Cours de Python

Le cours « Analyse scientifique en python » (L3/M1/M2) est disponible en ligne. Les suggestions et commentaires sont les bienvenus.

Taylor diagram

A Taylor diagram is a plot that shows the similarities of different models with respect to a given data set (correlation and RMS). It seems to be mostly used for climate models, but I've never seen it in action in astronomy fields.

I made a python/numpy/matplotlib implementation, which was then used by Michael Rawlins (from Univ. Massachusetts) to produce plots to be published in Rawlins et al., 2012. Assessment of regional climate model simulation estimates over the Northeast U.S., Journal of Geophysical Research (in review).

N.B.: do not trust this illustration for (geo)scientific purposes, but refer directly to the published version of the plot if needed.

Search for a unicode character.

If you're fed up with opening the so-called gucharmap (or any other equivalent) to look for a specific unicode character, you might be interested in this python script based on the standard unicodedata module.

$ searchUnicode.py letter lamda
Λ GREEK CAPITAL LETTER LAMDA &#923 0x39b
λ GREEK SMALL LETTER LAMDA &#955 0x3bb
ᴧ GREEK LETTER SMALL CAPITAL LAMDA &#7463 0x1d27

$ searchUnicode.py "letter lamda"
Λ GREEK CAPITAL LETTER LAMDA &#923 0x39b
λ GREEK SMALL LETTER LAMDA &#955 0x3bb

Encode intensity vector in an HSL image

I wanted to mimic emission and absorption spectra of gas lamps. One solution is to encode the spectrum intensity into a specific color component of a simple "rainbow" image, e.g.:

• the luminance component of the HLS color system for an emission spectrum,
• the value component of the HSV color system for a transmission spectrum.

Loading ....

The following Gist is an example in python/numpy/matplotlib, using standard module colorsys for color system conversion.

Les glaces de Nardone

Les glaces de chez Nardone:

• Nutella 4/5
• Noisette 4/5
• Praline rose 1/5
• Rhubarbe 2/5
• Mangue 4/5
• Menthe 3/5
• Chocolat blanc 3/5
• Pêche 5/5
• Ananas 5/5

Colloquium SNfactory 2010 (en)

The webcast of the colloquium I gave at APC on 2010/10/19 about "The Nearby Supernova Factory". The presentation is in english (kind of...), the slides are in french though.

Une comédie sans prétention pour qui compte au moins un Suisse parmis ses amis !
Bonus: ABAM

Largest packages on a Ubuntu machine

When in need for space on your Ubuntu (or Debian-based) machine, you can list the largest installed packages with:

dpkg-query -Wf '${Installed-Size}\t${Package}\n' | sort -n

Tilt-shift photography

Première tentative (pas encore optimale...) de « Tilt shift/Miniature Faking » digital. Prise de vue : près de Séguret (Vaucluse). Traitement : Gimp + Focus Blur.

Lien bonus : Bathtub IV

DOA - Citoyens clandestins

Une plongée dans les services secrets français, espionnant une bande d'islamistes radicaux, mais surtout se surveillant les uns les autres! Un très bon pavé de plus de 700 pages, avec un style très percutant.

Manoir de Lan Kerellec

Un restaurant Relais & Chateaux avec une vue magnifique sur l'ile Milliau, cuisine à tendance moléculaire hypo-calorique...
Menu Clin d'oeil, fixe :

• Amuse-bouche (oeuf de caille poché sur gélée de concombre et citron vert)
• Araignée décortiquée, fois gras de canard et émulsion de jus de pomme verte
• Homard bleu poêlé, transparence carotte/mangue au Quinoa, jus coraillé
• Cigare, crème légère Cognac, bananes safranées et glace au rhum

Le tout acompagné d'un Savennière Roche aux Moines, vin blanc sec d'Anjou très aromatique, presque fumé.

PaperScope

"PaperScope is a tool for graphically exploring the Astrophysics Data System (ADS)" (under Java).

