Varun Agarwal

New at PyTorch Conference - Fine-Tuning Mini-Summit: https://hubs.la/Q02M6Bxc0 As open-source LLMs have become more capable, a thriving ecosystem has emerged around fine-tuning these models. This mini-summit brings together researchers, developers, practitioners, and hobbyists to discuss cutting-edge topics in LLM fine-tuning. Highlights include: - Keynote by Joe Spisak on the Llama ecosystem - Talks from founders of Axolotl, Unsloth, and torchtune - Expert panel featuring Tim Dettmers (bitsandbytes, QLoRA), Hailey Schoelkopf (LM Eval Harness, EleutherAI), Aakanksha Chowdhery (PaLM, Gemini), and Alexis Conneau (OpenAI) Explore crucial areas like: - Memory efficiency - Parameter-efficient fine-tuning - Quantization - Performance at scale - Reproducible evaluations Attend the Fine Tuning Mini-Summit for FREE September 18 with your PyTorch Conference registration: https://hubs.la/Q02M6Bl_0 #PyTorchConf #PyTorch #PyTorchFoundation #AI #MachineLearning #ML #DeepLearning #OpenSource #OpenSourceSoftware #OpenSourceDevelopment #OpenSourceCommunity #OSS #LinuxFoundation #events #LLM #NLP

43

Looking to optimize your RAG retrieval system but short on time? We've got you covered! 🙌 Our team has created a video summary of our article on evaluation of RAG Retrieval Chunking Methods. In just one minute, you'll discover insights and pro tips you need to boost your RAG retrieval performance. 📈 🔍 Key highlights: - The top-performing chunking method that outshined the competition by 10% - The surprising effectiveness of the TinyBERT reranker - Why sentences are the ideal units for chunking - Actionable strategies for maximizing your RAG retrieval pipeline Whether you're a seasoned professional or just starting out, this video is packed with valuable information to help you stay ahead of the curve. 🌟 📹 Watch the video: https://buff.ly/4b2CqtQ 📄 Read the full article: https://buff.ly/4aZt5TF #RAGRetrieval #DataScience #MachineLearning #Efficiency #Optimization #Superlinked

1

At Pareto.AI, we've helped numerous customers maintain language model accuracy and prevent model drift by leveraging high-quality human training data. But... what is model drift anyway? 🤔 Imagine you’re playing a sport you’re great at, but suddenly, the rules change without you knowing. Your strategies that once led you to victory now lead to confusion and mistakes. Just like needing to learn the new rules and adapt, language models require regular adjustments to stay accurate. Model drift happens due to shifts in underlying data patterns, changes in user behavior, or emerging trends. To combat this, models require continuous monitoring, retraining with fresh data, and sometimes even structural updates to their algorithms. This week, we decided to write a blog on this very important topic. Learn how to spot and fix model drift in our latest blog post below! #MachineLearning #DataScience #ModelDrift #AI #ParetoAI #RLHF

28

1 Kommentar

This week marks the first stop on the second half of the Databricks Data + AI World Tour in Munich! 🌍 Check out this blog, written by Matt Turner, which reflects on some highlights from the previous stops in London and New York, and what we’re looking forward to over the next month in Munich, Chicago, Amsterdam, and Atlanta! 🚀 Read the blog here 🔗 https://lnkd.in/e9ZYdTGM #Prophecy #Databricks #DAIS #WorldTour #AI #DataTransformation

17

Run experiments to learn... then learn from all the experiments you ran! Throughout the summer we’ve been rolling out a series of new views that enable meta-analysis across your corpus of experiments 🧐 Have you checked out the views we've shipped? 1. Experiment timeline 2. Metric batting average 3. Metric correlation #experimentation #datascience

25

Join PyTorch at Snowflake's #DataCloudDevDay on June 6 in San Francisco to learn how to accelerate GenAI performance with Native PyTorch. Register: https://hubs.la/Q02xQStt0

18

Models like OpenAI's CLIP combine text and images for search, but they have limitations, particularly with longer texts and complex relationships. This article talks about JINA's new model that improves on CLIP by integrating both text and image embeddings in one model, making it faster and more efficient for multimodal search tasks. Experiments and Findings: → Search with Both Text and Images: Combining both embeddings leads to better and more diverse search results, improving relevance and variety in results. → Image-Driven Diversification: By using images for diversification, Jina-CLIP v1 can present a broader range of results, offering users more varied options. Results Comparison: → Using both text and image embeddings (via averaging) consistently outperformed searches based solely on text or images, providing more precise and diverse retrieval across both broad and fine-grained categories. https://lnkd.in/guNRQpBU

15

Data Scientists, This is BIG! Snowflake's Arctic LLM is Here – and Open-Source! 🤯 Get your hands on it now: Interactive Demo: Explore Arctic's capabilities with this Streamlit app → https://lnkd.in/gJ7y38Ng Replicate: Ready to experiment? Find it on Replicate → https://lnkd.in/gyDXiMju Why This Matters: Arctic lets you quickly build demos of your models. This is a game-changer for showcasing your work and making your models more accessible! Huge thanks to Snowflake for making this happen! Are you excited to try Arctic? Let me know in the comments! 👇 #LLM #OpenSource #DataScience #Streamlit #Snowflake

8

Meet Alex Lau, a Data Scientist here at Selector 👋 Alex has been working on expanding Selector's range of machine learning models, targeting ones tailored to different types of network infrastructure and metrics. Machine learning models help drive many of Selector's cutting-edge features, such as anomaly detection and root cause analysis. Being at Selector puts Alex at the forefront of machine learning within the networking space. 📹 Watch Alex's short video to learn more about his machine learning innovations here at Selector. #machinelearning #network #innovation #datascience #Selector #SelectorAI

38

1 Kommentar

In data partitioning, there are no universal rules. Tailor your approach based on the specific characteristics of your data and the problem at hand for optimal results. Check out Federico Rucci's latest article on creating appropriate test sets. #MachineLearning #DataScience

16

🎙️ New Episode: Klaviyo's Data Science Podcast🎙️ https://lnkd.in/gE5UqxeY Wish you could give your younger self some advise? 4 senior members of our data science team sit down to reflect on what they know now and what they wish they'd know earlier in their career: 🎙️ How taking a more scenic or indirect career route can impart valuable experiences 🎙️ The growth and development opportunities that truly unlocked new phases of their careers, and how you can make the most of similar opportunities 🎙️ Why mistakes often make the best learning opportunities

5

Running pandas at scale with slow single-threaded execution or painful out-of-memory errors? Snowflake is thrilled to announce public preview support of Snowpark pandas API, enabling seamless and secure execution of distributed pandas in Snowflake. Just by changing the import statement and a few lines of code users will be able to get the same pandas-native experience they know and love at Snowflake speed, governance and scale.

20

2 Kommentare

NEW BLOG POST 🧠 How to create embeddings with Sentence-Transformers in PySpark. #pyspark #sentencetransformers #nlp #generativeai #genai #datascience #datascientist #deeplearning https://lnkd.in/dMvgSDwP

3