#MAC #TríTuệNhiệt #Thunderbolt #GộpServer Clustering Mac Mini (Thunderbolt 4) có thể chậm 10x nhưng Thunderbolt 5+RDMA hứa hẹn tăng tốc 2x. Tối ưu hóa cluster phi chính thức (non-RDMA) đang được phát triển, phù hợp cho người sở hữu Mac base model. Tìm hiểu phần mềm mới & kiến thức về phân phối mô hình LLM. #CôngNghệ #AI #Optimize

*(#MAC #AI #Thunderbolt #Clustering Clustering Thunderbolt-4 Macs may incur 10x slowdown, but Thunderbolt-5 RDMA clusters boost speed 2x. Unofficial optimizations for

Data-driven empathy is the future of UX. Moving beyond the limitations of subjective persona creation, we now utilize clustering algorithms to uncover latent user patterns. This methodological shift ensures design decisions are rooted in empirical evidence rather than intuition. This guide at WebHeads United explores the technical frameworks for this synthesis. Refine your audience insights with algorithmic rigour.

https://webheadsunited.com/using-clustering-algorithms-persona-creation/

#DataScience #UXDesign #Clustering #WebDev #Fediverse

Showing something is NOT the case is much harder than showing it is, but Nishant painstakingly went through every possible analysis loophole to show this. Well done Nishant! For everyone interested in #clustering, #heterogeneity, #classification and #modelling!

"Normalized mutual information is a biased measure for classification and community detection"

#algorithm #classification #clustering #ML #bias #metric

https://www.nature.com/articles/s41467-025-66150-8

#fedihelp
Suche jemanden, der Erfahrung mit #UMAP und k-Means hat #clustering.
Geht darum ob mein Workflow so stichhaltig ist
#statistik.
Bin um jegliche Unterstützung dankbar - mehr per PN

GaMAC: Открытая библиотека для автоматической кластеризации мультимодальных данных под GPU

На сегодняшний день не существует полноценного инструментария для кластеризации на графическом процессоре, что стало основным стимулом для создания универсальной библиотеки, способной автоматически решать задачи кластеризации данных различных представлений. Мы представляем GaMAC - библиотека автоматической оптимизации кластеризации с поддержкой с GPU.

https://habr.com/ru/articles/973364/

#кластеризация #clustering #мультимодальность #автоматическое_машинное_обучение #gpu #мультимодальная_кластеризация

When You Need Clustering or SVM Developers for ML Projects

Not sure when your ML project requires expert help? Clustering specialists and SVM developers can make a huge difference when working with complex datasets, uncovering hidden patterns, or building high-accuracy predictive models.

📌 Read the full article on the Amplework blog: https://www.amplework.com/blog/when-you-need-clustering-or-svm-developers/

#MachineLearning #Clustering #SVM #DataScienceExperts #MLProjects #AIEngineering #PredictiveModels #ScalableML #AIDevelopment

@raccoon granted, #Valve did already commit themselves to it being "open like a PC" so they can't sell at a loss. They at least need to break even in terms of cost to avoid businesses like datacenters to go: "Hey, that's pretty good!" and order like 1000+...

• My University did that with the #PS3 and ran them as clusters becaude even tho #YellowDogLinux and any #OtherOS only gets 2 threads, the #Cell was still a competent and affordable #ppc64 machine that one could use #MPI and do #clustering at full blast without breaking a sweat. And mind you #Sony lost out €100+ just in parts cost on every PS3 phat sold at launch because winning the #FormatWar with #BluRay against #HDDVD was more important long-term.

Behind the scenes of “beautiful scientific graphics”

If you think that clean, elegant visuals in scientific reports are produced in a couple of clicks — I’m here to disappoint you. 🙂

Here’s one of many failed attempts to visualize results based on thermodynamic modeling of geochemical processes.

Turning raw multidimensional data into something that a human reader can intuitively grasp is a separate challenge — with dozens of input parameters and only a few truly useful outcomes.

Good science is not only about computation, but also about communication.

