alexandre.gramfort.net

fa.bianp.net

I say things - SciPy

http://fa.bianp.net/tag/scipy.html

Numerical optimizers for Logistic Regression. Mon 20 May 2013. In this post I compar several implementations of Logistic Regression. The task was to implement a Logistic Regression model using standard optimization tools from. And compare them against state of the art implementations such as LIBLINEAR. In this blog post I'll write down all the implementation details of this model, in the hope that not only the conclusions but also the process would be useful for future comparisons and benchmarks. Begin{a...

fa.bianp.net

Publications

http://fa.bianp.net/pages/publications.html

Here are my most recent publications. A similar list can also be found in my Google Scholar profile. On the convergence rate of the three operator splitting scheme. Word meaning in the ventral visual path: a perceptial to conceptual gradient of semantic coding.". Borghesani, Valentina, Fabian Pedregosa, Marco Buiatti, Alexis Amadon, Evelyn Eger, and Manuela Piazza. NeuroImage (2016) HTTP. Asaga: Asynchronous Parallel Saga. Rémi Leblond, Fabian Pedregosa, Simon Lacoste-Julien, preprint, ArXiv. Borghesani,...

fa.bianp.net

I say things - NumPy

http://fa.bianp.net/tag/numpy.html

PyData Paris - April 2015. Tue 07 April 2015. Last Friday was PyData Paris. In words of the organizers, ' a gathering of users and developers of data analysis tools in Python' . The organizers did a great job in putting together and the event started already with a full room for Gael's. My personal highlights were (note that there were two parallel tracks). Talk on Cleaning Confused Collections of Characters. Many tips and many pointers to libraries. In particular, I discovered fixes text for you. Static...

fa.bianp.net

I say things - Job

http://fa.bianp.net/tag/job.html

Job Offer: data scientist in Paris. Fri 10 July 2015. And I are looking looking for a data scientist to work on large-scale non-smooth optimization methods and other topics. You can find more information in this link. The job description is intentionally vague, and depending on the candidate this can be a postdoc-like job with most of their time devoted to research or an engineering-like job with more emphasis on coding tasks and contributing to open source projects. Page 1 / 1.

fa.bianp.net

I say things - ranking

http://fa.bianp.net/tag/ranking.html

Thu 02 May 2013. TL;DR: I've implemented a logistic ordinal regression or proportional odds model. Here is the Python code. Model, also known as the proportional odds was introduced in the early 80s by McCullagh [. And is a generalized linear model specially tailored for the case of predicting ordinal variables, that is, variables that are discrete (as in classification) but which can be ordered (as in regression). It can be seen as an extension of the logistic regression model to the ordinal setting.

fa.bianp.net

I say things - machine learning

http://fa.bianp.net/tag/machine-learning.html

A fast, fully asynchronous variant of the SAGA algorithm. Wed 12 October 2016. My friend Rémi Leblond. Has recently uploaded to ArXiv our preprint on an asynchronous version of the SAGA optimization algorithm. The main contribution is to develop a parallel (fully asynchronous, no locks) variant of the SAGA algorighm. The core of the asynchronous algorithm is similar to Hogwild! The speedups obtained versus the sequential version are quite impressive. For example, we have observed to commonly obtain 5...

fa.bianp.net

I say things - Python

http://fa.bianp.net/tag/python.html

Fri 25 March 2016. Announce: first public release of lightning. A library for large-scale linear classification, regression and ranking in Python. The library was started a couple of years ago by Mathieu Blondel. Who also contributed the vast majority of source code. I joined recently its development and decided it was about time for a v0.1! Prebuild conda packages are available for all operating systems (god thank appveyor). More information on lightning's website. Sun 06 March 2016. Mon 22 February 2016.

fa.bianp.net

I say things - consistency

http://fa.bianp.net/tag/consistency.html

On the consistency of ordinal regression methods. Fri 09 October 2015. My latests work (with Francis Bach. Is on the consistency of ordinal regression methods. It has the wildly imaginative title of "On the Consistency of Ordinal Regression Methods" and is currently under review but you can read the draft of it on ArXiv. If you have any thoughts about it, please leave me a comment! Having Fisher consistency is an important property that "allows us to design good loss functions with desirable properties"[.

fa.bianp.net

I say things - Jupyter

http://fa.bianp.net/tag/jupyter.html

Tue 21 April 2015. TL;DR I created a gallery for IPython/Jupyter notebooks. Check it out :-). A couple of months ago I put online a website that displays a collection of IPython/Jupyter notebooks. The is a website that collects user-submitted and publicly available notebooks and displays them with a nice screenshot. The great thing about this website compared to other. I would like to make a website where it is possible to. Under the hood there's a django app, for which the source code lives here. Don't ...

fa.bianp.net

I say things - fMRI

http://fa.bianp.net/tag/fmri.html

Data-driven hemodynamic response function estimation. Fri 05 December 2014. My latest research paper. Deals with the estimation of the hemodynamic response function (HRF) from fMRI data. This is an important topic since the knowledge of a hemodynamic response function is what makes it possible to extract the brain activation maps that are used in most of the impressive applications of machine learning to fMRI, such as (but not limited to) the reconstruction of visual images from brain activity [.