Crea un profiloAccedi

Ricerche recenti

Nessuna ricerca recente

Opzioni di ricerca

Disponibile solo dopo aver effettuato l'accesso.
mastodon.uno è uno dei tanti server Mastodon indipendenti che puoi usare per partecipare al fediverso.
Mastodon.Uno è la principale comunità mastodon italiana. Con 77.000 iscritti è il più grande nodo Mastodon italiano: anima ambientalista a supporto della privacy e del mondo Open Source.

Amministrato da:

Statistiche del server:

6,2K
utenti attivi

mastodon.uno: InfoStatoCartella dei profiliPolitica sulla privacy

Mastodon: InfoScarica l'appScorciatoie da tastieraVisualizza il codice sorgentev4.3.7

#datavisualisation

1 post1 partecipante0 post oggi
Pubblico
It's Richie

Is there a way to graph trending hashtags on mastodon/the Fediverse?

What I'm interested in is to graph say my local city of #Adelaide and use that as a bit of an indicator of if #Mastodon usage and engagemement is growing locally in my town

And yes im interested in data across the fediverse or at least across a few selected servers, as I want data beyond just a single instance.

I can see this on my desktop web interface right now.. so Im guessing there *should* be a way.

Image of the 'Trending now' section on a mastodon instance feed page. Shows the hashtage, number of people in last 2 days and a little mini graph.
#GrowingMastodon#DataVisualisation#TrendingTopics
Pubblico
d'archives et d'ailleurs
Des sources historiques sur des #femmes pendant les révolutions de 1789, 1830 (surtout) et 1848 !
A défaut d'avoir sous la main des photographies de documents originaux, une #datavisualisation du jeu de données en #opendata

Source : Archives nationales, jeu de données "Pensions et secours aux combattants et victimes des révolutions de 1789, 1830 et 1848" sur data.culture : https://data.culture.gouv.fr/explore/dataset/pensions-et-secours-aux-combattants-et-victimes-des-revolutions

#archives #1jour1archive #archivesnationales #journéeinternationalesdesdroitsdesfemmes #herstory #19esiècle #Révolutionfrançaise #TroisGlorieuses #Révolution1848 #womeninhistory #dataviz #sourceshistoriques
Visualisation des 175 dossiers de pensions ou secours concernant des femmes conservés aux Archives nationales, par année d'événement (révolution de 1789, de 1830 ou 1848)
#sourceshistoriques
Pubblico
Starbeamrainbowlabs

While I try to find time to write a PhD update blog post, here's a Cool Graph I plotted recently for my rainfall radar model.

One of the things I have been asked to do in my PhD corrections is to assess the model's stability. To this end, I trained a series of models at different learning rates, and this is the result.

Now I just need to figure out why the output prediction looks like it has a case of chickenpox.....

A Cool Graph showing validation accuracy over training for my rainfall radar model. Unsurprisingly, a lower learning rate (lr=0.00001) results in higher and more stable validation accuracy than a higher learning rate (lr=0.01). Particularly higher learning rates are very variable in accuracy; stretching from 0.3 - 0.8. Lower learning rates are ~0.8 - ~0.95.
#PhD#PhDThesis#Corrections
Pubblico
Laetitia Bracco

🎥 Le webinaire sur la datavisualisation, organisé par le GTSO #Données de Couperin, est désormais disponible en replay !

📅 Tenu le 17 décembre 2024, ce webinaire explore l’art de transformer des données complexes en récits visuels clairs, fonctionnels et esthétiques. Mais attention : une belle image ne suffit pas, elle doit aussi rendre les informations compréhensibles et pertinentes.

Ce webinaire propose :
✨ Des concepts et outils issus de différents univers disciplinaires.
✨ Des exemples concrets de projets et solutions pratiques.
✨ Une réflexion sur la place de l’image comme vecteur d’information.

👩‍🏫 Intervenants :
Jean-Daniel Fekete (Inria Paris-Saclay)
Étienne Côme (Université Gustave Eiffel)

🎥 Découvrez le replay et les supports sur notre nouvelle chaîne CanalU :canal-u.tv/chaines/couperin/de

Canal-UDerrière la beauté des images, des données | Canal ULa datavisualisation, ou visualisation des données, consiste à mettre en scène les données graphiquement pour obtenir une compréhension instantanée, simple, fonctionnelle et esthétique de phénomènes complexes ou ayant nécessité une étape préalable de fouille. Née au sein de la recherche en mathématique statistique et en informatique, elle s’est depuis appliquée à la recherche en sciences sociales.
#scienceouverte#datavisualisation#dataviz
Pubblico
Datasaurus
2024 first year to pass 1.5ºC global warming limit

The #BBC has produced this news item based on Copernicus data that highlights the following findings about #climatechange:

1. Global average temperatures for 2024 were around 1.6C above those of the pre-industrial period (1850-1900). This is used as a baseline, as it was the time before humans started burning large amounts of fossil fuels.

2. Whereas temperature increase from baseline was mainly located in specific areas by early 2000s, this is now observed everywhere almost equally, including oceans.

3. 2024 was the year with more number of days of temperature exceeding the average, and where the excess was more acute.

Source: https://www.bbc.co.uk/news/articles/cd7575x8yq5o


#datavisualization #datavisualisation #dataviz #datavis
Barplot showing temperature difference by year in relation to pre-industrial average (1850-1900). 2024 is the first to surpass 1.5ºC diff.
Heatmap maps of the world since 1940 to 2024, showing temperature differences. In 2024 everything is almost red.
Ridgeplot of temperature differences since 1940 to 2024. 2024 is the year with more days warmer than average
#datavis
Pubblico *
Sarah Lea

Everyone knows ggplot2 in R or matplotlib in Python. But what are the differences between theese 2 programming languages and what are the most important R & Python libraries for visualization? :blobcoffee:

towardsdatascience.com/the-ess

Towards Data Science · The Essential Guide to R and Python Libraries for Data VisualizationDi Sarah Lea
#data#datascience#dataviz
EsploraFeed dal vivo
Info