Data Science | Machine Learning with Python for Researchers
31.5K subscribers
1.58K photos
102 videos
22 files
1.86K links
Admin: @HusseinSheikho

The Data Science and Python channel is for researchers and advanced programmers

Buy ads: https://telega.io/c/dataScienceT
Download Telegram
📎 Instruction-tuning Stable Diffusion with InstructPix2Pix

InstructPix2Pix training strategy to follow more specific instructions related to tasks in image translation (such as cartoonization) and low-level image processing (such as image deraining).

🖥 Post: https://huggingface.co/blog/instruction-tuning-sd

⭐️ Training and inference code: https://github.com/huggingface/instruction-tuned-sd

📌 Demo: https://huggingface.co/spaces/instruction-tuning-sd/instruction-tuned-sd

InstructPix2Pix: https://huggingface.co/timbrooks/instruct-pix2pix

🔍Datasets and models from this post: https://huggingface.co/instruction-tuning-sd

https://t.iss.one/DataScienceT
❤‍🔥2
QLoRA: Efficient Finetuning of Quantized LLMs

Model name Guanaco, outperforms all previous openly released models on the Vicuna benchmark, reaching 99.3% of the performance level of ChatGPT while only requiring 24 hours of finetuning on a single GPU.

🖥 Github: https://github.com/artidoro/qlora

Paper: https://arxiv.org/abs/2305.14314

⭐️ Demo: https://huggingface.co/spaces/uwnlp/guanaco-playground-tgi

📌 Dataset: https://paperswithcode.com/dataset/ffhq

https://t.iss.one/DataScienceT
❤‍🔥2
Large Language Models as Tool Makers

In this work, we take an initial step towards removing this dependency by proposing a closed-loop framework, referred to as LLMs A s Tool Makers (LATM), where LLMs create their own reusable tools for problem-solving.

🖥 Github: https://github.com/ctlllll/llm-toolmaker

Paper: https://arxiv.org/pdf/2305.17126v1.pdf

📌 Dataset: https://paperswithcode.com/dataset/big-bench

https://t.iss.one/DataScienceT
👍21
Prompt-Free Diffusion: Taking "Text" out of Text-to-Image Diffusion Models

The performance of Text2Image is largely dependent on text prompts. In Prompt-Free Diffusion, no prompt is needed, just a reference images.

🖥 Github: https://github.com/shi-labs/prompt-free-diffusion

🔎 Demo: https://huggingface.co/spaces/shi-labs/Prompt-Free-Diffusion

Paper: https://arxiv.org/abs/2305.16223v1

📌 Dataset: https://paperswithcode.com/dataset/ffhq

https://t.iss.one/DataScienceT
❤‍🔥21👍1
Large Language Models as Tool Makers

In this work, we take an initial step towards removing this dependency by proposing a closed-loop framework, referred to as LLMs A s Tool Makers (LATM), where LLMs create their own reusable tools for problem-solving.

🖥 Github: https://github.com/ctlllll/llm-toolmaker

Paper: https://arxiv.org/pdf/2305.17126v1.pdf

📌 Dataset: https://paperswithcode.com/dataset/big-bench

https://t.iss.one/DataScienceT
❤‍🔥1👍1
🖥 A Practical Toolkit for Multilingual Question and Answer Generation

Multilingual/multidomain question generation datasets, models, and python library for question generation.

🖥 Github: https://github.com/asahi417/lm-question-generation

Paper: https://arxiv.org/abs/2305.17416v1

📌 Dataset: https://paperswithcode.com/dataset/squad

https://t.iss.one/DataScienceT
👍1
🦙 BigTrans 🚀

BigTrans which adapts LLaMA that covers only 20 languages and enhances it with multilingual translation capability on more than 100 languag

🖥 Github: https://github.com/ZNLP/BigTrans/tree/main

Paper: https://arxiv.org/abs/2305.18098v1

📌 Dataset: https://paperswithcode.com/dataset/flores-200

https://t.iss.one/DataScienceT
This media is not supported in your browser
VIEW IN TELEGRAM
🔥 GPT4Tools: Teaching LLM to Use Tools via Self-instruction

GPT4Tools is a centralized system that can control multiple visual foundation models. It is based on Vicuna (LLaMA), and 71K self-built instruction data.

🖥 Github: https://github.com/stevengrove/gpt4tools

Paper: https://arxiv.org/abs/2305.18752v1

📌 Project: https://gpt4tools.github.io/

https://t.iss.one/DataScienceT
Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistles

Hiera is a hierarchical vision transformer that is fast, powerful, and, above all, simple. It outperforms the state-of-the-art across a wide array of image and video tasks while being much faster.

pip install hiera-transformer

🖥 Github: https://github.com/stevengrove/gpt4tools

Paper: https://arxiv.org/abs/2306.00989v1

📌 Dataset: https://paperswithcode.com/dataset/inaturalist

https://t.iss.one/DataScienceT
❤‍🔥3👍1
Wuerstchen: Efficient Pretraining of Text-to-Image Models

Novel technique for text-to-image synthesis that unites competitive performance with unprecedented cost-effectiveness and ease of training on constrained hardwar

🖥 Github: https://github.com/dome272/wuerstchen

Paper: https://arxiv.org/abs/2306.00637v1

📌 Colab: https://colab.research.google.com/drive/1UTP9Xn2UIrVbAXyL-SKEvyLmgVWdw-Vy

https://t.iss.one/DataScienceT
❤‍🔥3