#Geochemistry #DataViz #Rstats #Clustering #EnvironmentalModelling #ScienceCommunication #PHREEQC #WaterPollution #SvystunovaGully #Contamimation #Groundwater #Infographics #PCA #Thermodynamics

@emilyyoung also I'm pretty shure folks like @geerlingguy would consider some decently spec'd #ThinClients as an option to leran #Kubernetes and #Clustering on a shoestring budget if one can't just buy several decently spec'd #RaspberryPi|s…

🧠 New preprint by Chintaluri et al. (2025): An ion channel #omnimodel for standardized #biophysical #neuron #modelling. A unified #HodgkinHuxley formalism applied to >3,500 ion channel models from #ModelDB. Enables cross-model comparison, #clustering, and reproducible simulation through a shared parametrization:

🌍 https://doi.org/10.1101/2025.10.03.680368

#CompNeuro #Neuroscience #Reproducibility

Wondering what the implications of the structure and attitude of the #Fediverse are for the small world and the spread of cooperation vs antagonism.
#networks #clustering
https://youtu.be/CYlon2tvywA?si=FKjKO1pHoT7DIJ0-

DSM Disorders Disappear in Statistical Clustering of Psychiatric Symptoms

https://www.psychiatrymargins.com/p/traditional-dsm-disorders-dissolve?r=2wyot6&triedRedirect=true

#HackerNews #DSM #Disorders #Statistical #Clustering #Psychiatric #Symptoms #MentalHealth #Research

🚀 TopicWatchdog – Week 3: Stable Topics with BERTopic

KMeans worked, but cluster IDs kept jumping across retrains. This week I added a Python BERTopic stage with a BigQuery registry → stable topic IDs!

🟢 UMAP + HDBSCAN
🟢 Stable IDs via registry
🟢 Auto-labels with Gemini
🟢 Looker Studio dashboards

📊 3,802 topics → 2,472 mapped, top clusters: migration, economy, climate, politics.

👉 Blog: https://dracoblue.net/dev/topicwatchdog-stable-topics-with-bertopic/

#TopicWatchdog #BERTopic #BigQuery
#Clustering
#MachineLearning
#FediScience

TopicWatchdog (Week 2)

Move beyond string matching: cluster political topics from German shorts with embeddings + KMeans in BigQuery ML.

642 videos → 620 topics → 20 clusters (AfD positions, debates, labor market, climate/economy, corruption …).

Blogpost with SQL & charts:
🔗 https://dracoblue.net/dev/topicwatchdog-embeddings-and-kmeans-clustering-of-topics-claims/

#TopicWatchdog #BigQuery #Embeddings #Clustering #MachineLearning #FediScience

Partitioning vs Bucketing vs Clustering – What Every Data Engineer Must Know!
https://youtu.be/tvsAgQsmqfM

#partitioning #bucketing #clustering #theDataChannel

Summer ☀️ read: a new paper on model-based clustering just appeared in Computo!

Julien Jacques and Brendan Thomas Murphy publish a new method for clustering multivariate count data. The method combines feature selection and clustering, and is based on conditionally independent Poisson mixture models and Poisson generalized linear models.

On simulations, the Adjusted Rand Index (ARI) of the model with selected variables is close to the optimal ARI obtained with the true clustering variables.

The paper and accompanying R code are available at https://computo-journal.org/published-202507-jacques-count-data/

#machineLearning #clustering #Rstats #openScience #openSource #openAccess

Fascinating work, @gaenplancher.bsky.social ! #cognitive #aphantasia #clustering #innovative_approach #cognitive_profiles "Uncovering spatial and verbal cognitive profiles in aphantasia through unsupervised clustering" www.researchgate.net/publication/...

(PDF) Uncovering spatial and v...

Don't pass by the new insightful lecture from Dr. Alejandro Rodriguez Garcia, Abdus Salam International Centre for Theoretical Physics (ICTP)!

In this one, Alex provides a comprehensive overview of various clustering methods, including flat, fuzzy, and hierarchical approaches. His lecture not only discusses the mathematical foundations of techniques like k-means and k-medoids but also highlights their practical applications across fields such as image recognition and data classification.

This lecture is an excellent opportunity to deepen your understanding of unsupervised learning and engage critically with advanced clustering methods.

Join Enabla to watch the lecture and interact with Dr. Rodriguez Garcia for free! Ask questions and spark discussions with both him and the rest of the Enabla community: https://enabla.com/pub/1109/about

#UnsupervisedLearning #MachineLearning #DataScience #Clustering #OpenAccess