What does it take to train #Llama2 on your own data? Only 10 lines of YAML with
@ludwig_ai
. These new capabilities are part of our upcoming v0.8 release. Get a sneak peek and #finetune LLaMa2 with this free colab notebook: https://
colab.research.google.com/drive/1c3AO8l_
H6V_x37RwQ8V7M6A-RmcBf2tG?usp=sharing
….
CODE
-
Train Llama2 on Custom Data with Just 10 Lines YAML
By
–
-
HuggingGPT Agent Connects HuggingFace Models via LangChain
By
–
HuggingGPT Another installment to langchain_experimental Thanks to @itricktreat you can now use an agent which connects to a variety of @huggingface models in @langchain Based on the paper here: https://
arxiv.org/abs/2303.17580 Docs: https://
python.langchain.com/docs/use_cases
/autonomous_agents/hugginggpt.html
… -
Customizing Agent Prompts and Tools in Framework
By
–
Most agent implementations customize the prompts & tools We rewrote our agent getting started guide to better describe all the parts of our agent framework as well as how to customize them
-
RealChar Team Highlights AI Companions with LangSmith Integration
By
–
The whole @RealCharAI team is awesome Really excited to have this opportunity to help highlight their work! Play around with their GitHub here: https://
github.com/Shaunwei/RealC
har
… Read blog here: https://
blog.langchain.dev/realchar-x-lan
gsmith-ai-companions/
… -
Mastering Small Object Detection with SAHI Technology
By
–
🚀New Video🚀https://t.co/eQuoNsQhxR
— Satya Mallick (@LearnOpenCV) 24 juillet 2023
Mastering small object detection in digital images can be tough! The key is overcoming limited spatial scope, low pixel coverage, and low signal-to-noise ratio!
Check out our latest video on SAHI to learn more.#ComputerVision… pic.twitter.com/NjB1ero9rZNew https://
youtube.com/watch?v=UuOjJK
xn-M8
… Mastering small object detection in digital images can be tough! The key is overcoming limited spatial scope, low pixel coverage, and low signal-to-noise ratio! Check out our latest video on SAHI to learn more. #ComputerVision -
Multi-Repl Deletion Feature Now Available
By
–
7/ Multi-Repl Deletion
— Replit ⠕ (@Replit) 24 juillet 2023
A much-requested feature! Now you can select and delete multiple Repls rather than each individually. pic.twitter.com/LtyWVofawq7/ Multi-Repl Deletion A much-requested feature! Now you can select and delete multiple Repls rather than each individually.
-
Inline Code Generation with Ghostwriter Right-Click Feature
By
–
3/ Inline Code Generation
— Replit ⠕ (@Replit) 24 juillet 2023
Right-click in a code file and select “Generate Code”. Enter a prompt, and Ghostwriter will provide an inline suggestion. pic.twitter.com/KO0UCJbPse3/ Inline Code Generation Right-click in a code file and select “Generate Code”. Enter a prompt, and Ghostwriter will provide an inline suggestion.
-
Colored Ghostwriter Suggestions with Syntax Highlighting
By
–
2/ Colored Ghostwriter Suggestions
— Replit ⠕ (@Replit) 24 juillet 2023
Ghostwriter Code Completion & Generation text now shows slightly-dimmed code with Syntax Highlighting. pic.twitter.com/acagZtMq7H2/ Colored Ghostwriter Suggestions Ghostwriter Code Completion & Generation text now shows slightly-dimmed code with Syntax Highlighting.
-
Ghostwriter Inline Code Transformation Now Seamless
By
–
1/ Inline Code Transformation
— Replit ⠕ (@Replit) 24 juillet 2023
Transforming Code with Ghostwriter is now a seamless experience as it works inline!
You can toggle between comparison mode, your original code, and Ghostwriter’s suggestion. pic.twitter.com/WIhvLKOYoT1/ Inline Code Transformation Transforming Code with Ghostwriter is now a seamless experience as it works inline! You can toggle between comparison mode, your original code, and Ghostwriter’s suggestion.
-
Replit launches seven new AI-powered development features
By
–
The Replit Changelog
— Replit ⠕ (@Replit) 24 juillet 2023
Three weeks, 823 Pull Requests.
Here are seven new Replit features that just dropped 🧵 pic.twitter.com/gA6GRKMNFZThe Replit Changelog Three weeks, 823 Pull Requests. Here are seven new Replit features that just dropped