Telluric correction

• Matheson et al. 2000 (see appendix)
• NOAO (see README)
• Wade et Horne 1998

Deconvolution

• 1998ApJ...494..472M: Magain, P.; Courbin, F.; Sohy, S., Deconvolution with Correct Sampling (site, pgm)
• 2000ApJ...529.1136C: Courbin, F.; Magain, P.; Kirkove, M.; Sohy, S., A Method for Spatial Deconvolution of Spectra
• 2007A&A...461..373M: P. Magain, F. Courbin, M. Gillon, S. Sohy, G. Letawe, V. Chantry and Y. Letawe, A deconvolution-based algorithm for crowded field photometry with unknown Point Spread Function
• 2003AJ....125.2266L: Lucy, L. B., Walsh, J. R., Iterative Techniques for the Decomposition of Long-Slit Spectra
• 2005A&A...439..401G: Giovannelli, J.-F., Coulais, A., Positive deconvolution for superimposed extended source and point sources
• 2005ApJ...627..701M: McNeil, Stephen R.; Moody, J. Ward, Photometric Accuracy of Deconvolved Simulated Astronomical Images
• 2005A&A...436..373P: Puschmann, K. G.; Kneer, F., On super-resolution in astronomical imaging
• 2002PASP..114.1051S: Starck, J. L.; Pantin, E.; Murtagh, F., Deconvolution in Astronomy: A Review
• 2000A&AS..147..139S: Starck, J.-L.; Bijaoui, A.; Valtchanov, I.; Murtagh, F., A combined approach for object detection and deconvolution
• 1999MNRAS.307..659P: Pijpers, F. P., Unbiased image reconstruction as an inverse problem

Atmospheric extinction

• Introduction
• CFHT Observer's Manual
• 1975ApJ...197..593H: Hayes, D. S., Latham, D. W.,A rediscussion of the atmospheric extinction and the absolute spectral-energy distribution of VEGA
• 1992A&A...266..619S: Sterken, C., Manfroid, J., The evolution of atmospheric extinction at La Silla derived from measurements in the Stromgren photometric system
• ELT Site evaluation WG

Optimal extraction algorithms

• 1998MNRAS.296..339N: Naylor, Tim, An optimal extraction algorithm for imaging photometry
  
• 1990A&A...240..216V: Verschueren, W.; Hensberge, H., Order extraction and background subtraction on CCD CASPEC echelle spectra
• 1990PASP..102..183M: Mukai, Koji, Optimal extraction of cross-dispersed spectra
• 1989PASP..101.1032M: Marsh, T. R., The extraction of highly distorted spectra
• 1986PASP...98..609H: Horne, K., An optimal extraction algorithm for CCD spectroscopy
• 1986PASP...98.1220R: Robertson, J. G., Optimal extraction of single-object spectra from observations with two-dimensional detectors

Sea-level time

OK, things are not working properly yet, but still it's sea-level time :) The red-channel is now commisionned, after the correction of a small tilt of the micro-lens array. But there's still plenty of things to do for the coming nights: of course, the guiding and acquisition, the astrometry of the photometric and spectrographic channels, the confocality and so on. On the software side, the online seems to be late... But the 3D data-reduction is mostly ready, here is my first reconstructed image of a star (obviously out of the field-of-view!):

First reconstructed image from SNIFS!

Not very impressive, uh, but I am nonetheless a happy man!

Snifs is completed

Snifs is now completed... but unfortunately not running yet. There are still some optical alignements to be fine tuned, the infamous Cam I is again off, some PCs are a bit sick, the control soft is shakky, etc. But everything is under control :)

So here is the long awaited picture of Snifs on the telescope (at zenith):

OK, not as beautiful as Sauron, but still fine...

First image of the multi-filter!

Tonight is the imaging night! Richard is at the summit, seconded by Greg at Hilo, and they're playing with the photometric channel to get the guiding working.

And here is the first (dome) image of the multi-filter:

Congrats!

Here I am!

After two days of flights, stops at Frankfurt, Charlotte NC, Los Angeles LA, here I am on Big Island, right in UH dome, looking at the mechanicians plugging some new unexpected devices to the telescope... I must admit 12h jet-lag and 4200m altitude doesn't help for the understanding of all the details.

Not sure we'll have access to the telescope in the coming hours, and not sure I'll be able to do anything except snoring after that. But Manu(s) and Blandine are ready to jump on the first exposure.

A little picture (note the shadow of the Mauna Kea on the clouds):

The same, but for Arlette: