Argilla
@argilla_io
Followers
3,415
Following
28
Media
287
Statuses
1,293
Explore trending content on Musk Viewer
#GMMTVOuting2024
• 308232 Tweets
George Floyd
• 279355 Tweets
Sant Rampal Ji Maharaj
• 162045 Tweets
#HolyHinduScriptures_Vs_Hindu
• 138979 Tweets
Gaga
• 118377 Tweets
ダービー
• 98658 Tweets
Celtics
• 91024 Tweets
Libertarians
• 80301 Tweets
コミティア
• 57707 Tweets
OUTING 2024 X GEMINI FOURTH
• 42585 Tweets
May Special
• 34051 Tweets
キュアニャミー
• 29008 Tweets
設営完了
• 25014 Tweets
プリキュア
• 24255 Tweets
ジャスティンミラノ
• 23224 Tweets
Jrue
• 22702 Tweets
Millonarios
• 19558 Tweets
東京競馬場
• 19260 Tweets
ホッパー1
• 18717 Tweets
Saint MSG Insan
• 17649 Tweets
レガレイラ
• 16966 Tweets
ガッチャード
• 15460 Tweets
Ross Ulbricht
• 15427 Tweets
#超超超超ゲーマーズday2
• 14714 Tweets
コスモキュランダ
• 12788 Tweets
ユキちゃん
• 12229 Tweets
アーバンシック
• 11260 Tweets
イーウィニャ
• 10865 Tweets
OffGun Outing Verr
• 10213 Tweets
Pinned Tweet
💥After months of work, we're thrilled to introduce ⚗️distilabel 1.0.0!
🚀More flexible, robust, and powerful.
🙌 Let's empower the community to build the most impactful datasets for Open Source AI!
Blogpost:
Github:
4
19
100
🚀 Open-source AI strikes again! Announcing Notux 8x7B, a fine-tune of Mixtral Instruct with high-quality chat data and DPO.
Notux now the top ranked MoE on the Open LLM leaderboard.
8
84
436
🚀🧙🏼♂️Introducing OpenHermesPreferences: the largest open dataset for RLHF & DPO
Built together with the
@huggingface
H4 team, it's a 1M preferences dataset on top of the amazing
@Teknium1
's dataset.
Let's dive in! 🧵
6
58
288
Build a news classifier from scratch with weak supervision
1. Programatically label 38.000 examples with rules and Snorkel.
2. Train a downstream classifier with scikit-learn to achieve 0.81 macro avg. f1-score.
Tutorial link below 👇
#nlproc
#ml
#datascience
#opensource
2
54
277
🔥Open-source, open-science, and data curation for the win!
Meet Notus 7B, a new LLM tuned with DPO on a new curated UltraFeedback dataset, surpassing Zephyr and Claude 2 on AlpacaEval.
Built on the shoulders of giants: 🙌
@huggingface
Alignment Handbook
1
64
252
🔥 More is less for DPO, high quality matters!
📢 Dropping our first open dataset and LLM of the year:
💾Meet distilabel Orca Pairs DPO, an improved version of the now famous dataset from
@intel
🏛️And a new OpenHermes model outperforming baselines with 54% less DPO pairs
🧵
5
46
231
🔥 Introducing a new open dataset for the Open Source AI community:
OpenHermes2.5-dpo-binarized-alpha built atop the amazing dataset by
@Teknium1
This time we use OSS models for everything, even for the preference step!
🧵
39
51
181
Training a text classifier without labelled data using
@PyTorch
End-to-end weak supervision with Weasel &
@huggingface
transformers
Guide and more details below 👇
#nlproc
#datascience
#python
#opensource
1
44
201
🚀Data labeling from the
@huggingface
Hub is here.
No more excuses to build great NLP datasets!
0
33
201
🤖Yesterday, we shared a large AI feedback dataset
👩💻Today,
@argilla_io
&
@huggingface
are thrilled to release a high-quality human feedback dataset, built with and for the community!
10K Prompts Ranked: +14K human ratings from +300 contributors!
2
20
142
🚀 The OSS AI community needs more open datasets for improving LLMs:
🎁 Excited to ship a new open DPO dataset for boosting chat models:
⚗️ distilabel capybara-dpo, a multi-turn preference dataset built atop the awesome dataset by
@ldjconfirmed
🧵
4
24
120
⚗️ distilabel, our OSS framework for building LLM finetuning & alignment datasets just crossed 300 stars
Let's do a quick overview of its current features
2
14
118
We're building a high-quality prompt dataset together with the community.
