@javendaxtw.bsky.social
Procesamiento Lenguaje Natural-LLM´s, aprendiz Ciencia de Datos, economista, datos abiertos, #RespetoALosDDHH-YA!, #useR , me gusta la música. de y en🇻🇪 javenda.netlify.app/about/
«El cáncer en el esófago me está colonizando el hígado. No me cabe ni un tratamiento bioquímico ni la cirugía porque mi cuerpo no lo aguanta. Lo que pido es que me dejen tranquilo. Se terminó mi ciclo hace rato. Sinceramente, me estoy muriendo y el guerrero tiene derecho a su descanso».
09.01.2025 12:42 — 👍 875 🔁 240 💬 33 📌 32Phil Schmid’s tutorial on fine tuning ModernBERT for classification (hope this is easy with prodigy/spacy) www.philschmid.de/fine-tune-mo...
26.12.2024 07:28 — 👍 50 🔁 8 💬 2 📌 0Some of my thoughts on OpenAI's o3 and the ARC-AGI benchmark
aiguide.substack.com/p/did-openai...
There is something very strange about cancer and how we react to it. It feels as if people show up all at once, and often, in the same way, they disappear all at once too.
20.12.2024 21:54 — 👍 4 🔁 1 💬 0 📌 0I'll get straight to the point.
We trained 2 new models. Like BERT, but modern. ModernBERT.
Not some hypey GenAI thing, but a proper workhorse model, for retrieval, classification, etc. Real practical stuff.
It's much faster, more accurate, longer context, and more useful. 🧵
ModernBERT is available as a slot-in replacement for any BERT-like model, with both 139M param and 395M param sizes.
It has a 8192 sequence length, is extremely efficient, is uniquely great at analyzing code, and much more. Read this for details:
huggingface.co/blog/modern...
L'Humanité du 19 décembre : Gisèle Pelicot, merci madame
📰 humanite.fr
Buen intento.
19.12.2024 12:34 — 👍 201 🔁 35 💬 5 📌 3very interesting to see that in posit RStudio the @github.com copilot is also working in its free version, so we #useRs , #userR #Rstats community also benefit from it
19.12.2024 21:42 — 👍 9 🔁 1 💬 0 📌 0Ayer fue asesinado en New York el CEO de la empresa más grande de prepagas de Salud en EEUU. Aparentemente el asesino dejó mensajes de frustración con la empresa escritos en las balas. Un poco sorprende el NYT publicando notas sobre el odio popular a las prepagas. www.nytimes.com/2024/12/05/n...
05.12.2024 18:19 — 👍 46 🔁 20 💬 3 📌 0¿No os da la sensación de que empezamos con la temporada final de la serie? ¿No?
03.12.2024 16:20 — 👍 166 🔁 27 💬 24 📌 2Es improbable que Bluesky llegue a alcanzar las cifras de Twitter. Mastodon no es divertido, y Threads penaliza la conversación. Es posible que el tiempo de las grandes fiestas se haya acabado.
Mi columna hoy en @elpais.com
elpais.com/opinion/2024...
Espero que tengan controladas sus alucinaciones
=======================
I hope they have their hallucinations under control
Pretty remarkable for an open model that can run in your Macbook!
30.11.2024 20:08 — 👍 1 🔁 1 💬 0 📌 0techniques facilitating the learning process and queries on the study materials.
This course is taught at the School of Economics of the @UCV.
all components of this system are open source or open-weights
I placed at the disposal of the students of the R programming course for Data Science a system based on @OpenWebUi and @ollama that together with the model @Alibaba_Qwen 2.5 coder, allow the application of retrieval augmented generation(RAG)
01.12.2024 21:58 — 👍 1 🔁 1 💬 1 📌 0agreed!
my support and solidarity with @danielvanstrien.bsky.social
Learn how to turn your Pandas data frame into a knowledge graph using LLMs. Build your own LLM graph-builder from scratch, implement LLMGraph Rransformer by LangChain, and QA your KG
towardsdatascience.com/building-a-k...
Es crucial que las universidades aprovechen los avances de la IA para mantenerse a la vanguardia y preparar a los estudiantes para los desafíos y oportunidades del futuro
#IA #Universidad #ética
www.diariodesevilla.es/sevilla/pasa...
Hello `World/BlueSky`
15.11.2024 05:33 — 👍 4 🔁 0 💬 0 📌 0