & Heidrich, W. (2011, December). Displacement interpolation using Lagrangian mass transport. In ACM Transactions on Graphics (TOG) (Vol. 30, No. 6, p. 158). ACM.

[2] Cuturi, M. (2013). Sinkhorn distances: Lightspeed computation of optimal transport. In Advances in Neural Information Processing Systems (pp. 2292-2300).

[3] Benamou, J. D., Carlier, G., Cuturi, M., Nenna, L., & Peyré, G. (2015). Iterative Bregman projections for regularized transportation problems. SIAM Journal on Scientific Computing, 37(2), A1111-A1138.

[4] S. Nakhostin, N. Courty, R. Flamary, D. Tuia, T. Corpetti, Supervised planetary unmixing with optimal transport, Whorkshop on Hyperspectral Image and Signal Processing : Evolution in Remote Sensing (WHISPERS), 2016.

[5] N. Courty; R. Flamary; D. Tuia; A. Rakotomamonjy, Optimal Transport for Domain Adaptation, in IEEE Transactions on Pattern Analysis and Machine Intelligence , vol.PP, no.99, pp.1-1

[6] Ferradans, S., Papadakis, N., Peyré, G., & Aujol, J. F. (2014). Regularized discrete optimal transport. SIAM Journal on Imaging Sciences, 7(3), 1853-1882.

[7] Rakotomamonjy, A., Flamary, R., & Courty, N. (2015). Generalized conditional gradient: analysis of convergence and applications. arXiv preprint arXiv:1510.06567.

[8] M. Perrot, N. Courty, R. Flamary, A. Habrard (2016), Mapping estimation for discrete optimal transport, Neural Information Processing Systems (NIPS).

[9] Schmitzer, B. (2016). Stabilized Sparse Scaling Algorithms for Entropy Regularized Transport Problems. arXiv preprint arXiv:1610.06519.

[10] Chizat, L., Peyré, G., Schmitzer, B., & Vialard, F. X. (2016). Scaling algorithms for unbalanced transport problems. arXiv preprint arXiv:1607.05816.

[11] Flamary, R., Cuturi, M., Courty, N., & Rakotomamonjy, A. (2016). Wasserstein Discriminant Analysis. arXiv preprint arXiv:1608.08063.

[12] Gabriel Peyré, Marco Cuturi, and Justin Solomon (2016), Gromov-Wasserstein averaging of kernel and distance matrices International Conference on Machine Learning (ICML).

[13] Mémoli, Facundo (2011). Gromov–Wasserstein distances and the metric approach to object matching. Foundations of computational mathematics 11.4 : 417-487.

[14] Knott, M. and Smith, C. S. (1984).On the optimal mapping of distributions, Journal of Optimization Theory and Applications Vol 43.

[15] Peyré, G., & Cuturi, M. (2018). Computational Optimal Transport .

[16] Agueh, M., & Carlier, G. (2011). Barycenters in the Wasserstein space. SIAM Journal on Mathematical Analysis, 43(2), 904-924.

[17] Blondel, M., Seguy, V., & Rolet, A. (2018). Smooth and Sparse Optimal Transport. Proceedings of the Twenty-First International Conference on Artificial Intelligence and Statistics (AISTATS).

[18] Genevay, A., Cuturi, M., Peyré, G. & Bach, F. (2016) Stochastic Optimization for Large-scale Optimal Transport. Advances in Neural Information Processing Systems (2016).

[19] Seguy, V., Bhushan Damodaran, B., Flamary, R., Courty, N., Rolet, A.& Blondel, M. Large-scale Optimal Transport and Mapping Estimation. International Conference on Learning Representation (2018)

[20] Cuturi, M. and Doucet, A. (2014) Fast Computation of Wasserstein Barycenters. International Conference in Machine Learning

[21] Solomon, J., De Goes, F., Peyré, G., Cuturi, M., Butscher, A., Nguyen, A. & Guibas, L. (2015). Convolutional wasserstein distances: Efficient optimal transportation on geometric domains. ACM Transactions on Graphics (TOG), 34(4), 66.

[22] J. Altschuler, J.Weed, P. Rigollet, (2017) Near-linear time approximation algorithms for optimal transport via Sinkhorn iteration, Advances in Neural Information Processing Systems (NIPS) 31

[23] Aude, G., Peyré, G., Cuturi, M., Learning Generative Models with Sinkhorn Divergences, Proceedings of the Twenty-First International Conference on Artficial Intelligence and Statistics, (AISTATS) 21, 2018

[24] Vayer, T., Chapel, L., Flamary, R., Tavenard, R. and Courty, N. (2019). Optimal Transport for structured data with application on graphs Proceedings of the 36th International Conference on Machine Learning (ICML).

[25] Frogner C., Zhang C., Mobahi H., Araya-Polo M., Poggio T. (2015). Learning with a Wasserstein Loss Advances in Neural Information Processing Systems (NIPS).

[26] Alaya M. Z., Bérar M., Gasso G., Rakotomamonjy A. (2019). Screening Sinkhorn Algorithm for Regularized Optimal Transport, Advances in Neural Information Processing Systems 33 (NeurIPS).

[27] Redko I., Courty N., Flamary R., Tuia D. (2019). Optimal Transport for Multi-source Domain Adaptation under Target Shift, Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics (AISTATS) 22, 2019.

[28] Caffarelli, L. A., McCann, R. J. (2010). Free boundaries in optimal transport and Monge-Ampere obstacle problems, Annals of mathematics, 673-730.

[29] Chapel, L., Alaya, M., Gasso, G. (2020). Partial Optimal Transport with Applications on Positive-Unlabeled Learning, Advances in Neural Information Processing Systems (NeurIPS), 2020.

[30] Flamary R., Courty N., Tuia D., Rakotomamonjy A. (2014). Optimal transport with Laplacian regularization: Applications to domain adaptation and shape matching, NIPS Workshop on Optimal Transport and Machine Learning OTML, 2014.

[31] Bonneel, Nicolas, et al. Sliced and radon wasserstein barycenters of measures, Journal of Mathematical Imaging and Vision 51.1 (2015): 22-45

Comments

[WIP] small tentative for EMD 1D in torch

I saw the torch branch for LP stuff. Would you be interested in my implementation for the 1d EMD (and the sliced wasserstein with it)?

I'm not a huge fan of Pytorch so I can't vouch that what I'm doing here is the best implementation, but it feels to me like it should be fairly ok for batched inputs which is what you want for slice stuff anyway.

opened by AdrienCorenflos 47
GPU changes:
Replace cudamat by cupy for GPU implementations (cupy is still in active development, while cudamat is not)

Use the new DA class instead of the old deprecated one

TODO for another PR:

Performances are still a bit lower than with cudamat (even if better than CPU for large matrices). Some speedups should be possible by tweaking the code
opened by toto6 38
Domain adaptation Classes
first proposal of DA class structure

BaseEstimator: OTDA wrapper (does not work as a stand-alone but implements the methods common to any OTDA algorithm)

SinkhornTransport: implements Sinkhorn algorithm for OTDA

try doc strings compliant with numpy requirements
opened by Slasnista 29
Domain adaptation Classes
We should change the domain adaptation Classes to be more sklearn compliant.

Main issues:

Use CamelCase for classes

Use init for setting parameters and instead of fit.

@agramfort proposed to Creat new Clases with proper names and begin deprecating the old classes.

I think it is a good move.
enhancement
opened by rflamary 28
Not in simplex -- two sets of largely different sizes
I am trying to calculate the EMD of two sets. When one set has a few hundred entries and the other has only 2, the EMD calculation fails and returns Problem Infeasible.

Steps to reproduce the behavior: ** SEE BELOW COMMENT FOR FIXED SCRIPT **

Expected behavior Should return EMD around 1, instead says that the sets spherEng1 and pencilEnergy are not in the simplex

Screenshots Here is comparing the EMDs calculated for less densely tiled to most densely tiled (number of particles = number of segments) with the two element set

Desktop (please complete the following information):

OS: [MacOSX]

Python version [3.6]

POT installed with pip

import platform; print(platform.platform()) Darwin-16.7.0-x86_64-i386-64bit import sys; print("Python", sys.version) ('Python', '2.7.15 |Anaconda, Inc.| (default, Dec 14 2018, 13:10:39) \n[GCC 4.2.1 Compatible Clang 4.0.1 (tags/RELEASE_401/final)]') import numpy; print("NumPy", numpy.version) ('NumPy', '1.15.4') import scipy; print("SciPy", scipy.version) ('SciPy', '1.1.0') import ot; print("POT", ot.version) ('POT', '0.5.1')
opened by caricesarotti 25
[MRG] Sliced wasserstein
Types of changes

[ ] Docs change / refactoring / dependency upgrade

[ ] Bug fix (non-breaking change which fixes an issue)

[X] New feature (non-breaking change which adds functionality)

[ ] Breaking change (fix or feature that would cause existing functionality to change)

Motivation and context / Related issue

Implement SWD: https://github.com/PythonOT/POT/issues/202

How has this been tested (if it applies)

Added specific tests (positive definiteness + matching the EMD in the 1D case)

Checklist

[X] The documentation is up-to-date with the changes I made.

[X] I have read the CONTRIBUTING document.

[x] All tests passed, and additional code has been covered with new tests.

Not sure why yet but the stuff doesn't build.

I'm publishing this as a draft as I have some other changes in my branch that are pending for another merge (cf this: https://github.com/PythonOT/POT/issues/200)
opened by AdrienCorenflos 19
[WIP] torch implementation of the Sliced Wasserstein Distance
Types of changes

[ ] Docs change / refactoring / dependency upgrade

[ ] Bug fix (non-breaking change which fixes an issue)

[x] New feature (non-breaking change which adds functionality)

[ ] Breaking change (fix or feature that would cause existing functionality to change)

Motivation and context / Related issue

Torch implementation of the SWD (or sliced OT loss? how do you want to call it?)

https://github.com/PythonOT/POT/issues/225

How has this been tested (if it applies)

Added a few unittests, needs to be tested further (WIP)

Checklist

[ ] The documentation is up-to-date with the changes I made.

[x] I have read the CONTRIBUTING document.

[ ] All tests passed, and additional code has been covered with new tests.
opened by AdrienCorenflos 16
[MRG] Improved docs and changed scipy version

I changed the scipy version requirements since version scipy 1.2.1 made my POT crash (cannot remember on which call, sorry, it happened while building the docs) and the issue was fixed when upgrading to scipy 1.3

Apart from that, the main goal of this PR is to homogenize a bit the presentation in the docs.

opened by rtavenar 15
fail when using "pip install POT"

When I use "pip install POT", it failed. It depended on Cython. However, it seems that it forgets to tell pip that it depends on Cython.

I solve this problem by install Cython first. However, if we write both Cython and POT into requirements.txt, the installation will fail.