The result will be published with an open, commercial-friendly license, anyone can use to build eval, sft and DPO datasets.
We need your help! This is how simple is to contribute:
4
26
113
A simple active learning loop with the amazing ModAL library
Active learning for YT spam classification inside a Jupyter notebook, a tutorial by
@vid_algo
Rubrix:
ModAL:
#nlproc
#python
#opensource
0
25
108
🚀 Want to train LLMs in your own language? For commercial use? Introducing
@databricks
Dolly Multilingual Datasets
Currently includes translations into Spanish, French, and German. Want to see more languages? Join us!
1
15
105
🤖 Turing-Complete Transformers: Two Transformers are more Powerful than One.
In this paper , currently under review as a conference paper at ICLR2023, the researchers present Find+Replace, a family of transformer architectures that are Turing-complete.
7
18
104
🌊Introducing DPO Mix 7K, a small DPO dataset that does wonders!
Yesterday,
@_philschmid
&
@_lewtun
showcased its strength with Zephyr Gemma
If you're looking for a small, diverse, high quality DPO dataset check it out!
2
25
103
🔥 Data is better together🔥
At
@argilla_io
&
@huggingface
, we believe in the collective intelligence of the OSS AI community
So we have partnered to let everyone contribute to AI datasets!
You can start contributing now to the first initiative:
🧵
2
18
102
Yesterday, we shared 10k_prompts_ranked: +14K human ratings by +300 amazing contributors
To understand the data, we collaborated with our friends
@graphext
, the data analysis tool for visual thinkers
Now it's available to everyone👇
Let's deep dive!
🧵
4
20
102
⚗ How distilabel works?
🚀 Yesterday, we announced our new open-source project, fully integrated with Argilla.
First of all, if you don't yet know the project, go to 👉
Today we want to share more details about how it works 🧵
15
30
99
Fine-tune a Hugging Face transformers for your own domain
Iteratively build a training set and fine-tune a sentiment classifier for the banking domain
Tutorial by
@dvilasuero
#NLProc
#python
#opensource
1
30
94
🔥Thrilled to share our new tutorial:
Collecting human preference data and training a reward model with the awesome trl by
@huggingface
The very first end-to-end example of the new Reward Trainer of trl
Colab with
@younesbelkada
&
@lvwerra
1
25
93
Training a Hugging Face text classifier directly from search queries?
User Weasel & Rubrix 👉
Weasel: End-to-End weak supervision by
@CachaySalva
&
@BenBoecking
:
Rubrix:
#nlproc
#python
#opensource
0
34
89
Zero-shot sentiment classification with human-readable explantions with
@OpenAI
GPT-3. Is it any good?
🤖 You can now run this yourself using Colab and 🤗 Spaces:
👀 Or see by yourself on this Argilla Space (argilla/1234):
1
16
92
🔥 Open source, open datasets & open collaboration go a long way
🍿The story behind NeuralBeagle14, a top performing 7B model released by
@maximelabonne
4
14
89
Data quality, fueled by human feedback, is the next frontier of LLMs
We are stoked to introduce the first open-source, enterprise-grade solution for the scalable collection of human feedback to power the next wave of custom LLMs
1/7
5
21
78
🔥 We'll be doubling down on open DPO datasets, why?
A few months ago:
UltraFeedback 62K examples --> got Zephyr beta sft from 7.0 to 7.34 on MTBench
Last week:
@argilla_io
mix 7K examples --> got Zephyr Gemma sft from 7.17 to 7.82 on MTBench
2x improvement with 9x less data
1
21
74
🌏🌎🌍 Interested in evaluating LLMs for your language?
We've selected 500 high-quality prompts.
We'll be supporting the community to validate the translations and create an open multilingual benchmark
Get involved & nominate yourself as language lead
1
21
73
🌎 Better AI is better data, and for better data we need expertise!
As part of the 'Data is Better Together' project in collaboration with Hugging Face, we bring the Domain Specific Datasets.
You can read more in this post: 🤗
1
26
73
🦒 Improving Text Embeddings with LLMs, a new distilabel tutorial:
In this distilabel tutorial, we will be replicating the process described in "Improving Text Embeddings with Large Language Models" by Liang Wang et al. ().
