🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Here's an animation of a PyTorch DataLoader. It turns your dataset into a shuffled, batched tensors iterator. (This is my first animation using Manim Community, the community fork of Grant Sanderson's manim) Here's a little summary of the different parts for those curious: 1/5

How do you build a 𝗟𝗟𝗠 𝗯𝗮𝘀𝗲𝗱 𝗖𝗵𝗮𝘁𝗯𝗼𝘁 𝘁𝗼 𝗾𝘂𝗲𝗿𝘆 𝘆𝗼𝘂𝗿 𝗣𝗿𝗶𝘃𝗮𝘁𝗲 𝗞𝗻𝗼𝘄𝗹𝗲𝗱𝗴𝗲 𝗕𝗮𝘀𝗲? Let’s find out. First step is to store the knowledge of your internal documents in a format that is suitable for querying. We do so by embedding it using an

Gzip + kNN beats transformers on text classification. (Gzip as in good old zip file compression)

AI generative fill extending scenes, movies shot in portrait format by Alex Cerrato

We are looking for a research engineer to work on domain adaptation and transfer learning École polytechnique near Paris. Come with us to do research, open source Python software and benchmarks. Contact me by email if interested. Please RT (free users need to help each other).

Allez hop pti giveaway sur mon compte ça faisait longtemps, on vous fait gagner un vélo électrique d’une valeur de 2500€ avec EDF ! Pour participer, RT/like le tweet et follow PA 🕺🏻 & @energiedusport, tirage au sort le 04/10 ⚡ #CollaborationCommerciale

Meet #PVCC's first image 😍 Ten years after its siblings, #PVCC's instrument took its first in-orbit multispectral image 🌍 Those first space steps, taken only 3️⃣ days after launch, allow to stretch the impressive swath of 350km and survey Switzerland's vegetation in one sweep!

Presenting GraphCast: our state-of-the-art AI model delivering 10-day weather forecasts with unprecedented accuracy in under one minute. 🌦️ It can even help predict the potential paths of cyclones further into the future. Here's how it works. 🧵 dpmd.ai/presenting-gra…

Happy to announce our latest paper:  Grounding Everything: Emerging Localization Properties in Vision-Language Transformers 🎉🎉🎉 GEM allows a training-free adaptation of Vision-Language models (e.g., CLIP ...) to perform zero-shot open-vocabulary segmentation.

I'll be presenting Score-based Data Assimilation at NeurIPS Conference on December 12th 🥳 Come talk about generative models, score priors, Markov blankets, dynamical systems and much more! 📜 arxiv.org/abs/2306.10574 👨‍💻 github.com/francois-rozet… 💬 neurips.cc/virtual/2023/p…

me trying to keep up with ai

✅Most important Training and Optimization Techniques for Transformers in ML - Explained in Simple terms. A quick thread 👇🏻🧵 #MachineLearning #Coding #100DaysofCode #deeplearning #DataScience PC : Research Gate

The FLAIR dataset 🛩️🛰️ is also available on Hugging Face : - over 20 billion annotated pixels of very high resolution aerial imagery at 0.2 m spatial resolution - along with Sentinel-2 satellite 1-year time series with 10 spectral band at 10m huggingface.co/datasets/IGNF/…

My top 8 predictions for the geospatial space in 2024:

Exciting opportunity! Join my new research group at the University of Zurich as a PhD candidate in Reinforcement Learning. Contribute to cutting-edge AI research and shape the future of the field 🔥 To apply follow the instructions on my website: gioramponi.github.io #UZH #RL

Need a dense multi-vector retrieval model in your language? We've got you covered with ColBERT-XM, a new modular retriever that supports 81+ languages and can easily be extended to many more. [1/4] 📑 Paper: arxiv.org/abs/2402.15059 🤗 Model: huggingface.co/antoinelouis/c…

🚀 Yesterday, March 4th, marked a momentous occasion as our team gathered to witness the successful deployment of four of our satellites into orbit. After months of hard work and meticulous planning, we watched with anticipation as our birds were released from Falcon 9 for

We collaborated with the European Space Agency to open-source the largest ever earth observation dataset: Major TOM Core! About half of the entire planet is covered. That's 2,245,886 patches of 1068 x 1068 pixels. At 10m resolution, we've got 256 million square km with over 2.5

I strongly believe AI papers should standardize including model size and benchmarking model speed against others. In my opinion, this is more crucial than gaining a few percentage points on a dataset, especially from a production point of view.