You are currently browsing the category archive for the ‘Uncategorized’ category.
It was announced earlier today that Google and NASA have formed a joint lab called “Quantum Artificial Intelligence Lab” to research machine learning applications of quantum computers provided by D-Wave. From the press release:
Researchers at Google, NASA and USRA expect to use the D-Wave system to develop applications for a broad range of complex problems such as machine learning, web search, speech recognition, planning and scheduling, search for exoplanets, and support operations in mission control centers.
Although Lockheed Martin recently decided to upgrade their D-Wave system (after testing an earlier version for 2 years), there is still some skepticism from Quantum Computing experts (for example, claims that the D-Wave implementation does not provide actual speedup compared to classical computing). In any case, exciting stuff…
Paris Rive Gauche is the neighbourhood of Paris surrounding the National Library of France (BnF). The district is on the left bank of the Seine river and is bordered by the railway tracks of Gare d’Austerlitz and the Périphérique.
Above, buildings of Paris Diderot University.
Above, “Monochrome for Paris” by Nancy Rubins.
Above, one of the buildings of the BnF.
Above, Les Rochers dans le Ciel (“The Rocks in the Sky”) by Didier Marcel.
It might be overkill but is clearly more refined than the old-school method:
The “GIS and Agent-Based Modeling” blog reports on GAMA, an open source GIS-ABM tool that looks quite interesting:
GAMA is a simulation platform, which aims at providing field experts, modelers, and computer scientists with a complete modeling and simulation development environment for building spatially explicit agent-based simulations. It is being developed by several French and Vietnamese research teams under the umbrella of the IRD/UPMC International Research Unit UMMISCO since 2007.
GAMA provides the modeler with:
- The ability to use arbitrarily complex GIS data as environments for the agents.
- The possibility to run simulations composed of vast numbers of agents (up to millions).
- A way to conduct automated controlled experiments on various scenarios, with a systematic, guided or “intelligent” exploration of the space of parameters of models.
- The possibility to let users interact with the agents in the course of the simulations.
A demo video is also available.
On the Esri side, check out the (free) Agent Analyst extension.
Above, the church of Saint-Jean-de-Montierneuf.
Above inside the church of Notre-Dame-la-Grande.
Above, view of Poitiers from Notre-Dame-des-Dunes.
Above, the empty parking lot of Futuroscope (closed for the winter).
Above, boats in the morning fog in La Rochelle.
Above, city hall of La Rochelle.
Above, watching sharks at the Aquarium.
Above, flying kite on a beach near Port des Minimes.
Above, Tour de la Chaine and Tour de la Lanterne, seen from the top of Tour Saint-Nicolas.
Above, plage de la Concurrence, before sunset.
I got a Kobo Glo eBook reader some time ago. I find it quite awesome: With numerous public libraries in Paris and this device (together with sources of free books like FeedBooks, Baen Free Library or less authoritative sources), it is hard to think I will ever buy another physical book again.
I was recently looking for a way to convert .mobi files to .epub: The Kobo supports .mobi ebooks natively but the support is not as good as with .epub. It turns out Calibre allows just that: It is an open source e-book application for managing a library of ebooks and it does a lot of stuff. It is possible to use the GUI to perform the conversion but Calibre also provides some command-line tools to perform (among other things) conversions in batch. Here is an example of a Python script that uses the ebook-convert tool to convert all .mobi files in a directory to .epub:
# -*- coding: utf-8 *-* import sys, os, subprocess _, dirPath = sys.argv convertedDirPath = os.path.join(dirPath, "__converted") if not os.path.exists(convertedDirPath): os.makedirs(convertedDirPath) for fileName in os.listdir(dirPath): if fileName.endswith(".mobi"): rootFileName, _ = os.path.splitext(fileName) epubFileName = rootFileName + ".epub" mobiFilePath = os.path.join(dirPath, fileName) convertedMobiFilePath = os.path.join(convertedDirPath, fileName) epubFilePath = os.path.join(dirPath, epubFileName) conversionResult = subprocess.call(["ebook-convert", mobiFilePath, epubFilePath]) if conversionResult == 0: os.rename(mobiFilePath, convertedMobiFilePath)
It is then possible to copy the converted files directly to the eReader after plugging it into the computer.
One more nice thing with the Kobo is that it stores its settings in a SQLite database on the device. Therefore there are many tools available to manipulate the content. For example, I saw this Python library that can do just that (although I haven’t tested yet with my version of Kobo).
The photo above was taken on New Year’s eve inside the Madeleine Church. I had walked by the building a few times but I had never realized it was a church (since it looks like a Greek temple) so I entered to check it out. Inside, there was this modern-style nativity scene depicting Mary and Joseph watching a picture of baby Jesus on a flat screen TV. The faces of the white mannequins (an army of angels apparently) also change from time to time. Very strange and slightly creepy… It was designed by Gaëtan Duthu, a young French designer.