1
12
72
💬 Did you know you can finetune an LLM, Mistral 7B in particular, on a chat-style instruction dataset? This Argilla tutorial has a step-by-step guide!
Argilla:
distilabel:
#nlproc
#llms
#python
#opensource
2
5
70
🤖👩💻Do humans prefer synthetic or human generated prompts?
If you wanna know, participate in the prompt collective event and share with your friends & teammates!
With +230 participants and ~4,300 ratings already, here's some interesting results
1/7🧵
4
18
70
What a week!
> We launched Notus, one of the most powerful commercial LLMs out there.
>
@weights_biases
, the iconic AI company, shows how they use Argilla for LLM eval
> A DPO tune of OpenHermes by our admired
@Teknium1
shows great gains using our curated preferences dataset.
3
10
71
Data is one of the best ways to contribute to open source AI
Two weeks ago we shared the distilabel orca pairs dataset.
13k downloads, 10s of models fine tuned with it, still trending on
@huggingface
But that's not the most important
🧵
1
15
67
🔥🦜🔗 Stoked to announce our integration with
@LangChainAI
Monitor and collect human feedback from LangChain apps with a few lines of code
🚀Great job by
@alvarobartt
Thanks
@hwchase17
and team!
✍️LLM feedback with
@argilla_io
@argilla_io
is a great tool for LLM data labeling, curation, and monitoring. Use the new Argilla callbacks to easily log and provide feedback for you LangChain LLMs!
Integration docs:
Argilla docs:
2
4
40
1
14
69
🌸 Synthetic Haiku DPO 🌸
🙌A DPO dataset by
@vanstriendaniel
generated with OSS models
⚗️ Built with distilabel using the awesome OpenHermes by
@Teknium1
Let's dive in!
🧵
1
11
65
Finding and 𝗰𝗼𝗿𝗿𝗲𝗰𝘁𝗶𝗻𝗴 label errors
1⃣ Train a text classifier, predict over the test set
2⃣ Find label errors with the built-in cleanlab integration
3⃣ Correct errors with the UI
Practical tutorial by
@vid_algo
#nlproc
#opensource
#ml
0
24
66
Building a text classifier with Flyingsquid it's never been easier
Flyingsquid is a label model for fast and accurate weak supervision
Weak supervision guide:
Flyingsquid:
Rubrix:
#nlproc
#opensource
#ml
1
16
62
🙋❓ Curious about the differences between DPO, KTO, ORPO, and other preference alignment algorithms? Check out a comprehensive overview in the latest blog post of our series with MantisNLP.
2
13
61
🧮 Meta's LIMA, a model trained by fine-tuning LLaMa with only 1000 curated prompts and responses, demonstrates strong performance when compared to similar models with a much bigger set of prompts used for alignment.
#nlproc
#llms
#python
#opensource
3
12
57
In 2024 we're going to build some awesome open datasets for the AI community.
Our focus will be on instructions and preference datasets for DPO/RLHF
❓What should we build?
8
9
52
🔥 Community and Data Quality Are More For Alignment
A recipe to replicate SPIN with 30x less data:
🗣️ 50K samples vs 1.8K prompts curated by the 350+ amazing DIBT contributors.
⚗️ Distillation of
@MistralAI
Large instead of OpenAI
🙌 Open data & code with ⚗️distilabel
👇
1
19
53
🦦We present CapybaraHermes-2.5-Mistral-7B, a model trained with the capybara-dpo dataset, built with ⚗️ distilabel. It's a preference-tuned OpenHermes-2.5-Mistral-7B.
#nlproc
#llms
#python
#opensource
2
5
52
⚗️ Preference data is the key ingredient for DPO/RLHF pipelines
distilabel implements generation and labeling focusing on scalability and SoTA methods like UltraFeedback or JudgeML
Get started and help us shape the project:
#rlhf
#rlai
#opensource
3
14
52
🎉⚗️ distilabel 0.5.0 is out!
Packed with new features and enhancements to build and improve LLM fine-tuning datasets with the latest techniques.
Let's see what's inside this release!
🔥 Deita components for evolution and automatic data selection of instruction tuning data
🧵
2
13
52
What is a text2text model?
