Sinkhorn l1lp transport does not return "log" details

[sayali7]

Describe the bug

when you initialize and run ot.da.SinkhornLpl1Transport() with log=True, and then the fit() function, I get an error that says "ValueError: too many values to unpack (expected 2)".

To Reproduce

Steps to reproduce the behavior:
I ran the following two lines of code:
1.ot_lpl1 = ot.da.SinkhornLpl1Transport(reg_e=1e02,reg_cl=1e-2,log=True,verbose=True)
2. ot_lpl1.fit(Xs=Xs, ys=ys, Xt=Xt)

Error Message:

---

ValueError Traceback (most recent call last)
/tmp/ipykernel_176879/2895611854.py in
1 # Sinkhorn Transport with Group lasso regularization
2 ot_lpl1 = ot.da.SinkhornLpl1Transport(reg_e=1e02,reg_cl=1e-2,log=True,verbose=True)
----> 3 ot_lpl1.fit(Xs=Xs, ys=ys, Xt=Xt)
4 transp_Xs_lpl1 = ot_lpl1.transform(Xs=Xs)
5 pd.DataFrame(transp_Xs_lpl1).head()

~/.local/lib/python3.8/site-packages/ot/da.py in fit(self, Xs, ys, Xt, yt)
1748 # deal with the value of log
1749 if self.log:
-> 1750 self.coupling_, self.log_ = returned_
1751 else:
1752 self.coupling_ = returned_

ValueError: too many values to unpack (expected 2)

Code sample

ot_lpl1 = ot.da.SinkhornLpl1Transport(reg_e=1e02,reg_cl=1e-2,log=True,verbose=True)
ot_lpl1.fit(Xs=Xs, ys=ys, Xt=Xt)

Expected behavior

The fit() function should return an object containing both "coupling" and "log" details.

Environment (please complete the following information):

• OS (e.g. MacOS, Windows, Linux): Ubuntu 20
• Python version: 3.8
• How was POT installed (source, pip, conda): pip

Output of the following code snippet:

import platform; print(platform.platform())
import sys; print("Python", sys.version)
import numpy; print("NumPy", numpy.__version__)
import scipy; print("SciPy", scipy.__version__)
import ot; print("POT", ot.__version__)

Linux-5.4.0-132-generic-x86_64-with-glibc2.29
Python 3.8.10 (default, Jun 22 2022, 20:18:18)
[GCC 9.4.0]
NumPy 1.21.4
SciPy 1.8.0
POT 0.8.1.0

Additional context

[rflamary]

Good catch!

I fixed it in PR #413 that shoud be merged soon. You can use branch bug_log_lpl1 in the meantime.