Could anyone solve that?
documentation

opened by Adoni 15
[MRG] Add Unbalanced KL Wasserstein distance + barycenter
new unbalanced module for UOT with KL relaxation with the funcs:

sinkhorn_unbalanced: generalized Sinkhorn to compute W

barycenter_unbalanced: unbalanced Wasserstein barycenter

Tests of convergence for both algorithms

Examples plot_UOT_1D and plot_UOT_barycenter_1D with unbalanced gaussian distributions.

new feature
opened by hichamjanati 14

An Error (RuntimeWarning: invalid value encountered in log) in ot.da.SinkhornL1lL2 after some iterations

Using ot.da.SinkhornL1l2Transport for a domain adaptation problem, I faced an error as follows:

Datasets used:

Xs.txt Xt.txt ys.txt

To Reproduce

If you download the input files in C:\ , then the code is:

import numpy as np
import to

Xs = np.loadtxt("C: / Xs.txt").reshape(604, 5)
Xt = np.loadtxt("C: / Xt.txt").reshape(601, 5)
ys = np.loadtxt("C: / ys.txt")

ot_base = ot.da.SinkhornL1l2Transport(reg_e=10000, reg_cl=100, max_iter=100, verbose=True)
ot_base.fit(Xs=Xs, ys=ys, Xt=Xt)

Result and Error

 It.  |Loss        |Relative loss|Absolute loss
------------------------------------------------
0|5.193677e+06|0.000000e+00|0.000000e+00
1|3.150847e+05|1.548343e+01|4.878593e+06
2|2.668420e+05|1.807914e-01|4.824274e+04
3|2.663638e+05|1.795333e-03|4.782117e+02
4|2.663590e+05|1.786689e-05|4.759007e+00
5|2.663590e+05|1.312580e-07|3.496174e-02
6|2.663588e+05|7.339658e-07|1.954982e-01
7|2.663106e+05|1.808094e-04|4.815146e+01

C:\Users\enayat.aria\PycharmProjects\pythonProject\venv\lib\site-packages\ot\optim.py:357: RuntimeWarning: 
invalid value encountered in log
return np.sum(M * G) + reg1 * np.sum(G * np.log(G)) + reg2 * f(G)
Traceback (most recent call last):
File "C:\Program Files\JetBrains\PyCharm Community Edition 2021.1.2\plugins\python- 
ce\helpers\pydev\pydevd.py", line 1483, in _exec
pydev_imports.execfile(file, globals, locals)  # execute the script
File "C:\Program Files\JetBrains\PyCharm Community Edition 2021.1.2\plugins\python- 
ce\helpers\pydev\_pydev_imps\_pydev_execfile.py", line 18, in execfile
exec(compile(contents+"\n", file, 'exec'), glob, loc)
File "C:/Users/enayat.aria/PycharmProjects/pythonProject/OT_for_DA.py", line 258, in <module>
ot_base.fit(Xs=Xs, ys=ys, Xt=Xt)
File "C:\Users\enayat.aria\PycharmProjects\pythonProject\venv\lib\site-packages\ot\da.py", line 1950, in fit
returned_ = sinkhorn_l1l2_gl(
File "C:\Users\enayat.aria\PycharmProjects\pythonProject\venv\lib\site-packages\ot\da.py", line 239, in 
sinkhorn_l1l2_gl
return gcg(a, b, M, reg, eta, f, df, G0=None, numItermax=numItermax,
File "C:\Users\enayat.aria\PycharmProjects\pythonProject\venv\lib\site-packages\ot\optim.py", line 388, in gcg
G = G + alpha * deltaG
TypeError: unsupported operand type(s) for *: 'NoneType' and 'float'

Checking the parameters, I found that the G matrix obtained in the optim.py code has negative values in the last iteration; due to the last update.

Please let me know how to solve the problem, or if I should provide more information.

Best,

Environment (please complete the following information):

OS (e.g. MacOS, Windows, Linux): Windows 10
Python version: 3.9
How was POT installed (source, pip, conda): pip

bug help wanted

opened by EnayatAria 12

Bugs for GPU based ot.sinkhorn, and very slow speed for GPU based ot.sinkhorn2

Hi, I met a bug when I intend to use GPU to calculate sinkhorn OT:

RuntimeError: CUDA error: device-side assert triggered CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1

Also I will meet this bug after running ot.sinkhorn2 for a while.

The errors come from my inside network codes, which is caused by choosing ot.sinkorn. If I choose ot.emd, then I will not meet such an error. Could you please help me? Thanks.
bug help wanted

opened by HelloWorldLTY 1

Problem installing POT on Python 3.6

when I tried to install POT use the commond "pip install POT", I always meet problem “Could not build wheels for POT, which is required to install pyproject.toml-based projects”

Collecting POT
  Using cached POT-0.8.2.tar.gz (255 kB)
  Installing build dependencies ... done
  Getting requirements to build wheel ... done
  Preparing metadata (pyproject.toml) ... done
Requirement already satisfied: numpy>=1.16 in ./anaconda3/envs/py36/lib/python3.6/site-packages (from POT) (1.19.2)
Requirement already satisfied: scipy>=1.0 in ./anaconda3/envs/py36/lib/python3.6/site-packages (from POT) (1.5.4)
Building wheels for collected packages: POT
  Building wheel for POT (pyproject.toml) ... error
  ERROR: Command errored out with exit status 1:
   command: /home/ll20/anaconda3/envs/py36/bin/python3 /home/ll20/anaconda3/envs/py36/lib/python3.6/site-packages/pip/_vendor/pep517/in_process/_in_process.py build_wheel /tmp/tmpw4i_wpdh
       cwd: /tmp/pip-install-9kg4vxnm/pot_4e7b690d6a354c238d684ac7e449ad42
  Complete output (346 lines):
  running bdist_wheel
  running build
  running build_py
  creating build
  creating build/lib.linux-x86_64-3.6
  creating build/lib.linux-x86_64-3.6/ot
  copying ot/__init__.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/backend.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/bregman.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/da.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/datasets.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/dr.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/factored.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/gromov.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/optim.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/partial.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/plot.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/regpath.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/sliced.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/smooth.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/stochastic.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/unbalanced.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/utils.py -> build/lib.linux-x86_64-3.6/ot
  copying ot/weak.py -> build/lib.linux-x86_64-3.6/ot
  creating build/lib.linux-x86_64-3.6/ot/helpers
  copying ot/helpers/__init__.py -> build/lib.linux-x86_64-3.6/ot/helpers
  copying ot/helpers/openmp_helpers.py -> build/lib.linux-x86_64-3.6/ot/helpers
  copying ot/helpers/pre_build_helpers.py -> build/lib.linux-x86_64-3.6/ot/helpers
  creating build/lib.linux-x86_64-3.6/ot/lp
  copying ot/lp/__init__.py -> build/lib.linux-x86_64-3.6/ot/lp
  copying ot/lp/cvx.py -> build/lib.linux-x86_64-3.6/ot/lp
  copying ot/lp/solver_1d.py -> build/lib.linux-x86_64-3.6/ot/lp
  running build_ext
  building 'ot.lp.emd_wrap' extension
  C compiler: gcc -pthread -B /home/ll20/anaconda3/envs/py36/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC

  creating build/temp.linux-x86_64-3.6
  creating build/temp.linux-x86_64-3.6/ot
  creating build/temp.linux-x86_64-3.6/ot/lp
  compile options: '-Iot/lp -I/tmp/pip-build-env-dgob8pab/overlay/lib/python3.6/site-packages/numpy/core/include -I/tmp/pip-install-9kg4vxnm/pot_4e7b690d6a354c238d684ac7e449ad42/ot/lp -I/home/ll20/anaconda3/envs/py36/include/python3.6m -c'
  extra options: '-O3 -fopenmp -DOMP'
  gcc: ot/lp/emd_wrap.cpp
  cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++ [enabled by default]
  In file included from /tmp/pip-build-env-dgob8pab/overlay/lib/python3.6/site-packages/numpy/core/include/numpy/ndarraytypes.h:1809:0,
                   from /tmp/pip-build-env-dgob8pab/overlay/lib/python3.6/site-packages/numpy/core/include/numpy/ndarrayobject.h:18,
                   from /tmp/pip-build-env-dgob8pab/overlay/lib/python3.6/site-packages/numpy/core/include/numpy/arrayobject.h:4,
                   from ot/lp/emd_wrap.cpp:792:
  /tmp/pip-build-env-dgob8pab/overlay/lib/python3.6/site-packages/numpy/core/include/numpy/npy_1_7_deprecated_api.h:15:2: warning: #warning "Using deprecated NumPy API, disable it by " "#defining NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION" [-Wcpp]
   #warning "Using deprecated NumPy API, disable it by " \
    ^
  gcc: ot/lp/EMD_wrapper.cpp
  cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++ [enabled by default]
  In file included from /usr/include/c++/4.8.2/cstdint:35:0,
                   from ot/lp/full_bipartitegraph.h:29,
                   from ot/lp/network_simplex_simple.h:67,
                   from ot/lp/EMD_wrapper.cpp:16:
  /usr/include/c++/4.8.2/bits/c++0x_warning.h:32:2: error: #error This file requires compiler and library support for the ISO C++ 2011 standard. This support is currently experimental, and must be enabled with the -std=c++11 or -std=gnu++11 compiler options.
   #error This file requires compiler and library support for the \
    ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h:119:3: error: ‘unordered_map’ in namespace ‘std’ does not name a type
     std::unordered_map<size_t, T> data;
     ^
  ot/lp/network_simplex_simple_omp.h: In member function ‘T lemon_omp::SparseValueVector<T>::operator[](size_t) const’:
  ot/lp/network_simplex_simple_omp.h:102:18: error: ‘unordered_map’ in namespace ‘std’ does not name a type
      typename std::unordered_map<size_t, T>::const_iterator it = data.find(id);
                    ^
  ot/lp/network_simplex_simple_omp.h:102:31: error: expected unqualified-id before ‘<’ token
      typename std::unordered_map<size_t, T>::const_iterator it = data.find(id);
                                 ^
  ot/lp/network_simplex_simple_omp.h:106:8: error: ‘it’ was not declared in this scope
      if (it == data.end())
          ^
  ot/lp/network_simplex_simple_omp.h:106:14: error: ‘data’ was not declared in this scope
      if (it == data.end())
                ^
  ot/lp/network_simplex_simple_omp.h: In member function ‘lemon_omp::ProxyObject<T>::operator T()’:
  ot/lp/network_simplex_simple_omp.h:142:18: error: ‘unordered_map’ in namespace ‘std’ does not name a type
      typename std::unordered_map<size_t, T>::iterator it = _v->data.find(_idx);
                    ^
  ot/lp/network_simplex_simple_omp.h:142:31: error: expected unqualified-id before ‘<’ token
      typename std::unordered_map<size_t, T>::iterator it = _v->data.find(_idx);
                                 ^
  ot/lp/network_simplex_simple_omp.h:146:8: error: ‘it’ was not declared in this scope
      if (it == _v->data.end())
          ^
  ot/lp/network_simplex_simple_omp.h: In member function ‘void lemon_omp::ProxyObject<T>::operator+=(T)’:
  ot/lp/network_simplex_simple_omp.h:156:18: error: ‘unordered_map’ in namespace ‘std’ does not name a type
      typename std::unordered_map<size_t, T>::iterator it = _v->data.find(_idx);
                    ^
  ot/lp/network_simplex_simple_omp.h:156:31: error: expected unqualified-id before ‘<’ token
      typename std::unordered_map<size_t, T>::iterator it = _v->data.find(_idx);
                                 ^
  ot/lp/network_simplex_simple_omp.h:160:8: error: ‘it’ was not declared in this scope
      if (it == _v->data.end())
          ^
  ot/lp/network_simplex_simple_omp.h: In member function ‘void lemon_omp::ProxyObject<T>::operator-=(T)’:
  ot/lp/network_simplex_simple_omp.h:175:18: error: ‘unordered_map’ in namespace ‘std’ does not name a type
      typename std::unordered_map<size_t, T>::iterator it = _v->data.find(_idx);
                    ^
  ot/lp/network_simplex_simple_omp.h:175:31: error: expected unqualified-id before ‘<’ token
      typename std::unordered_map<size_t, T>::iterator it = _v->data.find(_idx);
                                 ^
  ot/lp/network_simplex_simple_omp.h:179:8: error: ‘it’ was not declared in this scope
      if (it == _v->data.end())
          ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h: In member function ‘Number lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::totalCost() const’:
  ot/lp/network_simplex_simple_omp.h:1006:18: error: ‘unordered_map’ in namespace ‘std’ does not name a type
      typename std::unordered_map<size_t, Value>::const_iterator it;
                    ^
  ot/lp/network_simplex_simple_omp.h:1006:31: error: expected unqualified-id before ‘<’ token
      typename std::unordered_map<size_t, Value>::const_iterator it;
                                 ^
  ot/lp/network_simplex_simple_omp.h:1010:9: error: ‘it’ was not declared in this scope
      for (it = _flow.data.begin(); it!=_flow.data.end(); ++it)
           ^
  In file included from ot/lp/EMD_wrapper.cpp:16:0:
  ot/lp/network_simplex_simple.h: In instantiation of ‘lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::NetworkSimplexSimple(const GR&, bool, int, ArcsType, size_t) [with GR = lemon::FullBipartiteDigraph; V = double; C = double; NodesType = unsigned int; ArcsType = long int; size_t = long unsigned int]’:
  ot/lp/EMD_wrapper.cpp:57:116:   required from here
  ot/lp/network_simplex_simple.h:523:18: warning: ‘lemon::NetworkSimplexSimple<lemon::FullBipartiteDigraph, double, double, unsigned int>::_init_nb_arcs’ will be initialized after [-Wreorder]
           ArcsType _init_nb_arcs;
                    ^
  ot/lp/network_simplex_simple.h:365:21: warning:   ‘const Value lemon::NetworkSimplexSimple<lemon::FullBipartiteDigraph, double, double, unsigned int>::MAX’ [-Wreorder]
           const Value MAX;
                       ^
  ot/lp/network_simplex_simple.h:236:9: warning:   when initialized here [-Wreorder]
           NetworkSimplexSimple(const GR& graph, bool arc_mixing, int nbnodes, ArcsType nb_arcs, size_t maxiters) :
           ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h: In instantiation of ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::NetworkSimplexSimple(const GR&, bool, int, ArcsType, size_t, int) [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int; size_t = long unsigned int]’:
  ot/lp/EMD_wrapper.cpp:159:128:   required from here
  ot/lp/network_simplex_simple_omp.h:661:12: warning: ‘lemon_omp::NetworkSimplexSimple<lemon_omp::FullBipartiteDigraph, double, double, unsigned int>::_init_nb_arcs’ will be initialized after [-Wreorder]
     ArcsType _init_nb_arcs;
              ^
  ot/lp/network_simplex_simple_omp.h:392:15: warning:   ‘const Value lemon_omp::NetworkSimplexSimple<lemon_omp::FullBipartiteDigraph, double, double, unsigned int>::MAX’ [-Wreorder]
     const Value MAX;
                 ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h:247:3: warning:   when initialized here [-Wreorder]
     NetworkSimplexSimple(const GR& graph, bool arc_mixing, int nbnodes, ArcsType nb_arcs, size_t maxiters = 0, int numThreads=-1) :
     ^
  In file included from ot/lp/EMD_wrapper.cpp:16:0:
  ot/lp/network_simplex_simple.h: In instantiation of ‘lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::ProblemType lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::start() [with PivotRuleImpl = lemon::NetworkSimplexSimple<lemon::FullBipartiteDigraph, double, double, unsigned int>::BlockSearchPivotRule; GR = lemon::FullBipartiteDigraph; V = double; C = double; NodesType = unsigned int; ArcsType = long int]’:
  ot/lp/network_simplex_simple.h:1419:25:   required from ‘lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::ProblemType lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::start() [with GR = lemon::FullBipartiteDigraph; V = double; C = double; NodesType = unsigned int; ArcsType = long int]’
  ot/lp/network_simplex_simple.h:700:26:   required from ‘lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::ProblemType lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::run() [with GR = lemon::FullBipartiteDigraph; V = double; C = double; NodesType = unsigned int; ArcsType = long int]’
  ot/lp/EMD_wrapper.cpp:97:21:   required from here
  ot/lp/network_simplex_simple.h:1436:434: warning: format ‘%d’ expects argument of type ‘int’, but argument 3 has type ‘size_t {aka long unsigned int}’ [-Wformat=]
                       sprintf( errMess, "RESULT MIGHT BE INACURATE\nMax number of iteration reached, currently \%d. Sometimes iterations go on in cycle even though the solution has been reached, to check if it's the case here have a look at the minimal reduced cost. If it is very close to machine precision, you might actually have the correct solution, if not try setting the maximum number of iterations a bit higher\n",iter_number );
                                                                                                                                                                                                                                                                                                                                                                                                                                                    ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h: In instantiation of ‘lemon_omp::ProxyObject<T>& lemon_omp::ProxyObject<T>::operator=(const T&) [with T = double]’:
  ot/lp/network_simplex_simple_omp.h:1145:16:   required from ‘bool lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::init() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:833:14:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::run() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/EMD_wrapper.cpp:199:21:   required from here
  ot/lp/network_simplex_simple_omp.h:134:13: error: ‘class lemon_omp::SparseValueVector<double>’ has no member named ‘data’
       _v->data[_idx] = v;
               ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h: In instantiation of ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::start() [with PivotRuleImpl = lemon_omp::NetworkSimplexSimple<lemon_omp::FullBipartiteDigraph, double, double, unsigned int>::BlockSearchPivotRule; GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’:
  ot/lp/network_simplex_simple_omp.h:1555:16:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::start() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:838:17:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::run() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/EMD_wrapper.cpp:199:21:   required from here
  ot/lp/network_simplex_simple_omp.h:1614:419: warning: format ‘%d’ expects argument of type ‘int’, but argument 3 has type ‘size_t {aka long unsigned int}’ [-Wformat=]
        sprintf( errMess, "RESULT MIGHT BE INACURATE\nMax number of iteration reached, currently \%d. Sometimes iterations go on in cycle even though the solution has been reached, to check if it's the case here have a look at the minimal reduced cost. If it is very close to machine precision, you might actually have the correct solution, if not try setting the maximum number of iterations a bit higher\n",iter_number );
                                                                                                                                                                                                                                                                                                                                                                                                                                     ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h: In instantiation of ‘lemon_omp::ProxyObject<T>::operator T() [with T = double]’:
  ot/lp/network_simplex_simple_omp.h:1659:19:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::start() [with PivotRuleImpl = lemon_omp::NetworkSimplexSimple<lemon_omp::FullBipartiteDigraph, double, double, unsigned int>::BlockSearchPivotRule; GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:1555:16:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::start() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:838:17:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::run() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/EMD_wrapper.cpp:199:21:   required from here
  ot/lp/network_simplex_simple_omp.h:146:11: error: ‘class lemon_omp::SparseValueVector<double>’ has no member named ‘data’
      if (it == _v->data.end())
             ^
  ot/lp/network_simplex_simple_omp.h: In instantiation of ‘void lemon_omp::ProxyObject<T>::operator+=(T) [with T = double]’:
  ot/lp/network_simplex_simple_omp.h:1302:19:   required from ‘void lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::changeFlow(bool) [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:1595:23:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::start() [with PivotRuleImpl = lemon_omp::NetworkSimplexSimple<lemon_omp::FullBipartiteDigraph, double, double, unsigned int>::BlockSearchPivotRule; GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:1555:16:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::start() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:838:17:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::run() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/EMD_wrapper.cpp:199:21:   required from here
  ot/lp/network_simplex_simple_omp.h:160:11: error: ‘class lemon_omp::SparseValueVector<double>’ has no member named ‘data’
      if (it == _v->data.end())
             ^
  ot/lp/network_simplex_simple_omp.h:161:13: error: ‘class lemon_omp::SparseValueVector<double>’ has no member named ‘data’
       _v->data[_idx] = val;
               ^
  ot/lp/network_simplex_simple_omp.h:166:6: error: ‘class lemon_omp::SparseValueVector<double>’ has no member named ‘data’
        _v->data.erase(it);
        ^
  ot/lp/network_simplex_simple_omp.h: In member function ‘lemon_omp::ProxyObject<T>::operator T() [with T = double]’:
  ot/lp/network_simplex_simple_omp.h:150:3: warning: control reaches end of non-void function [-Wreturn-type]
     }
     ^
  ot/lp/network_simplex_simple_omp.h: In member function ‘T lemon_omp::SparseValueVector<T>::operator[](size_t) const [with T = double; size_t = long unsigned int]’:
  ot/lp/network_simplex_simple_omp.h:110:3: warning: control reaches end of non-void function [-Wreturn-type]
     }
     ^
  cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++ [enabled by default]
  In file included from /usr/include/c++/4.8.2/cstdint:35:0,
                   from ot/lp/full_bipartitegraph.h:29,
                   from ot/lp/network_simplex_simple.h:67,
                   from ot/lp/EMD_wrapper.cpp:16:
  /usr/include/c++/4.8.2/bits/c++0x_warning.h:32:2: error: #error This file requires compiler and library support for the ISO C++ 2011 standard. This support is currently experimental, and must be enabled with the -std=c++11 or -std=gnu++11 compiler options.
   #error This file requires compiler and library support for the \
    ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h:119:3: error: ‘unordered_map’ in namespace ‘std’ does not name a type
     std::unordered_map<size_t, T> data;
     ^
  ot/lp/network_simplex_simple_omp.h: In member function ‘T lemon_omp::SparseValueVector<T>::operator[](size_t) const’:
  ot/lp/network_simplex_simple_omp.h:102:18: error: ‘unordered_map’ in namespace ‘std’ does not name a type
      typename std::unordered_map<size_t, T>::const_iterator it = data.find(id);
                    ^
  ot/lp/network_simplex_simple_omp.h:102:31: error: expected unqualified-id before ‘<’ token
      typename std::unordered_map<size_t, T>::const_iterator it = data.find(id);
                                 ^
  ot/lp/network_simplex_simple_omp.h:106:8: error: ‘it’ was not declared in this scope
      if (it == data.end())
          ^
  ot/lp/network_simplex_simple_omp.h:106:14: error: ‘data’ was not declared in this scope
      if (it == data.end())
                ^
  ot/lp/network_simplex_simple_omp.h: In member function ‘lemon_omp::ProxyObject<T>::operator T()’:
  ot/lp/network_simplex_simple_omp.h:142:18: error: ‘unordered_map’ in namespace ‘std’ does not name a type
      typename std::unordered_map<size_t, T>::iterator it = _v->data.find(_idx);
                    ^
  ot/lp/network_simplex_simple_omp.h:142:31: error: expected unqualified-id before ‘<’ token
      typename std::unordered_map<size_t, T>::iterator it = _v->data.find(_idx);
                                 ^
  ot/lp/network_simplex_simple_omp.h:146:8: error: ‘it’ was not declared in this scope
      if (it == _v->data.end())
          ^
  ot/lp/network_simplex_simple_omp.h: In member function ‘void lemon_omp::ProxyObject<T>::operator+=(T)’:
  ot/lp/network_simplex_simple_omp.h:156:18: error: ‘unordered_map’ in namespace ‘std’ does not name a type
      typename std::unordered_map<size_t, T>::iterator it = _v->data.find(_idx);
                    ^
  ot/lp/network_simplex_simple_omp.h:156:31: error: expected unqualified-id before ‘<’ token
      typename std::unordered_map<size_t, T>::iterator it = _v->data.find(_idx);
                                 ^
  ot/lp/network_simplex_simple_omp.h:160:8: error: ‘it’ was not declared in this scope
      if (it == _v->data.end())
          ^
  ot/lp/network_simplex_simple_omp.h: In member function ‘void lemon_omp::ProxyObject<T>::operator-=(T)’:
  ot/lp/network_simplex_simple_omp.h:175:18: error: ‘unordered_map’ in namespace ‘std’ does not name a type
      typename std::unordered_map<size_t, T>::iterator it = _v->data.find(_idx);
                    ^
  ot/lp/network_simplex_simple_omp.h:175:31: error: expected unqualified-id before ‘<’ token
      typename std::unordered_map<size_t, T>::iterator it = _v->data.find(_idx);
                                 ^
  ot/lp/network_simplex_simple_omp.h:179:8: error: ‘it’ was not declared in this scope
      if (it == _v->data.end())
          ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h: In member function ‘Number lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::totalCost() const’:
  ot/lp/network_simplex_simple_omp.h:1006:18: error: ‘unordered_map’ in namespace ‘std’ does not name a type
      typename std::unordered_map<size_t, Value>::const_iterator it;
                    ^
  ot/lp/network_simplex_simple_omp.h:1006:31: error: expected unqualified-id before ‘<’ token
      typename std::unordered_map<size_t, Value>::const_iterator it;
                                 ^
  ot/lp/network_simplex_simple_omp.h:1010:9: error: ‘it’ was not declared in this scope
      for (it = _flow.data.begin(); it!=_flow.data.end(); ++it)
           ^
  In file included from ot/lp/EMD_wrapper.cpp:16:0:
  ot/lp/network_simplex_simple.h: In instantiation of ‘lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::NetworkSimplexSimple(const GR&, bool, int, ArcsType, size_t) [with GR = lemon::FullBipartiteDigraph; V = double; C = double; NodesType = unsigned int; ArcsType = long int; size_t = long unsigned int]’:
  ot/lp/EMD_wrapper.cpp:57:116:   required from here
  ot/lp/network_simplex_simple.h:523:18: warning: ‘lemon::NetworkSimplexSimple<lemon::FullBipartiteDigraph, double, double, unsigned int>::_init_nb_arcs’ will be initialized after [-Wreorder]
           ArcsType _init_nb_arcs;
                    ^
  ot/lp/network_simplex_simple.h:365:21: warning:   ‘const Value lemon::NetworkSimplexSimple<lemon::FullBipartiteDigraph, double, double, unsigned int>::MAX’ [-Wreorder]
           const Value MAX;
                       ^
  ot/lp/network_simplex_simple.h:236:9: warning:   when initialized here [-Wreorder]
           NetworkSimplexSimple(const GR& graph, bool arc_mixing, int nbnodes, ArcsType nb_arcs, size_t maxiters) :
           ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h: In instantiation of ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::NetworkSimplexSimple(const GR&, bool, int, ArcsType, size_t, int) [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int; size_t = long unsigned int]’:
  ot/lp/EMD_wrapper.cpp:159:128:   required from here
  ot/lp/network_simplex_simple_omp.h:661:12: warning: ‘lemon_omp::NetworkSimplexSimple<lemon_omp::FullBipartiteDigraph, double, double, unsigned int>::_init_nb_arcs’ will be initialized after [-Wreorder]
     ArcsType _init_nb_arcs;
              ^
  ot/lp/network_simplex_simple_omp.h:392:15: warning:   ‘const Value lemon_omp::NetworkSimplexSimple<lemon_omp::FullBipartiteDigraph, double, double, unsigned int>::MAX’ [-Wreorder]
     const Value MAX;
                 ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h:247:3: warning:   when initialized here [-Wreorder]
     NetworkSimplexSimple(const GR& graph, bool arc_mixing, int nbnodes, ArcsType nb_arcs, size_t maxiters = 0, int numThreads=-1) :
     ^
  In file included from ot/lp/EMD_wrapper.cpp:16:0:
  ot/lp/network_simplex_simple.h: In instantiation of ‘lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::ProblemType lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::start() [with PivotRuleImpl = lemon::NetworkSimplexSimple<lemon::FullBipartiteDigraph, double, double, unsigned int>::BlockSearchPivotRule; GR = lemon::FullBipartiteDigraph; V = double; C = double; NodesType = unsigned int; ArcsType = long int]’:
  ot/lp/network_simplex_simple.h:1419:25:   required from ‘lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::ProblemType lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::start() [with GR = lemon::FullBipartiteDigraph; V = double; C = double; NodesType = unsigned int; ArcsType = long int]’
  ot/lp/network_simplex_simple.h:700:26:   required from ‘lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::ProblemType lemon::NetworkSimplexSimple<GR, V, C, NodesType, ArcsType>::run() [with GR = lemon::FullBipartiteDigraph; V = double; C = double; NodesType = unsigned int; ArcsType = long int]’
  ot/lp/EMD_wrapper.cpp:97:21:   required from here
  ot/lp/network_simplex_simple.h:1436:434: warning: format ‘%d’ expects argument of type ‘int’, but argument 3 has type ‘size_t {aka long unsigned int}’ [-Wformat=]
                       sprintf( errMess, "RESULT MIGHT BE INACURATE\nMax number of iteration reached, currently \%d. Sometimes iterations go on in cycle even though the solution has been reached, to check if it's the case here have a look at the minimal reduced cost. If it is very close to machine precision, you might actually have the correct solution, if not try setting the maximum number of iterations a bit higher\n",iter_number );
                                                                                                                                                                                                                                                                                                                                                                                                                                                    ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h: In instantiation of ‘lemon_omp::ProxyObject<T>& lemon_omp::ProxyObject<T>::operator=(const T&) [with T = double]’:
  ot/lp/network_simplex_simple_omp.h:1145:16:   required from ‘bool lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::init() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:833:14:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::run() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/EMD_wrapper.cpp:199:21:   required from here
  ot/lp/network_simplex_simple_omp.h:134:13: error: ‘class lemon_omp::SparseValueVector<double>’ has no member named ‘data’
       _v->data[_idx] = v;
               ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h: In instantiation of ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::start() [with PivotRuleImpl = lemon_omp::NetworkSimplexSimple<lemon_omp::FullBipartiteDigraph, double, double, unsigned int>::BlockSearchPivotRule; GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’:
  ot/lp/network_simplex_simple_omp.h:1555:16:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::start() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:838:17:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::run() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/EMD_wrapper.cpp:199:21:   required from here
  ot/lp/network_simplex_simple_omp.h:1614:419: warning: format ‘%d’ expects argument of type ‘int’, but argument 3 has type ‘size_t {aka long unsigned int}’ [-Wformat=]
        sprintf( errMess, "RESULT MIGHT BE INACURATE\nMax number of iteration reached, currently \%d. Sometimes iterations go on in cycle even though the solution has been reached, to check if it's the case here have a look at the minimal reduced cost. If it is very close to machine precision, you might actually have the correct solution, if not try setting the maximum number of iterations a bit higher\n",iter_number );
                                                                                                                                                                                                                                                                                                                                                                                                                                     ^
  In file included from ot/lp/EMD_wrapper.cpp:17:0:
  ot/lp/network_simplex_simple_omp.h: In instantiation of ‘lemon_omp::ProxyObject<T>::operator T() [with T = double]’:
  ot/lp/network_simplex_simple_omp.h:1659:19:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::start() [with PivotRuleImpl = lemon_omp::NetworkSimplexSimple<lemon_omp::FullBipartiteDigraph, double, double, unsigned int>::BlockSearchPivotRule; GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:1555:16:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::start() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:838:17:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::run() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/EMD_wrapper.cpp:199:21:   required from here
  ot/lp/network_simplex_simple_omp.h:146:11: error: ‘class lemon_omp::SparseValueVector<double>’ has no member named ‘data’
      if (it == _v->data.end())
             ^
  ot/lp/network_simplex_simple_omp.h: In instantiation of ‘void lemon_omp::ProxyObject<T>::operator+=(T) [with T = double]’:
  ot/lp/network_simplex_simple_omp.h:1302:19:   required from ‘void lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::changeFlow(bool) [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:1595:23:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::start() [with PivotRuleImpl = lemon_omp::NetworkSimplexSimple<lemon_omp::FullBipartiteDigraph, double, double, unsigned int>::BlockSearchPivotRule; GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:1555:16:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::start() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/network_simplex_simple_omp.h:838:17:   required from ‘lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::ProblemType lemon_omp::NetworkSimplexSimple<GR, V, C, ArcsType>::run() [with GR = lemon_omp::FullBipartiteDigraph; V = double; C = double; ArcsType = unsigned int]’
  ot/lp/EMD_wrapper.cpp:199:21:   required from here
  ot/lp/network_simplex_simple_omp.h:160:11: error: ‘class lemon_omp::SparseValueVector<double>’ has no member named ‘data’
      if (it == _v->data.end())
             ^
  ot/lp/network_simplex_simple_omp.h:161:13: error: ‘class lemon_omp::SparseValueVector<double>’ has no member named ‘data’
       _v->data[_idx] = val;
               ^
  ot/lp/network_simplex_simple_omp.h:166:6: error: ‘class lemon_omp::SparseValueVector<double>’ has no member named ‘data’
        _v->data.erase(it);
        ^
  ot/lp/network_simplex_simple_omp.h: In member function ‘lemon_omp::ProxyObject<T>::operator T() [with T = double]’:
  ot/lp/network_simplex_simple_omp.h:150:3: warning: control reaches end of non-void function [-Wreturn-type]
     }
     ^
  ot/lp/network_simplex_simple_omp.h: In member function ‘T lemon_omp::SparseValueVector<T>::operator[](size_t) const [with T = double; size_t = long unsigned int]’:
  ot/lp/network_simplex_simple_omp.h:110:3: warning: control reaches end of non-void function [-Wreturn-type]
     }
     ^
  error: Command "gcc -pthread -B /home/ll20/anaconda3/envs/py36/compiler_compat -Wl,--sysroot=/ -Wsign-compare -DNDEBUG -g -fwrapv -O3 -Wall -Wstrict-prototypes -fPIC -Iot/lp -I/tmp/pip-build-env-dgob8pab/overlay/lib/python3.6/site-packages/numpy/core/include -I/tmp/pip-install-9kg4vxnm/pot_4e7b690d6a354c238d684ac7e449ad42/ot/lp -I/home/ll20/anaconda3/envs/py36/include/python3.6m -c ot/lp/EMD_wrapper.cpp -o build/temp.linux-x86_64-3.6/ot/lp/EMD_wrapper.o -O3 -fopenmp -DOMP" failed with exit status 1
  ----------------------------------------
  ERROR: Failed building wheel for POT
Failed to build POT
ERROR: Could not build wheels for POT, which is required to install pyproject.toml-based projects```