In simple words: a model which given a text returns another text
Text summarization is one of such models
This is how to use the scitldr, summarization dataset by
@allen_ai
👇
More task examples:
#nlproc
#python
#opensource
0
18
50
🚀 Introducing Rubrix Weak Labeling 🚀
⚡️ Supercharge your NLP data annotation with interactive weak supervision
👩🔬 Leverage the latest weak supervision methods
👩🏭 Production-ready and open-source
#nlproc
#datacentricAI
#opensource
#python
0
20
49
📰 Last minute update:
With 54% less data & decontaminated gsm8k training prompts, distilabeled Hermes highly improves GSM8K results.
Second time we see this with DPO: last time was
@winglian
on TruthfulQA using
@argilla_io
UltraFeedback cleaned.
🔥 More is less for DPO, high quality matters!
📢 Dropping our first open dataset and LLM of the year:
💾Meet distilabel Orca Pairs DPO, an improved version of the now famous dataset from
@intel
🏛️And a new OpenHermes model outperforming baselines with 54% less DPO pairs
🧵
5
46
231
1
6
48
Do you use
@FastAPI
for serving NLP models? Monitor data & predictions with Rubrix
Unlock data, prediction & model monitoring for text classification and NER
Rubrix, open-source framework for data-centric NLP 👉
0
6
49
📢 distilabel 0.3 is out!
🎉Welcome 3 LLM engines for generation & DPO labeling:
✔️
@GoogleAI
VertexAI: Gemini, PaLM, Codey & custom deployments
✔️
@anyscalecompute
&
@togethercompute
OSS LLM inference
💾 plus a new checkpointing method
🧵👇
2
10
49
Rubrix: Python framework for data-centric NLProc
🎉 New release 0.5.0
Support for text2text tasks (text summarization, OCR post-processing & many more)
Kudos
@abdrahman_issam
& other new contributors
#python
#opensource
#NLProc
1
18
47
🔥 Top 3 trending datasets for DPO built with ⚗️ distilabel, curated with
@argilla_io
Let's publish some more next week!
What should we build next?
1
7
46
🔥 Zero-shot then few-shot with SetFit 🔥
SetFit + Argilla + Vector search is a game changer in terms of speed and the ability to go from no labels to a decent dataset and model in just some iterations.
0
9
45
Welcome Zephyr 141B to Hugging Chat🔥
You gotta love its system prompt 🤩🤩🤩🤩
Welcome Zephyr 141B to Hugging Chat🔥
🎉A Mixtral-8x22B fine-tune
⚡️Super fast generation with TGI
🤗Fully open source (from the data to the UI)
12
85
386
1
7
45
Can the community build impactful datasets in just 10 days?
Yesterday we shared the code matching SPIN (Self-Play Fine Tuning) with 30x less data!
Big win of quality vs quantity
This with 10K rated prompts, rate some prompts to see where 20K gets us 👇
2
7
44
Yesterday, we released Notux 8x7b v1, a Mixtral 8x7B Instruct v0.1 fine-tune using a second iteration of DPO and our cleaned and binarized version of the UltraFeedback dataset.
Today, we release a new space to chat with it 🚀
1
9
43
🥁 Launching a new dataset: Capybara-Preferences, built with distilabel 1.0 ⚗️!
Hard at work fine-tuning Llama 3?
Here's the dataset you've been waiting for.
Initial results with ORPO & this dataset are 🔥
🧵What makes this dataset so special?
2
13
43
🤔 HELM: A holistic framework for evaluating foundation models.
Partnering with some of the most influential AI companies, several researchers from Stanford University have presented the (HELM) framework, to improve the transparency of language models.
1
12
39
🚀Training high-quality models from Argilla just got a lot easier
Use Argilla +
@huggingface
AutoTrain to train NLP models without a single line of code.
Don't wait for your LLaMA weights and start building powerful
#opensource
NLP models!
1
9
42
🌀 We're taking 10k_prompts_ranked for a SPIN & the results are 🤯
What's 10k_prompts_ranked?
What's SPIN?
Stay tuned for another small open data win!
In the meantime, discover the dataset and contribute!
1
8
42
🔂 We ran Self-Play fIne-tuNing (SPIN) on the DIBT prompt collective data so we figured writing a blog with
@mantisnlp
would be an awesome way to get the community ready for yet another awesome model release that highlights the need for high-quality data.
0
10
41
The answer is yes and the result is:
distilabeled Hermes 2.5, a model fine-tuned on top of the amazing OpenHermes by
@Teknium1
.
Unlike other DPO fine-tunes is trained with only 6K examples.
It outperforms
@maximelabonne
's NeuralHermes with the same recipe but 54% less samples
1
3
40
🧙 Create an evol-instruct dataset with distilabel
In this tutorial, we develop an evol-instruct dataset by employing the approaches outlined in and using distilabel.
#nlproc
#llms
#python
#opensource
1
5
40
Always wanted to gather user feedback from your
@Gradio
apps for data labeling?
Check this example Space
1. Users can flag specific responses from the sentiment classifier
2. Flagged predictions are logged into Argilla for labeling & validation
🧵 👇
1
12
38
Do you use
@FastAPI
for serving NLP models? Monitor data & predictions with Rubrix 👉
📓 Log input data and predictions to detect live issues.
📈 Collect labelled data to evaluate models in productions.
#mlops
#nlp
#python
#opensource
1
12
36
Phi-2 is now open source!
Microsoft has recently changed the license of their phi-2 model from research to MIT, it can be used for commercial purposes now .
#nlproc
#llms
#python
#opensource
2
1
38
fast few-shot learning 🫱🏾🫲🏼 active learning
Active learning with classy-classification and
@argilla_io
You can now run this tutorial on Google Colab and Hugging Face Argilla Spaces
#nlp
#opensource
#python
#argilla
#nlproc
0
11
37
🔥 Zero-shot then few-shot with SetFit 🔥
SetFit + Argilla + Vector search is a game changer in terms of speed and the ability to go from no labels to a decent dataset and model in just some iterations.
1
9
37
🆕 Open replication of
@cohere
's "Replacing Judges with Juries" using distilabel!
🧑💻 Post by
@alvarobartt
:
📄 Paper:
1
10
36
We will say it once again:
Contributing open datasets is one of the most impactful ways to accelerate OSS AI
Thrilled to see the community building at this pace
🌟 ORPO, a technique that replaces SFT+DPO/PPO was released recently. I saw
@_philschmid
's post regarding it yesterday. Gave ORPO a shot with phi-2 and
@argilla_io
dpo-mix-7k.
Model:
Try out LazyORPO (Automated):
0
6
64
2
8
37
💬 Did you know you can finetune an LLM, Mistral 7B in particular, on a chat-style instruction dataset? This Argilla tutorial has a step-by-step guide!
#nlproc
#llms
#python
#opensource
2
3
35
🌟 Exciting News! 📷 We've been featured in
@CBinsights
prestigious top 100 AI company list for 2024! 📷📷 It's an honor to be recognized among the companies driving innovation and shaping the future of AI .✨
0
10
34
Yesterday, we launched a new feature to enable sign in with
@huggingface
into
@argilla_io
spaces.
We also launched a collective initiative to collect high-quality prompts.
In just a few hours: ~80 contributors and more 1500 data points!
Join us!
1
11
34
Starting a project with little labelled data?
Tutorials for few and zero shot classification 👇
#python
#argilla
#nlproc
#opensource
1
10
33
Did you know that Argilla and distilabel datasets have over 6 million hub downloads on the Hub? 🤯
Now, distilabel datasets will be even easier to identify thanks to the new icon added to the
@huggingface
Hub—a nice addition to yesterday's release!
1
10
34
📚 📚 Part 6: Identity Preference Optimization (IPO) of our collab with
@mantisnlp
. Can IPO be a valid alternative to DPO? Don't worry we will soon cover an overview of KTO and ORPO as well 🤓
1
6
32
🚀So excited to see the progress of open-weights models these past days!
At
@argilla_io
we build open datasets so people can build better models.
Our datasets now power hundreds of open models.
Here's a recap of the datasets we've shared in just a few months
1
8
33
We used distilabel, our open-source AI Feedback framework to build a preference dataset with ratings for each pair and natural language critiques.
It took around 3 hours to build the full dataset and just a few lines of code.
🧵
2
3
33
💨 Argilla - Notus 7B v1 is already quantized thanks to the awesome TheBloke!
You can find them in the 🤗 Hub at:
- GGUF
- AWQ
- GPTQ
#nlproc
#llms
#data
#quality
#opensource
1
9
32
From few-shot to production datasets:
Label, train, and predict using SetFit!
Iterate on your NLP tasks with the new Argilla training module
Supports SetFit, Transformers, and spaCy.
2
10
32
🚀 Building open datasets is the best way to contribute to Open AI
Let's keep pushing!
