Label Studio, by HumanSignal, is the most flexible open source platform for AI evaluation and data labeling.
San Francisco
Joined May 2019
- Tweets478
- Following119
- Followers1.1K
- Likes234
Label Studio retweeted
How to build a tool to prepare and annotate your data for #AI model fine tuning with @LabelStudioHQ, @Gradio (UI) and a little code from #ChatGPT :
youtu.be/5HWeVZXQ-E4
Label Studio retweeted
Excited to be in SF this Wednesday for the AI Engineer meetup, where the @LabelStudioHQ team will be demoing, Adala, a new open source data labeling agent! Hope to see you there!
Nobody is doing a Devday Recap, so here it is:
OpenAI Devday Recap event next Wednesday November 8th 5:30PM with @humansignal_ and @aiengfoundation
partiful.com/e/5eKTJcttKkD5H…
Come join AI Engineers discussing what the announcements all mean. We also have spots for you if you'd like to present. Please reach out if so. We will try to make the event content available on youtube afterwards.
Label Studio retweeted
I'm so excited about this project and can't wait to see what people will build
💫 We're excited to announce our latest experimental open-source project for autonomous data labeling: Adala
github.com/humansignal/adala
Label Studio retweeted
The future of @LabelStudioHQ and @humansignal_ is AI agents
HumanSignal launches Adala open source framework for autonomous data labeling agents venturebeat.com/ai/humansign…
Label Studio retweeted
I hate labeling data, but it's an important task for Machine Learning engineers.
Currently, I'm using the following tools for labeling:
💁GPT-4 and GPT-3.5 from @OpenAI
🔨and @LabelStudioHQ
How do you label data? What tools do you use?
Label Studio retweeted
I collaborated with the @LabelStudioHQ folks to do a quick and simple example for fine-tuning LLM applications - in example, creating a tunable chat bot for your GitHub repo. Check out the blog and the code!
labelstud.io/blog/from-found…
Label Studio retweeted
🏷️Log to @LabelStudioHQ
Label Studio is an open-source data labeling platform. Use it in 🦜🔗to label data for fine-tuning LLMs, and for collection and evaluation of human feedback.
s/o to GH user niklub for the addition!
Docs: python.langchain.com/docs/in…
Label Studio retweeted
Excited about this work! I see a world in which we can use LMMs + tools like Prodigy (@spacy_io), @LabelStudioHQ @argilla_io to annotate a bunch of domain-specific data very efficiently and then train tiny models that do one thing but do it well + have a lower environmental cost.
I have no idea who has been downloading my 0-shot model 4+ million times this month via @huggingface but that's v motivating. I'm working on an update, should be done in a few weeks. Encoder 0-shot models are very efficient: they can run on a Raspberry Pi, no A100 GPU needed 1/2
Label Studio retweeted
I'm excited for what the future holds for @LabelStudioHQ ✨
Label Studio retweeted
In a nod that humans are still useful in an AI world, Heartex (leading open-source data labeling tool) is re-branding to HumanSignal #OSS #AI
Read more about how we're betting big on people.✨
Help us celebrate this step forward as HumanSignal, harmonizing human insights and feedback with AI progression.
Full announcement ⤵️
🔗 humansignal.com/blog/introdu…
Label Studio retweeted
🤖 This Open Source Friday we're exploring the role that data labeling plays in LLMs.
📆 Join us on June 16 at 1pm ET as @erinmikail and @hogepodge discuss how @LabelStudioHQ empowers AI's evolution.
👾 Live on twitch.tv/github
✅ RSVP: gh.io/heartex
ALT How Data Labeling Empowers AI's Evolution featuring Erin Mikail Staples, Developer Advocate at Heartex on June 16 at 1pm
Label Studio retweeted
I am very excited for my friends @heartex for getting their @labelstudiohq 1.8 open source release out today. I think that training enterprise LLMs will be critical in the future. And labeling tools that make humans more productive for that time-consuming task is critical.
Label Studio retweeted
The @LabelStudioHQ crew killing it (as usual)!
Label Studio retweeted
FRIENDS! FANS! STRANGERS!
Who’s ready to get their geek out on today?!?
Join me, @JimmyMWhitaker and @lubimovnik as we nerd out about #RLHF and how to retrain large language models.
2 PM ET today!
RSVP: lu.ma/RLHF
ALT me wearing a green patterned dress smiling and selfie-ing for the workshop
Mark your calendars! 📅
Tomorrow, we're hosting a workshop to get folks started with #RLHF!
Join @JimmyMWhitaker, @erinmikail, and @lubimovnik for an entry-level workshop on Reinforcement Learning with Human Feedback.
🔗lu.ma/RLHF
Label Studio retweeted
💥 New episode of Practical AI!
📌 Creating instruction tuned models
😎 with @erinmikail from @LabelStudioHQ
⚡️ hosted by @dwhitena
📍 recorded at #odsceast @_odsc
💚 practicalai.fm/223
Label Studio retweeted
Loved geeking out with @dwhitena at @_odsc about tuning models, improving machine learning from human feedback, and the exploding AI/ML world.
Give it a listen, and reach out if you've got any questions!
💥 New episode of Practical AI!
📌 Creating instruction tuned models
😎 with @erinmikail from @LabelStudioHQ
⚡️ hosted by @dwhitena
📍 recorded at #odsceast @_odsc
💚 practicalai.fm/223
Label Studio retweeted
Repping @pyladies / @NYCPyLadies at @_odsc!
Looking forward to chatting with @dwhitena shortly and geeking out on the importance of data labeling and the role of open source tools like @LabelStudioHQ! – at Boston, MA
ALT Taking a selfie in front of a sign that says “the future of machine learning starts here”
Label Studio retweeted
In this @LabelStudioHQ webinar, our Sr #AI Solutions Architect, @aaronschliem gave insights into the data design lifecycle, sharing tips, potential pitfalls, & solutions to support your #data collection efforts
Watch the recording here ➡ bit.ly/3GSea1C
Label Studio retweeted
It's pretty easy to use @LabelStudioHQ for this (danielvanstrien.xyz/huggingf…). Hopefully there will be an even easier integration soon :) cc @hogepodge @erinmikail
Label Studio retweeted
🧑🏫 This week @TimescaleDB’s @HarlemCavalier schooled us on timeseries data.
Listen to the show to learn all about the differences and similarities between warehouse data and timeseries data.
bit.ly/3J11khH