bug help wanted

opened by LL0912 1

Negative Gromov-Wasserstein distance
Hi,

Thank you for the extremely helpful software! Sometimes I encounter a negative value of gw_dist returned by gromov.entropic_gromov_wasserstein() function:

gw1, log1 = ot.gromov.entropic_gromov_wasserstein( C1, C2, p, q, 'square_loss', epsilon=0.01, log=True, verbose=True) print(log1['gw_dist'])

However, from my understanding, the distance part has to be nonnegative given the square loss. Could you help me with my confusion? If helpful at all, my cost matrices are nonnegative, symmetric integer matrices of sizes 267267 and 250250, and the marginals p and q are uniform. The negative gw_dist value appears for epsilon=0.01, but not for epsilon=0.02 or 0.05.

The version of packages is:

macOS-10.16-x86_64-i386-64bit Python 3.8.8 (default, Apr 13 2021, 12:59:45) [Clang 10.0.0 ] NumPy 1.20.1 SciPy 1.6.2 POT 0.8.1.0
opened by WenjunZHAOwO 4
[WIP] Semi-relaxed (fused) gromov-wasserstein divergence with updates of current gromov-wasserstein solvers and dependencies
Types of changes

Integration of the semi-relaxed (fused) gromov-wasserstein solvers using conditional gradients --- Related new functions in gromov.py: init_matrix_semirelaxed / semirelaxed_gromov_wasserstein / semirelaxed_gromov_wasserstein2 / semirelaxed_fused_gromov_wasserstein/ semirelaxed_fused_gromov_wasserstein2/ --- Related new functions in optim.py: solve_semirelaxed_gromov_linesearch.

Factorization of the conditional gradient solvers in optim.py to take as entry any kind of linear program solver for the conditional gradient direction, and any kind of line search solver for the optimal line search step. --- Related new functions in optim.py: generic_cg / Might replace functions in optim.py: cg, gcg (an instance of how to replace cgc is given in a potentially temporary function named new_gcg)

New factoring for the (fused) gromov-wasserstein functions to fit the new function generic_cg, and support asymmetric matrices as started in another pull request [PR: Gromov-Wasserstein with asymmetric cost matrices #399] --- Related modified functions in gromov.py: gromov_wasserstein / gromov_wasserstein2 / fused_gromov_wasserstein / fused_gromov_wasserstein2 / gromov_barycenters / fgw_barycenters --- Related new functions in optim.py: solve_gromov_linesearch (generalizing the line seach coded in solve_linesearch, plus speeding up the function).

Updates of the (fused) gromov-wasserstein dictionary learning functions to leverage the new (fused) gromov-wasserstein solvers to also support asymmetric matrices.

[ TO DO]

update test_gromov.py

update related examples

update dependencies with cg and gcg if necessary.

add semi-relaxed (fused) gromov-wasserstein barycenters ?

Motivation and context / Related issue

The current conditional gradient solver (optim.py/cg) does not allow an easy integration of new lp solver, not new line search methods. The upgrade to the function generic_cg allows a such flexiblity.

The integration of the semi-relaxed fgw (srfgw) requires new lp solver and line-search solver, hence fits well with the function generic_cg. Moreover the implementation srfgw supports symmetric and asymmetric matrices, so fgw solvers have been updated to support asymmetric matrices for the sake of consistency.

The implementation of the (fused) gromov-wasserstein dictionary learning was also restrained to symmetric matrices. Actually wrong if asymmetric matrices were provided or if the dictionary atoms were not projected onto the set of symmetric matrices.

How has this been tested (if it applies)

Test to come.

PR checklist

[x] I have read the CONTRIBUTING document.

[ ] The documentation is up-to-date with the changes I made (check build artifacts).

[ ] All tests passed, and additional code has been covered with new tests.

[ ] I have added the PR and Issue fix to the RELEASES.md file.
opened by cedricvincentcuaz 1
[WIP] Gromov-Wasserstein with asymmetric cost matrices
Types of changes

Support of a wider class of arguments for Gromov-Wasserstein functions

Signature of 2 functions modified

Docstrings updated

Motivation and context / Related issue

In [12] "Gromov-Wasserstein Averaging of Kernel and Distance Matrices", the authors make no explicit assumption about the symmetry of the cost mastrices C and C'. However, their formula (9) for the computation of the gradient misses at least a 2 factor in the case of symmetric matrices, and even a second term in the case of asymmetric matrices, as explained in subsection 5.1 of "Multilingual Alignment of Word Embedding Spaces", Dan Meller and Gonzague de Carpentier, 2021 (https://www.researchgate.net/publication/357505092_Multilingual_alignment_of_word_embedding_spaces). In the code of ot/gromov.py, the 2 factor has been corrected, but it is not sufficient to support asymmetric cost matrices, which can be useful in the case of directed graphs for example. This contribution corrects the expression of the gradient in ot.gromov.gwggrad to support asymmetric distance matrices, by adding the appropriate term instead of just multiplying by 2.

How has this been tested (if it applies)

PR checklist

[x] I have read the CONTRIBUTING document.

[x] The documentation is up-to-date with the changes I made (check build artifacts).

[x] All tests passed, and additional code has been covered with new tests.

[x] I have added the PR and Issue fix to the RELEASES.md file.
opened by decarpentierg 3

Releases(0.8.2)

0.8.2(Apr 21, 2022)
This releases introduces several new notable features. The less important but most exiting one being that we now have a logo for the toolbox (color and dark background) :

This logo is generated using with matplotlib and using the solution of an OT problem provided by POT (with ot.emd). Generating the logo can be done with a simple python script also provided in the documentation gallery.

New OT solvers include Weak OT and OT with factored coupling that can be used on large datasets. The Majorization Minimization solvers for non-regularized Unbalanced OT are now also available. We also now provide an implementation of GW and FGW unmixing and dictionary learning. It is now possible to use autodiff to solve entropic an quadratic regularized OT in the dual for full or stochastic optimization thanks to the new functions to compute the dual loss for entropic and quadratic regularized OT and reconstruct the OT plan on part or all of the data. They can be used for instance to solve OT problems with stochastic gradient or for estimating the dual potentials as neural networks.

On the backend front, we now have backend compatible functions and classes in the domain adaptation ot.da and unbalanced OT ot.unbalanced modules. This means that the DA classes can be used on tensors from all compatible backends. The free support Wasserstein barycenter solver is now also backend compatible.

Finally we have worked on the documentation to provide an update of existing examples in the gallery and and several new examples including GW dictionary learning and weak Optimal Transport.

New features

Remove deprecated ot.gpu submodule (PR #361)

Update examples in the gallery (PR #359)

Add stochastic loss and OT plan computation for regularized OT and backend examples(PR #360)

Implementation of factored OT with emd and sinkhorn (PR #358)

A brand new logo for POT (PR #357)

Better list of related examples in quick start guide with minigallery (PR #334)

Add optional log-domain Sinkhorn implementation in WDA to support smaller values of the regularization parameter (PR #336)

Backend implementation for ot.lp.free_support_barycenter (PR #340)

Add weak OT solver + example (PR #341)

Add backend support for Domain Adaptation and Unbalanced solvers (PR #343)

Add (F)GW linear dictionary learning solvers + example (PR #319)

Add links to related PR and Issues in the doc release page (PR #350)

Add new minimization-maximization algorithms for solving exact Unbalanced OT + example (PR #362)

Closed issues

Fix mass gradient of ot.emd2 and ot.gromov_wasserstein2 so that they are centered (Issue #364, PR #363)

Fix bug in instantiating an autograd function ValFunction (Issue #337, PR #338)

Fix POT ABI compatibility with old and new numpy (Issue #346, PR #349)

Warning when feeding integer cost matrix to EMD solver resulting in an integer transport plan (Issue #345, PR #343)

Fix bug where gromov_wasserstein2 does not perform backpropagation with CUDA tensors (Issue #351, PR #352)