The best performing 7B model on the Open LLM leaderboard:
> a merge of a model DPO'd with
@argilla_io
distilabel orca pairs
> with an additional DPO fine-tuning using the same dataset
🐶 NeuralBeagle14-7B
It's the best-performing 7B parameter model on the Open LLM Leaderboard.
Remarkably, it also ranks as the 10th best-performing model overall on the Open LLM Leaderboard. In just 7B parameters!
Merge + DPO = profit
16
47
360
4
5
30
🎉 New release 0.13.0 🎉
🗂 Multi-label weak supervision
🤗 Read ANY dataset from the Hugging Face Hub
👥 Redesigned team workspace for team collaboration
#python
#opensource
#nlproc
1
11
29
⚗️ distilabel v.0.6.0 released 🎉
> JSONOpenAILLM to get JSON responses from generation models
> HF InferenceEndpoints to run Free/Pro API endpoints
If you have a HF Pro account, you can now generate and label with Mixtral!
Release notes:
1
7
29
Fast few-shot classification with active learning to speed up data labelling ❤️
Classy-classification few-shot learning within an active learning loop
Tutorial:
#machinelearning
#opensource
#python
#argilla
#nlproc
1
9
28
🚀We're thrilled by the progress we've made so far
We can reach the goal of annotating 10,000 prompts by the end of this week!
Once reached, we'll publish v1 of the dataset
Do you have an account on
@huggingface
? It takes 5 seconds to contribute
👇
2
10
28
Do you want to improve OSS LLMs for your domain or language?
We can support you running a community effort!
Join our first cohort. It's fun and you get to work with amazing people from
@huggingface
like
@vanstriendaniel
and from
@argilla_io
like
@alvarobartt
👇👇
The Data is Better Together Discord channel is already engaging in discussions about some incredibly important and interesting use cases. If you're interested in contributing to the creation of better datasets, we welcome you to join us!
Learn more here:
0
4
15
0
8
29
🔥 distilabel orca pairs is trending on
@huggingface
> 1k downloads in less than a week
17 models using it!
If you're doing dpo try it out!
1
4
29
👋 Twitter
I'm Rubrix, a free and open-source Python framework to explore, label, and monitor data for natural language processing.
Follow this account for updates and learning materials about practical NLP
#python
#opensource
#nlproc
0
15
28
We set up a distributed annotation for RLHF with our open LLM CrowdEval and the newly announced persistent storage for Hugging Face Spaces.
Claim your login:
Contribute here:
About HF persistent storage:
3
10
29
🤯 Labeling data without leaving Google Colab?
1⃣ Pick your favorite Argilla tutorial
2⃣ Use the Open in Colab button
3⃣ Deploy Argilla using
@huggingface
Spaces
4⃣ Use %%html to embed Argilla using the iframe shown behind the Embed this space button.
🚀 Start labeling
0
7
28
🗂️ Multi-label text classification weak labeling
Get started with this brand new feature with this tutorial by
@vid_algo
#python
#opensource
#datacentricai
0
17
27
👩🏼🏫 Have you heard about Self-Instruct? This framework allows bootstrapping the ability of LLMs to generate their own instructions. You can try it out in distilabel!
Argilla's GitHub:
distilabel:
#nlproc
#llms
#python
#opensource
0
11
26
Do you have a
@huggingface
account? It takes 5 seconds to contribute to OSS AI 👉
We are running a collaborative effort to annotate prompts and improve data for future OSS LLMs.
📱You can even do it from your phone!
🗣️ Every voice counts!
1
18
26
🎉 Big news!
We raised $1.6 Million to Transform Data Labeling for NLP, co-led by
@ZettaVentures
and
@CaixaCR
0
4
26
🐑 Learn how to curate instruction datasets & scale up your annotation team with our new tutorial, guided by our work on the Dolly Dataset by Databricks
#AI
#LLMs
#Opensource
0
9
26
🎉 New release 0.6.0
Tons of
#UX
improvements
Early support for weak supervision with any method (Snorkel, Flyingsquid)
Metrics for robust NLP
Kudos
@_Enamya
,
@abdrahman_issam
& our new contributors
#python
#opensource
#NLProc
#datacentricAI
0
9
24
🎉Today we crossed 900 stars on GitHub 🌟
If you don't know Rubrix yet 👉🏽
Thanks everyone for your support!
We're just getting started.
Follow us
@rubrixml
#Python
#opensource
#NLProc
1
10
25