#ListenBrainz / #MetaBrainz I'm confused. Aren't sponsors the true customer? Why use this? 
On one hand #Music: "Listen together", "Ethical forever"
On the other: #DATASETS
"Some of the world’s biggest platforms such as Google and Amazon, use our data"
"We ask commercial supporters to support us in order to help fund the creation and maintenance of these datasets."
"The following organizations make use of the data-sets published by MetaBrainz"
STAT: Gold-standard maternal mortality database in limbo as CDC staff placed on leave. “As part of the sweeping layoffs that rocked the Department of Health and Human Services on Tuesday, the entire staff that oversaw an annual survey to better understand infant and maternal health — and that was considered the gold standard in the field — was placed on administrative leave. The Pregnancy […]
This data may vanish under Trump, so we charted it
Some of most valuable #datasets in human history vanished from #US #government websites, felt like watching Library of Alexandria go up in smoke
Many have gone on record describing #Census Bureau’s #American Community Survey as wonder of modern world
Another loss? #HouseholdPulse survey, online survey that provided week-by-week data on income losses, economic struggles and precarious mental health
https://www.washingtonpost.com/business/2025/02/14/this-data-may-vanish-under-trump-so-we-charted-it/
https://archive.ph/mB512
#CDC #Datasets at archive.org: https://archive.org/details/20250128-cdc-datasets
#meettheexperts #CESSDA
New #video on #YouTube: Meet our experts Dr. Anja Perry @Datendealerin & Dr. Jonas Recker who present #LGBTIQ+ #datasets, their temporal and spatial coverage, and which LGBTIQ+ identities were sampled.
"Some federal #datasets are nearly irreplaceable. Hurricane Helene helped drive that fact home in September, when it flooded much of western North Carolina and temporarily knocked NOAA’s NCEI headquarters in Asheville offline. Scientists found they were unable to complete certain kinds of analyses until the databases were back up and running."
Hey #DataScience people!
I am about to start my first “Introduction to Data Science” course at #University, and our professor asked us to team up and think about a project that we want to do.
Nevertheless, since I don’t know anything about the topic yet, I would really appreciate any tips of entry-level data science projects that I could do with #OpenData #DataSets in #Python!
Probably, we will be using #pandas. Since you’re here, any additional learning resources or general suggestions are much welcome, too!
Thanks 
(Not sure how useful it is, but this is the course link: https://ois2.tlu.ee/tluois/subject/ULP6613-23265)
The #wasm build of #sqlite opens up great new possibilities in the #browser, especially when coupled with the Origin Private File System (#opfs). I've used it to implement importing huge #csv or #jsonl #datasets right in the browser. Import, validate, search, edit, close the browser and continue tomorrow, stream the database (with on-the-fly compression!) to the server when ready – it's all possible!
Another interesting use case I came across today: use it in #golang to get rid of #cgo!
[1/2]
Here, we display the performance of the algorithm on the DVSgesture dataset. For this gesture recognition task, the online HOTS accuracy remains close to the chance level for about 100 events. More evidence needs to be accumulated, and then the accuracy increases monotonically, outperforming the previous method after about 10.000 events (at an average of 9.3% of the number of events in the sample) :
Under the carpet of this article, Mark Zuckerberg is hiding the fact that—contrary to traditional free/open source software—the freedoms to study and modify #Llama models are asymmetric. Meta can exercise both with much more ease than everyone else, because they are the only ones with access to the training datasets (and full training pipeline): https://about.fb.com/news/2024/07/open-source-ai-is-the-path-forward/
#Opensource #AI needs open training #datasets.
Oh, and scientific #reproducibility needs them too. #openscience
The collection of vast #datasets has become crucial for discovering the secrets of the #universe.
What are the challenges and innovations in managing #astronomical #data and how do we deal with #metadata in #astronomy?
Read the interview with a #BigData #astronomer and an expert on #archiving data from #satellite-mounted astronomical instruments:
"[W]hen we bury code, we bury decisions beneath a million lines of if/then statements... and biases of machine learning. Unchallenged, the world that has slipped past us shapes the predictions of these systems in ways we cannot understand."
The Ghost Stays in the Picture by our research fellow @CyberneticForests
Have you checked out @opensuse's new space on @huggingface? Discover the #AI models & #datasets we already have for #cavil's #opensource #licensing. #collaborate & #contribute! 
#AI #openSUSE #HuggingFace https://news.opensuse.org/2024/05/24/osp-listed-as-org-on-ai-hugging-faces/
The Institute for Dissent and Datalove is a loose collective comprised of hackers, artists, activists and tinkerers. It overlaps with networks of solidarities involved in active defense of free speech and free/libre technologies, technology critics and political interventions.
The Institute for Dissent and Datalove has so far mostly been used for operations of de/re-contextualization of large datasets, de-formatting of formats and playful use of liberating algorithms.
It tries to criticize and deconstruct itself, while remaining grounded in uncompromising collective practices of autonomy and solidarity.
We even have a website: https://dissent-and-datalove.institute/
An amazing post on how big #data for training #AI models is created, using #LION-5B as the example, the authors illustrate the caveats, the issues, and other lesser known biases and limitations on these fundamental datasets.
"Models all the way down" https://knowingmachines.org/models-all-the-way
Someone I follow just launched this site. Looks pretty interesting.
Does anyone want a dataset of MDNI fedi accounts? One of the things that I do when blocking MDNI accounts is putting some basic info into a spreadsheet, and now, using a Python script, I have some JSON files with the same data as well.
"Discovering #Datasets Through Machine Learning: An Ensemble Approach to Uncovering the Prevalence of #Government-Funded Datasets" · Harvard Data Science Review
Hey #dataviz and #datascience and #research folks:
I'm trying to analyze the FBI's #NIBRS "master files". They have a deeply confusing (to me) structure, and I can't seem to find anything online showing how to extract the information e.g. to a CSV file (except some questions like mine with no good answers).
Anyone have experience? Anyone know where there's a good walkthrough or how-to?