Source code(tar.gz)
Source code(zip)
POT-0.8.2-cp310-cp310-macosx_10_9_universal2.whl(246.39 KB)
POT-0.8.2-cp310-cp310-macosx_10_9_x86_64.whl(194.08 KB)
POT-0.8.2-cp310-cp310-macosx_11_0_arm64.whl(185.91 KB)
POT-0.8.2-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.whl(644.66 KB)
POT-0.8.2-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(655.37 KB)
POT-0.8.2-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(653.80 KB)
POT-0.8.2-cp310-cp310-win32.whl(183.86 KB)
POT-0.8.2-cp310-cp310-win_amd64.whl(188.34 KB)
POT-0.8.2-cp37-cp37m-macosx_10_9_x86_64.whl(193.64 KB)
POT-0.8.2-cp37-cp37m-manylinux_2_12_i686.manylinux2010_i686.whl(639.02 KB)
POT-0.8.2-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(649.35 KB)
POT-0.8.2-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(640.02 KB)
POT-0.8.2-cp37-cp37m-win32.whl(183.79 KB)
POT-0.8.2-cp37-cp37m-win_amd64.whl(188.36 KB)
POT-0.8.2-cp38-cp38-macosx_10_9_universal2.whl(246.36 KB)
POT-0.8.2-cp38-cp38-macosx_10_9_x86_64.whl(194.06 KB)
POT-0.8.2-cp38-cp38-macosx_11_0_arm64.whl(185.92 KB)
POT-0.8.2-cp38-cp38-manylinux_2_12_i686.manylinux2010_i686.whl(653.25 KB)
POT-0.8.2-cp38-cp38-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(666.29 KB)
POT-0.8.2-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(654.69 KB)
POT-0.8.2-cp38-cp38-win32.whl(184.02 KB)
POT-0.8.2-cp38-cp38-win_amd64.whl(188.41 KB)
POT-0.8.2-cp39-cp39-macosx_10_9_universal2.whl(245.51 KB)
POT-0.8.2-cp39-cp39-macosx_10_9_x86_64.whl(193.63 KB)
POT-0.8.2-cp39-cp39-macosx_11_0_arm64.whl(185.53 KB)
POT-0.8.2-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.whl(643.49 KB)
POT-0.8.2-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(654.50 KB)
POT-0.8.2-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(652.65 KB)
POT-0.8.2-cp39-cp39-win32.whl(183.63 KB)
POT-0.8.2-cp39-cp39-win_amd64.whl(188.14 KB)
POT-0.8.2.tar.gz(249.79 KB)
0.8.1.0(Dec 31, 2021)
This is a bug fix release that will remove the benchmarks module form the installation and correct the documentation generation.

Closed issues

Bug in documentation generation (tag VS master push, PR #332)

Remove installation of the benchmarks in global namespace (Issue #331, PR #333)

Source code(tar.gz)
Source code(zip)
POT-0.8.1.0-cp310-cp310-macosx_10_9_universal2.whl(238.16 KB)
POT-0.8.1.0-cp310-cp310-macosx_10_9_x86_64.whl(185.84 KB)
POT-0.8.1.0-cp310-cp310-macosx_11_0_arm64.whl(177.68 KB)
POT-0.8.1.0-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.whl(636.52 KB)
POT-0.8.1.0-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(647.22 KB)
POT-0.8.1.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(645.64 KB)
POT-0.8.1.0-cp310-cp310-win32.whl(175.84 KB)
POT-0.8.1.0-cp310-cp310-win_amd64.whl(179.18 KB)
POT-0.8.1.0-cp37-cp37m-macosx_10_9_x86_64.whl(185.07 KB)
POT-0.8.1.0-cp37-cp37m-manylinux_2_12_i686.manylinux2010_i686.whl(630.01 KB)
POT-0.8.1.0-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(641.04 KB)
POT-0.8.1.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(632.47 KB)
POT-0.8.1.0-cp37-cp37m-win32.whl(175.51 KB)
POT-0.8.1.0-cp37-cp37m-win_amd64.whl(178.90 KB)
POT-0.8.1.0-cp38-cp38-macosx_10_9_universal2.whl(237.62 KB)
POT-0.8.1.0-cp38-cp38-macosx_10_9_x86_64.whl(185.48 KB)
POT-0.8.1.0-cp38-cp38-macosx_11_0_arm64.whl(177.52 KB)
POT-0.8.1.0-cp38-cp38-manylinux_2_12_i686.manylinux2010_i686.whl(644.25 KB)
POT-0.8.1.0-cp38-cp38-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(657.44 KB)
POT-0.8.1.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(646.88 KB)
POT-0.8.1.0-cp38-cp38-win32.whl(175.86 KB)
POT-0.8.1.0-cp38-cp38-win_amd64.whl(179.11 KB)
POT-0.8.1.0-cp39-cp39-macosx_10_9_universal2.whl(238.16 KB)
POT-0.8.1.0-cp39-cp39-macosx_10_9_x86_64.whl(185.83 KB)
POT-0.8.1.0-cp39-cp39-macosx_11_0_arm64.whl(177.68 KB)
POT-0.8.1.0-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.whl(635.81 KB)
POT-0.8.1.0-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(646.80 KB)
POT-0.8.1.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(644.78 KB)
POT-0.8.1.0-cp39-cp39-win32.whl(175.84 KB)
POT-0.8.1.0-cp39-cp39-win_amd64.whl(179.19 KB)
POT-0.8.1.0.tar.gz(236.13 KB)
0.8.1(Dec 27, 2021)
This release fixes several bugs and introduces two new backends: Cupy and Tensorflow. Note that the tensorflow backend will work only when tensorflow has enabled the Numpy behavior (for transpose that is not by default in tensorflow). We also introduce a simple benchmark on CPU GPU for the sinkhorn solver that will be provided in the backend documentation.

This release also brings a few changes in dependencies and compatibility. First we removed tests for Python 3.6 that will not be updated in the future. Also note that POT now depends on Numpy (>= 1.20) because a recent change in ABI is making the wheels non-compatible with older numpy versions. If you really need an older numpy POT will work with no problems but you will need to build it from source.

As always we want to that the contributors who helped make POT better (and bug free).

New features

New benchmark for sinkhorn solver on CPU/GPU and between backends (PR #316)

New tensorflow backend (PR #316)

New Cupy backend (PR #315)

Documentation always up-to-date with README, RELEASES, CONTRIBUTING and CODE_OF_CONDUCT files (PR #316, PR #322).

Closed issues

Fix bug in older Numpy ABI (<1.20) (Issue #308, PR #326)

Fix bug in ot.dist function when non euclidean distance (Issue #305, PR #306)

Fix gradient scaling for functions using nx.set_gradients (Issue #309, PR #310)

Fix bug in generalized Conditional gradient solver and SinkhornL1L2 (Issue #311, PR #313)

Fix log error in gromov_barycenters (Issue #317, PR #3018)

Source code(tar.gz)
Source code(zip)
POT-0.8.1-cp310-cp310-macosx_10_9_universal2.whl(241.02 KB)
POT-0.8.1-cp310-cp310-macosx_10_9_x86_64.whl(188.70 KB)
POT-0.8.1-cp310-cp310-macosx_11_0_arm64.whl(180.54 KB)
POT-0.8.1-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.whl(639.45 KB)
POT-0.8.1-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(650.15 KB)
POT-0.8.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(648.61 KB)
POT-0.8.1-cp310-cp310-win32.whl(178.74 KB)
POT-0.8.1-cp310-cp310-win_amd64.whl(182.08 KB)
POT-0.8.1-cp37-cp37m-macosx_10_9_x86_64.whl(187.92 KB)
POT-0.8.1-cp37-cp37m-manylinux_2_12_i686.manylinux2010_i686.whl(632.97 KB)
POT-0.8.1-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(644.00 KB)
POT-0.8.1-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(635.42 KB)
POT-0.8.1-cp37-cp37m-win32.whl(178.41 KB)
POT-0.8.1-cp37-cp37m-win_amd64.whl(181.80 KB)
POT-0.8.1-cp38-cp38-macosx_10_9_universal2.whl(240.48 KB)
POT-0.8.1-cp38-cp38-macosx_10_9_x86_64.whl(188.33 KB)
POT-0.8.1-cp38-cp38-macosx_11_0_arm64.whl(180.38 KB)
POT-0.8.1-cp38-cp38-manylinux_2_12_i686.manylinux2010_i686.whl(647.19 KB)
POT-0.8.1-cp38-cp38-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(660.38 KB)
POT-0.8.1-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(649.85 KB)
POT-0.8.1-cp38-cp38-win32.whl(178.75 KB)
POT-0.8.1-cp38-cp38-win_amd64.whl(182.01 KB)
POT-0.8.1-cp39-cp39-macosx_10_9_universal2.whl(241.02 KB)
POT-0.8.1-cp39-cp39-macosx_10_9_x86_64.whl(188.69 KB)
POT-0.8.1-cp39-cp39-macosx_11_0_arm64.whl(180.54 KB)
POT-0.8.1-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.whl(638.76 KB)
POT-0.8.1-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(649.72 KB)
POT-0.8.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(647.74 KB)
POT-0.8.1-cp39-cp39-win32.whl(178.73 KB)
POT-0.8.1-cp39-cp39-win_amd64.whl(182.08 KB)
POT-0.8.1.tar.gz(235.13 KB)
0.8.0(Nov 5, 2021)
This new stable release introduces several important features.

First we now have an OpenMP compatible exact ot solver in ot.emd. The OpenMP version is used when the parameter numThreads is greater than one and can lead to nice speedups on multi-core machines.

Second we have introduced a backend mechanism that allows to use standard POT function seamlessly on Numpy, Pytorch and Jax arrays. Other backends are coming but right now POT can be used seamlessly for training neural networks in Pytorch. Notably we propose the first differentiable computation of the exact OT loss with ot.emd2 (can be differentiated w.r.t. both cost matrix and sample weights), but also for the classical Sinkhorn loss with ot.sinkhorn2, the Wasserstein distance in 1D with ot.wasserstein_1d, sliced Wasserstein with ot.sliced_wasserstein_distance and Gromov-Wasserstein with ot.gromov_wasserstein2 . Examples of how this new feature can be used are now available in the documentation where the Pytorch backend is used to estimate a minimal Wasserstein estimator, a Generative Network (GAN), for a sliced Wasserstein gradient flow and optimizing the Gromov-Wassersein distance. Note that the Jax backend is still in early development and quite slow at the moment, we strongly recommend for Jax users to use the OTT toolbox when possible. As a result of this new feature, the old ot.gpu submodule is now deprecated since GPU implementations can be done using GPU arrays on the torch backends.

Other novel features include implementation for Sampled Gromov Wasserstein and Pointwise Gromov Wasserstein, Sinkhorn in log space with method='sinkhorn_log', Projection Robust Wasserstein, and deviased Sinkorn barycenters.

This release will also simplify the installation process. We have now a pyproject.toml that defines the build dependency and POT should now build even when cython is not installed yet. Also we now provide pe-compiled wheels for linux aarch64 that is used on Raspberry PI and android phones and for MacOS on ARM processors.

Finally POT was accepted for publication in the Journal of Machine Learning Research (JMLR) open source software track and we ask the POT users to cite this paper from now on. The documentation has been improved in particular by adding a "Why OT?" section to the quick start guide and several new examples illustrating the new features. The documentation now has two version : the stable version https://pythonot.github.io/ corresponding to the last release and the master version https://pythonot.github.io/master that corresponds to the current master branch on GitHub.

