Hugging Face's logo
Roberto Bon's picture

Roberto Bon

Bjock
·

AI & ML interests

None yet

Recent Activity

reacted to mmaguero's post with 👍 about 1 month ago
🚀 Multidimensional Affective Analysis for Guarani/Jopara! 🌎 This project explored affective computing for low-resource languages, focusing on emotion recognition, humor detection, and offensive language identification in Guarani and Jopara (a code-switching mix of Guarani and Spanish). Highlights: 🧵 Corpora: - Emotion Recognition - Humor Detection - Offensive Language Identification 💻 Base Models for Fine-Tuning (trained on Guarani Wiki): - From scratch: BERT-based tiny, small, base and large models - Continuously pre-trained models: Multilingual-BERT and BETO 📓 Baseline Notebooks: - Fine-tuning BERT-based models - NCRF++ models via GitHub 💡 Check the repo! https://github.com/mmaguero/guarani-multi-affective-analysis 📖 Check out the publication here: - https://digibug.ugr.es/handle/10481/98843 - https://link.springer.com/article/10.1007/s12559-023-10165-0 #NLP #AffectiveComputing #LowResourceLanguages #Guarani #Jopara #SentimentAnalysis #AIForAll
reacted to Narsil's post with ❤️ about 1 month ago
Performance leap: TGI v3 is out. Processes 3x more tokens, 13x faster than vLLM on long prompts. Zero config ! 3x more tokens. By reducing our memory footprint, we’re able to ingest many more tokens and more dynamically than before. A single L4 (24GB) can handle 30k tokens on llama 3.1-8B, while vLLM gets barely 10k. A lot of work went into reducing the footprint of the runtime and its effect are best seen on smaller constrained environments. 13x faster On long prompts (200k+ tokens) conversation replies take 27.5s in vLLM, while it takes only 2s in TGI. How so ? We keep the initial conversation around, so when a new reply comes in, we can answer almost instantly. The overhead of the lookup is ~5us. Thanks @Daniël de Kok for the beast data structure. Zero config That’s it. Remove all the flags your are using and you’re likely to get the best performance. By evaluating the hardware and model, TGI carefully selects automatic values to give best performance. In production, we don’t have any flags anymore in our deployments. We kept all existing flags around, they may come in handy in niche scenarios. Read more: https://huggingface.co/docs/text-generation-inference/conceptual/chunking
View all activity

Organizations

None yet

models

None public yet

datasets

None public yet