metric-learn

metric-learn copied to clipboard

SCML: Add warm_start parameter

7

Because [SCML](http://researchers.lille.inria.fr/abellet/papers/aaai14.pdf) optimization procedure is based on stochastic subgradient descent, we can save the weights after fitting the model, and use them in a following fit call (with a different...

maxi-marufo

Addresses my request in #210. 1. Introduces a _BaseLMNN/ LMNN class to operate on triplets, s.t. `d(triplets[i, 0],triplets[i, 1]) < d(triplets[i, 0], triplets[i, 2])` - the same setup as SCML....

zdk123

refactor lmnn code

1

#### Description LMNN's code could be refactored to be more simple to read (see comment https://github.com/metric-learn/metric-learn/pull/201#issuecomment-496948016) see also https://github.com/metric-learn/metric-learn/pull/201#discussion_r286032898

wdevazelhes

[DOC] [WIP] Developers documentation page

1

@bellet @perimosocordiae @terrytangyuan @wdevazelhes Motivated by #259 , I made this docs for new developers like a guide in _How to_ contribute to the package. I followed the scikit-learn guideline...

mvargas33

2. [MRG] Bilinear similarity

17

Hi! In the aim of implementing [OASIS](https://www.jmlr.org/papers/volume11/chechik10a/chechik10a.pdf) algorithm, I open this PR to discuss the implementation of the BilinearMixin class. Which is a requirement for the algorithm. I've made some...

mvargas33

3. [WIP] OASIS algorithm implementation

1

Hi! I am currently implementing the [OASIS](https://www.jmlr.org/papers/volume11/chechik10a/chechik10a.pdf) algorithm and I open this PR to make the implementation transparent while working on it. Any discussion, question or comments is very welcomed....

mvargas33

I created a model selection example for supervised Mahalanobis learners, to show the effectiveness of the linear transformation. I use a "large" dataset from sklearn: Labeled Faces in the Wild...

mvargas33

New LMNN version

33

Hi all, this is a PR to replace LMNN. I have submitted this version also to scikit-learn some time ago (see https://github.com/scikit-learn/scikit-learn/pull/8602) and @bellet suggested to submit here too. I...

johny-c

It would be nice to have an example in the doc which demonstrates how to calibrate the pairwise metric learners with respect to several scores as introduced in #168, as...

bellet

[DOC] RCA should maybe be in "Supervised algorithms"

2

Currently `RCA` is in Pairs Learners in the doc, where it shouldn't. It should probably be in "Supervised"

wdevazelhes