As usual, we want to thank all the POT contributors (now 37 people have contributed to the toolbox). But for this release we thank in particular Nathan Cassereau and Kamel Guerda from the AI support team at IDRIS for their support to the development of the backend and OpenMP implementations.

New features

OpenMP support for exact OT solvers (PR #260)

Backend for running POT in numpy/torch + exact solver (PR #249)

Backend implementation of most functions in ot.bregman (PR #280)

Backend implementation of most functions in ot.optim (PR #282)

Backend implementation of most functions in ot.gromov (PR #294, PR #302)

Test for arrays of different type and device (CPU/GPU) (PR #304, #303)

Implementation of Sinkhorn in log space with method='sinkhorn_log' (PR #290)

Implementation of regularization path for L2 Unbalanced OT (PR #274)

Implementation of Projection Robust Wasserstein (PR #267)

Implementation of Debiased Sinkhorn Barycenters (PR #291)

Implementation of Sampled Gromov Wasserstein and Pointwise Gromov Wasserstein (PR #275)

Add pyproject.toml and build POT without installing cython first (PR #293)

Lazy implementation in log space for sinkhorn on samples (PR #259)

Documentation cleanup (PR #298)

Two up-to-date documentations for stable release and for master branch.

Building wheels on ARM for Raspberry PI and smartphones (PR #238)

Update build wheels to new version and new pythons (PR #236, #253)

Implementation of sliced Wasserstein distance (Issue #202, PR #203)

Add minimal build to CI and perform pep8 test separately (PR #210)

Speedup of tests and return run time (PR #262)

Add "Why OT" discussion to the documentation (PR #220)

New introductory example to discrete OT in the documentation (PR #191)

Add templates for Issues/PR on Github (PR#181)

Closed issues

Debug Memory leak in GAN example (#254)

DEbug GPU bug (Issue #284, #287, PR #288)

set_gradients method for JAX backend (PR #278)

Quicker GAN example for CircleCI build (PR #258)

Better formatting in Readme (PR #234)

Debug CI tests (PR #240, #241, #242)

Bug in Partial OT solver dummy points (PR #215)

Bug when Armijo linesearch (Issue #184, #198, #281, PR #189, #199, #286)

Bug Barycenter Sinkhorn (Issue 134, PR #195)

Infeasible solution in exact OT (Issues #126,#93, PR #217)

Doc for SUpport Barycenters (Issue #200, PR #201)

Fix labels transport in BaseTransport (Issue #207, PR #208)

Bug in emd_1d, non respected bounds (Issue #169, PR #170)

Removed Python 2.7 support and update codecov file (PR #178)

Add normalization for WDA and test it (PR #172, #296)

Cleanup code for new version of flake8 (PR #176)

Fixed requirements in setup.py (PR #174)

Removed specific MacOS flags (PR #175)

Source code(tar.gz)
Source code(zip)
POT-0.8.0-cp310-cp310-macosx_10_9_universal2.whl(234.65 KB)
POT-0.8.0-cp310-cp310-macosx_10_9_x86_64.whl(181.93 KB)
POT-0.8.0-cp310-cp310-macosx_11_0_arm64.whl(173.97 KB)
POT-0.8.0-cp310-cp310-manylinux_2_12_i686.manylinux2010_i686.whl(632.13 KB)
POT-0.8.0-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(643.01 KB)
POT-0.8.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(641.57 KB)
POT-0.8.0-cp310-cp310-win32.whl(171.60 KB)
POT-0.8.0-cp310-cp310-win_amd64.whl(174.97 KB)
POT-0.8.0-cp36-cp36m-macosx_10_9_x86_64.whl(180.64 KB)
POT-0.8.0-cp36-cp36m-manylinux_2_12_i686.manylinux2010_i686.whl(622.29 KB)
POT-0.8.0-cp36-cp36m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(633.07 KB)
POT-0.8.0-cp36-cp36m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(625.38 KB)
POT-0.8.0-cp36-cp36m-win32.whl(170.94 KB)
POT-0.8.0-cp36-cp36m-win_amd64.whl(174.37 KB)
POT-0.8.0-cp37-cp37m-macosx_10_9_x86_64.whl(181.11 KB)
POT-0.8.0-cp37-cp37m-manylinux_2_12_i686.manylinux2010_i686.whl(624.48 KB)
POT-0.8.0-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(635.17 KB)
POT-0.8.0-cp37-cp37m-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(628.22 KB)
POT-0.8.0-cp37-cp37m-win32.whl(171.20 KB)
POT-0.8.0-cp37-cp37m-win_amd64.whl(174.60 KB)
POT-0.8.0-cp38-cp38-macosx_10_9_universal2.whl(234.00 KB)
POT-0.8.0-cp38-cp38-macosx_10_9_x86_64.whl(181.33 KB)
POT-0.8.0-cp38-cp38-macosx_11_0_arm64.whl(173.90 KB)
POT-0.8.0-cp38-cp38-manylinux_2_12_i686.manylinux2010_i686.whl(638.92 KB)
POT-0.8.0-cp38-cp38-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(652.18 KB)
POT-0.8.0-cp38-cp38-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(642.19 KB)
POT-0.8.0-cp38-cp38-win32.whl(171.56 KB)
POT-0.8.0-cp38-cp38-win_amd64.whl(174.86 KB)
POT-0.8.0-cp39-cp39-macosx_10_9_universal2.whl(234.64 KB)
POT-0.8.0-cp39-cp39-macosx_10_9_x86_64.whl(181.92 KB)
POT-0.8.0-cp39-cp39-macosx_11_0_arm64.whl(173.96 KB)
POT-0.8.0-cp39-cp39-manylinux_2_12_i686.manylinux2010_i686.whl(631.43 KB)
POT-0.8.0-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl(642.58 KB)
POT-0.8.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl(640.61 KB)
POT-0.8.0-cp39-cp39-win32.whl(171.59 KB)
POT-0.8.0-cp39-cp39-win_amd64.whl(174.97 KB)
POT-0.8.0.tar.gz(226.68 KB)
0.7.0(May 5, 2020)
This is the new stable release for POT. We made a lot of changes in the documentation and added several new features such as Partial OT, Unbalanced and Multi Sources OT Domain Adaptation and several bug fixes. One important change is that we have created the GitHub organization PythonOT that now owns the main POT repository https://github.com/PythonOT/POT and the repository for the new documentation is now hosted at https://PythonOT.github.io/.

This is the first release where the Python 2.7 tests have been removed. Most of the toolbox should still work but we do not offer support for Python 2.7 and will close related Issues.

A lot of changes have been done to the documentation that is now hosted on https://PythonOT.github.io/ instead of readthedocs. It was a hard choice but readthedocs did not allow us to run sphinx-gallery to update our beautiful examples and it was a huge amount of work to maintain. The documentation is now automatically compiled and updated on merge. We also removed the notebooks from the repository for space reason and also because they are all available in the example gallery. Note that now the output of the documentation build for each commit in the PR is available to check that the doc builds correctly before merging which was not possible with readthedocs.

The CI framework has also been changed with a move from Travis to Github Action which allows to get faster tests on Windows, MacOS and Linux. We also now report our coverage on Codecov.io and we have a reasonable 92% coverage. We also now generate wheels for a number of OS and Python versions at each merge in the master branch. They are available as outputs of this action. This will allow simpler multi-platform releases from now on.

In terms of new features we now have OTDA Classes for unbalanced OT, a new Domain adaptation class form multi domain problems (JCPOT), and several solvers to solve the Partial Optimal Transport problems.

This release is also the moment to thank all the POT contributors (old and new) for helping making POT such a nice toolbox. A lot of changes (also in the API) are comming for the next versions.

Features

New documentation on https://PythonOT.github.io/ (PR #160, PR #143, PR #144)

Documentation build on CircleCI with sphinx-gallery (PR #145,PR #146, #155)

Run sphinx gallery in CI (PR #146)

Remove notebooks from repo because available in doc (PR #156)

Build wheels in CI (#157)

Move from travis to GitHub Action for Windows, MacOS and Linux (PR #148, PR #150)

Partial Optimal Transport (PR#141 and PR #142)

Laplace regularized OTDA (PR #140)

Multi source DA with target shift (PR #137)

Screenkhorn algorithm (PR #121)

Closed issues

Bug in Unbalanced OT example (Issue #127)

Clean Cython output when calling setup.py clean (Issue #122)

Various Macosx compilation problems (Issue #113, Issue #118, PR#130)

EMD dimension mismatch (Issue #114, Fixed in PR #116)

2D barycenter bug for non square images (Issue #124, fixed in PR #132)

Bad value in EMD 1D (Issue #138, fixed in PR #139)

Log bugs for Gromov-Wassertein solver (Issue #107, fixed in PR #108)

Weight issues in barycenter function (PR #106)

Source code(tar.gz)
Source code(zip)
POT-0.7.0-cp35-cp35m-macosx_10_9_x86_64.whl(140.58 KB)
POT-0.7.0-cp35-cp35m-manylinux2010_i686.whl(402.91 KB)
POT-0.7.0-cp35-cp35m-manylinux2010_x86_64.whl(416.43 KB)
POT-0.7.0-cp35-cp35m-win32.whl(129.33 KB)
POT-0.7.0-cp35-cp35m-win_amd64.whl(133.99 KB)
POT-0.7.0-cp36-cp36m-macosx_10_9_x86_64.whl(142.45 KB)
POT-0.7.0-cp36-cp36m-manylinux2010_i686.whl(407.40 KB)
POT-0.7.0-cp36-cp36m-manylinux2010_x86_64.whl(418.77 KB)
POT-0.7.0-cp36-cp36m-win32.whl(130.23 KB)
POT-0.7.0-cp36-cp36m-win_amd64.whl(135.17 KB)
POT-0.7.0-cp37-cp37m-macosx_10_9_x86_64.whl(142.53 KB)
POT-0.7.0-cp37-cp37m-manylinux2010_i686.whl(409.40 KB)
POT-0.7.0-cp37-cp37m-manylinux2010_x86_64.whl(420.62 KB)
POT-0.7.0-cp37-cp37m-win32.whl(130.33 KB)
POT-0.7.0-cp37-cp37m-win_amd64.whl(135.25 KB)
POT-0.7.0-cp38-cp38-macosx_10_9_x86_64.whl(142.79 KB)
POT-0.7.0-cp38-cp38-manylinux2010_i686.whl(431.01 KB)
POT-0.7.0-cp38-cp38-manylinux2010_x86_64.whl(444.95 KB)
POT-0.7.0-cp38-cp38-win32.whl(131.31 KB)
POT-0.7.0-cp38-cp38-win_amd64.whl(135.91 KB)
POT-0.7.0.tar.gz(185.41 KB)
0.7.0-beta0(Apr 23, 2020)

This is a beta test pre release for the new version of POT.

