CODE

→ View original post on X — @rasbt

LLaMA-Adapter: Efficient LLM Fine-tuning with Tunable Prefixes

By

@rasbt

–

13 June 2023 16h51

I really like the LLaMA-Adapter method! In a nutshell, it's about finetuning LLMs more efficiently. In a nutshell, it's adding a tunable prefix to the key and value tensors in the self-attention layers. Fun fact: it's not specific to LLaMA. Use it with any LLM!

→ View original post on X — @learnopencv

IoU Loss Functions for Object Detection: GIoU, DIoU, CIoU

By

@learnopencv

–

13 June 2023 16h35

🚀Blog-tastic Tuesdays🚀

Dive into the world of object detection with #DeepLearning!https://t.co/egztROZxfW

Check out our latest read as we break down the IoU Loss Functions (GIoU, DIoU, CIoU)#AI #MachineLearning pic.twitter.com/AcwvIf33Qw
— Satya Mallick (@LearnOpenCV) 13 juin 2023

Blog-tastic Tuesdays Dive into the world of object detection with #DeepLearning! https://
learnopencv.com/iou-loss-funct
ions-object-detection/
… Check out our latest read as we break down the IoU Loss Functions (GIoU, DIoU, CIoU) #AI #MachineLearning

→ View original post on X — @axeleraai

No-Code Cloud Platform for Edge Computer Vision Applications

By

@axeleraai

–

13 June 2023 13h07

Discover the power of creating advanced #ComputerVision applications at the #edge without writing a single line of #code! In the final episode, Ioannis Koltsidas takes you through our state-of-the-art cloud platform's intuitive, user-friendly graphical interface. #NoCode pic.twitter.com/gMGMKJp9za
— Axelera AI (@AxeleraAI) 13 juin 2023

Discover the power of creating advanced #ComputerVision applications at the #edge without writing a single line of #code! In the final episode, Ioannis Koltsidas takes you through our state-of-the-art cloud platform's intuitive, user-friendly graphical interface. #NoCode

Llama.cpp Introduces Full CUDA GPU Acceleration Support

By

–

13 June 2023 6h13

Llama.cpp: Full CUDA GPU Acceleration PR: https://
github.com/ggerganov/llam
a.cpp/pull/1827
…

Cap3D: Scalable Automatic 3D Object Text Description System

By

–

13 June 2023 5h56

Scalable 3D Captioning with Pretrained Models paper page: https://
huggingface.co/papers/2306.07
279
…
dataset: https://
huggingface.co/datasets/tiang
e/Cap3D
… introduce Cap3D, an automatic approach for generating descriptive text for 3D objects. This approach utilizes pretrained models from image captioning, image-text

→ View original post on X — @rasbt

Humorous commitment to converting someone to PyTorch

By

@rasbt

–

13 June 2023 4h17

maybe not this year, but I'll convert you to PyTorch eventually haha

→ View original post on X — @rasbt

Pretrained Models Make Fine-tuning the Only Practical Approach

By

@rasbt

–

13 June 2023 4h13

Hah, and even further, with pretrained models available, anything more than finetuning feels like a pain

Weakly Supervised Information Extraction from Handwritten Document Images

By

–

13 June 2023 3h46

Weakly supervised information extraction from inscrutable handwritten document images paper page: https://
arxiv.org/abs/2306.06823 State-of-the-art information extraction methods are limited by OCR errors. They work well for printed text in form-like documents, but unstructured,

High-Fidelity Audio Compression with Improved RVQGAN

By

–

13 June 2023 3h30

High-Fidelity Audio Compression with Improved RVQGAN paper page: https://
huggingface.co/papers/2306.06
546
… Language models have been successfully used to model natural signals, such as images, speech, and music. A key component of these models is a high quality neural compression model that can