Do not use in production.
Source code(tar.gz)
Source code(zip)
0.6.0(Sep 10, 2019)
This is the first official stable release of POT and this means a jump to 0.6! The library has been used in the wild for a while now and we have reached a state where a lot of fundamental OT solvers are available and tested. It has been quite stable in the last months but kept the beta flag in its Pypi classifiers until now.

Note that this release will be the last one supporting officially Python 2.7 (See https://python3statement.org/ for more reasons). For next release we will keep the travis tests for Python 2 but will make them non necessary for merge in 2020.

The features are never complete in a toolbox designed for solving mathematical problems and research but with the new contributions we now implement algorithms and solvers from 24 scientific papers (listed in the README.md file). New features include a direct implementation of the empirical Sinkhorn divergence, a new efficient (Cython implementation) solver for EMD in 1D and corresponding Wasserstein 1D. We now also have implementations for Unbalanced OT and a solver for Unbalanced OT barycenters. A new variant of Gromov-Wasserstein divergence called Fused Gromov-Wasserstein has been also contributed with exemples of use on structured data and computing barycenters of labeld graphs.

A lot of work has been done on the documentation with several new examples corresponding to the new features and a lot of corrections for the docstrings. But the most visible change is a new quick start guide for POT that gives several pointers about which function or classes allow to solve which specific OT problem. When possible a link is provided to relevant examples.

We will also provide with this release some pre-compiled Python wheels for Linux 64bit on github and pip. This will simplify the install process that before required a C compiler and numpy/cython already installed.

Finally we would like to acknowledge and thank the numerous contributors of POT that has helped in the past build the foundation and are still contributing to bring new features and solvers to the library.

Features

Add compiled manylinux 64bits wheels to pip releases (PR #91)

Add quick start guide (PR #88)

Make doctest work on travis (PR #90)

Update documentation (PR #79, PR #84)

Solver for EMD in 1D (PR #89)

Solvers for regularized unbalanced OT (PR #87, PR#99)

Solver for Fused Gromov-Wasserstein (PR #86)

Add empirical Sinkhorn and empirical Sinkhorn divergences (PR #80)

Closed issues

Issue #59 fail when using "pip install POT" (new details in doc+ hopefully wheels)

Issue #85 Cannot run gpu modules

Issue #75 Greenkhorn do not return log (solved in PR #76)

Issue #82 Gromov-Wasserstein fails when the cost matrices are slightly different

Issue #72 Macosx build problem

Source code(tar.gz)
Source code(zip)
POT-0.6.0-cp27-cp27m-manylinux1_i686.whl(298.91 KB)
POT-0.6.0-cp27-cp27m-manylinux1_x86_64.whl(309.41 KB)
POT-0.6.0-cp27-cp27mu-manylinux1_i686.whl(298.87 KB)
POT-0.6.0-cp27-cp27mu-manylinux1_x86_64.whl(309.40 KB)
POT-0.6.0-cp34-cp34m-manylinux1_i686.whl(283.37 KB)
POT-0.6.0-cp34-cp34m-manylinux1_x86_64.whl(292.62 KB)
POT-0.6.0-cp35-cp35m-manylinux1_i686.whl(284.25 KB)
POT-0.6.0-cp35-cp35m-manylinux1_x86_64.whl(295.24 KB)
POT-0.6.0-cp36-cp36m-manylinux1_i686.whl(287.62 KB)
POT-0.6.0-cp36-cp36m-manylinux1_x86_64.whl(298.46 KB)
POT-0.6.0-cp37-cp37m-manylinux1_i686.whl(288.12 KB)
POT-0.6.0-cp37-cp37m-manylinux1_x86_64.whl(298.64 KB)
0.5.0(Oct 3, 2018)
POT is 2 years old! This release brings numerous new features to the toolbox as listed below but also several bug correction.

Among the new features, we can highlight a non-regularized Gromov-Wasserstein solver, a new greedy variant of sinkhorn, non-regularized, convolutional (2D) and free support Wasserstein barycenters and smooth and stochastic implementation of entropic OT.

POT 0.5 also comes with a rewriting of ot.gpu using the cupy framework instead of the unmaintained cudamat. Note that while we tried to keed changes to the minimum, the OTDA classes were deprecated. If you are happy with the cudamat implementation, we recommend you stay with stable release 0.4 for now.

The code quality has also improved with 92% code coverage in tests that is now printed to the log in the Travis builds. The documentation has also been greatly improved with new modules and examples/notebooks.

This new release is so full of new stuff and corrections thanks to the old and new POT contributors (you can see the list in the readme).

Features

Add non regularized Gromov-Wasserstein solver (PR #41)

Linear OT mapping between empirical distributions and 90% test coverage (PR #42)

Add log parameter in class EMDTransport and SinkhornLpL1Transport (PR #44)

Add Markdown format for Pipy (PR #45)

Test for Python 3.5 and 3.6 on Travis (PR #46)

Non regularized Wasserstein barycenter with scipy linear solver and/or cvxopt (PR #47)

Rename dataset functions to be more sklearn compliant (PR #49)

Smooth and sparse Optimal transport implementation with entropic and quadratic regularization (PR #50)

Stochastic OT in the dual and semi-dual (PR #52 and PR #62)

Free support barycenters (PR #56)

Speed-up Sinkhorn function (PR #57 and PR #58)

Add convolutional Wassersein barycenters for 2D images (PR #64)

Add Greedy Sinkhorn variant (Greenkhorn) (PR #66)

Big ot.gpu update with cupy implementation (instead of un-maintained cudamat) (PR #67)

Deprecation

Deprecated OTDA Classes were removed from ot.da and ot.gpu for version 0.5 (PR #48 and PR #67). The deprecation message has been for a year here since 0.4 and it is time to pull the plug.

Closed issues

Issue #35 : remove import plot from ot/__init__.py (See PR #41)

Issue #43 : Unusable parameter log for EMDTransport (See PR #44)

Issue #55 : UnicodeDecodeError: 'ascii' while installing with pip

Source code(tar.gz)
Source code(zip)
0.4.0(Sep 20, 2017)

Source code(tar.gz)
Source code(zip)
0.4(Sep 15, 2017)
This release contains a lot of contribution from new contributors.

Features

Automatic notebooks and doc update (PR #27)

Add gromov Wasserstein solver and Gromov Barycenters (PR #23)

emd and emd2 can now return dual variables and have max_iter (PR #29 and PR #25)

New domain adaptation classes compatible with scikit-learn (PR #22)

Proper tests with pytest on travis (PR #19)

PEP 8 tests (PR #13)

Closed issues

emd convergence problem du to fixed max iterations (#24)

Semi supervised DA error (#26)

Source code(tar.gz)
Source code(zip)
0.3.1(Jul 11, 2017)
correct bug in EMD on windows

Source code(tar.gz)
Source code(zip)
0.3(Jul 7, 2017)
emd* and sinkhorn* are now performed in parallel for multiple target distributions

emd and sinkhorn are for OT matrix computation

emd2 and sinkhorn2 are for OT loss computation

new notebooks for emd computation and Wasserstein Discriminant Analysis

relocate notebooks

update documentation

clean_zeros(a,b,M) for removimg zeros in sparse distributions

GPU implementations for sinkhorn and group lasso regularization

Source code(tar.gz)
Source code(zip)
V0.2(Apr 7, 2017)
New dimensionality reduction method (WDA)

Efficient method emd2 returns only tarnsport (in paralell if several histograms given)

Source code(tar.gz)
Source code(zip)
V0.1.12(Jan 5, 2017)
Add sphinx gallery for better documentation

Small efficiency tweak in sinkhorn

Add simple tic() toc() functions for timing

Source code(tar.gz)
Source code(zip)
V0.1.11(Nov 7, 2016)
correct important bug introduced in last release

Source code(tar.gz)
Source code(zip)
V0.1.10(Nov 7, 2016)
numerical stabilization for sinkhorn (log domain and epsilon scaling)

Source code(tar.gz)
Source code(zip)
V0.1.9.1(Nov 4, 2016)

Correct comment error
Source code(tar.gz)
Source code(zip)
V0.1.9(Nov 4, 2016)
Add classes and examples for domain adaptation

Joint OT matrix and mapping estimation

Source code(tar.gz)
Source code(zip)
V0.1.7(Oct 31, 2016)

Source code(tar.gz)
Source code(zip)
V0.1.6(Oct 31, 2016)

Source code(tar.gz)
Source code(zip)
V0.1.4(Oct 28, 2016)

Source code(tar.gz)
Source code(zip)
V0.1.3(Oct 28, 2016)

Source code(tar.gz)
Source code(zip)
V0.1(Oct 28, 2016)

Source code(tar.gz)
Source code(zip)

POT : Python Optimal Transport

Related tags

Overview

POT: Python Optimal Transport

Using and citing the toolbox

Installation

Pip installation

Anaconda installation with conda-forge

Post installation check

Dependencies

Examples

Short examples

Examples and Notebooks

Acknowledgements

Contributions and code of conduct

Support

References

Comments

Types of changes

Motivation and context / Related issue

How has this been tested (if it applies)

Checklist

Types of changes

Motivation and context / Related issue

How has this been tested (if it applies)

Checklist

Datasets used:

** To Reproduce**

Result and Error

Environment (please complete the following information):

Types of changes

Motivation and context / Related issue

How has this been tested (if it applies)

PR checklist

Types of changes

Motivation and context / Related issue

How has this been tested (if it applies)

PR checklist

Releases(0.8.2)

0.8.2(Apr 21, 2022)

New features

Closed issues

0.8.1.0(Dec 31, 2021)

Closed issues

0.8.1(Dec 27, 2021)

New features

Closed issues

0.8.0(Nov 5, 2021)

New features

Closed issues

0.7.0(May 5, 2020)

Features

Closed issues

0.7.0-beta0(Apr 23, 2020)

0.6.0(Sep 10, 2019)

Features

Closed issues

0.5.0(Oct 3, 2018)

Features

Deprecation

Closed issues

0.4.0(Sep 20, 2017)

0.4(Sep 15, 2017)

Features

Closed issues

0.3.1(Jul 11, 2017)

0.3(Jul 7, 2017)

V0.2(Apr 7, 2017)

V0.1.12(Jan 5, 2017)

V0.1.11(Nov 7, 2016)

V0.1.10(Nov 7, 2016)

V0.1.9.1(Nov 4, 2016)

V0.1.9(Nov 4, 2016)

V0.1.7(Oct 31, 2016)

V0.1.6(Oct 31, 2016)

V0.1.4(Oct 28, 2016)

V0.1.3(Oct 28, 2016)

V0.1(Oct 28, 2016)

Owner

Python Optimal Transport

To Reproduce