Kaggle

We’re less than 2 weeks away – don’t forget to register! Join thousands of Kagglers who’ve signed up for our 5-Day Gen AI Intensive Course with Google on Nov 11-15. Find more curriculum info & FAQs on the event website 👇 https://lnkd.in/drwDBhes 

You can now watch the full recording of our AMA with Cohere For AI on YouTube! Thank you Ahmet Üstün and Marzieh Fadaee for giving us a closer look at your new model. 👇 https://lnkd.in/enD7V_Z5

We’re LIVE! 🚨https://lnkd.in/enD7V_Z5

In ~1 hour (11am ET), we're hosting an AMA with Cohere For AI to learn more about their newest model, Aya Expanse. Join us! https://lnkd.in/gv6sVRe5

Looking forward to joining Kaggle tomorrow for an AMA event on Aya Expanse! ✨ Add the event to your calendar, and submit your questions for C4AI Sr Research Scientists Ahmet Üstün and Marzieh Fadaee. https://lnkd.in/giCti9Z2

Excited to see Thad Starner and Sam Sepah on the People of AI Podcast! They've hosted several sign language competitions on Kaggle and are at the forefront of AI-powered accessibility tech. Don't miss this episode. 🙌

𝐁𝐮𝐢𝐥𝐝 𝐚𝐧𝐝 𝐝𝐞𝐩𝐥𝐨𝐲 𝐨𝐧 𝐆𝐂𝐏 𝐰𝐢𝐭𝐡 𝐀𝐩𝐩 𝐄𝐧𝐠𝐢𝐧𝐞 𝐲𝐨𝐮𝐫 𝐜𝐡𝐚𝐭𝐛𝐨𝐭 𝐩𝐨𝐰𝐞𝐫𝐞𝐝 𝐛𝐲 𝐆𝐞𝐦𝐢𝐧𝐢 A couple of weeks ago I published an article about how you can create your own chatbot powered by a Gemma 2 model running locally on your computer. Now, in a new Medium article, I am showing how you can develop and deploy on GCP (using App Engine) a Streamlit chatbot powered by Gemini. In the article I explain how to get your Gemini API key, how to create and cache the connection function in and how to create the simple conversational experience in Streamlit. Then I show how to add and configure your Dockerfile and app.yaml file and perform the deployment using Google Cloud SDK. Full article link: https://lnkd.in/dUHCqujf GitHub project link: https://lnkd.in/dZyYpUsW #Gemini #Chatbot #GCP #GoogleCloudSDK #AppEngine #Streamlit #GDE

🤖 Cohere For AI's Aya Expanse is now available on #KaggleModels! Mark your calendars for October 31st at 11 am Eastern when we’ll bring your questions to Marzieh Fadaee and Ahmet Üstün in an “Ask Me Anything” about the new model, hosted by Kaggle. 👉 Learn more about Aya: https://lnkd.in/ekppvWjA ❓Ask your questions: https://lnkd.in/eGMEAkFe 

Today, we introduce ✨Aya Expanse ✨ – an open-weights state-of-art family of models to help close the language gap with AI. Many communities have been left unsupported due to the language limitations of current models. For the last few years, our fundamental research has sought to change this. Today, we release both the 8B and 32B models as open weights. These models advance the boundaries of multilingual performance. At a larger scale, Aya Expanse 32B outperforms Gemma 2 27B, Mistral 8x22B, and Llama 3.1 70B, a model more than 2x its size. Aya Expanse 8B outperforms the leading open-weights models in its parameter class, such as Gemma 2 9B, Llama 3.1 8B, and the recently released Ministral 8B. We hope this release helps researchers around the world further progress, ask deeper questions about multilingual AI, and develop tools that support global communities. Thank you to all our collaborators, and the Cohere For AI and Cohere team. An idea never becomes a breakthrough unless there is a shared belief we can do something different. Find our open weights on Kaggle and Hugging Face. Read more about Aya Expanse here: https://lnkd.in/enNw3w3h

Have a brilliant idea for a machine learning challenge? Now's the perfect time to bring it to life. Our community competition platform is available at no cost and used by thousands of organizations every year to host hands-on machine learning competitions. With several newly launched features, creating your own Kaggle community competition has never been easier. Here’s a summary of what’s new: Feature #1: ✨Create an AI generated Kaggle competition in 5 minutes This feature lets you pick a ML task and instantly create a new synthetic dataset based off of a source dataset. The new data will mimic the original's patterns without any identifiable info, ensuring a fun, rigorous competition with no risk of participants finding the original data. Learn More: https://lnkd.in/dyFZzWYv Feature: #2: 💸Offer Prizes Up to $10,000 USD Community competition hosts can offer prizes up to $10,000 USD. With this change, we hope community competitions will be more rewarding and that cutting-edge research will reach even more minds.  Learn More: https://lnkd.in/dGjMUwaG Feature #3: 🎯 Write your own Python competition evaluation metric in a Kaggle Notebook Customize your competition to your learning goal. Now you can write your own Python evaluation metric and share it with participants. We've also released 63 new open-source metrics for you to use.  Learn More: https://lnkd.in/du_fGVWq Whether you're an educator, researcher, startup, or simply a passionate data enthusiast, we hope you'll consider hosting a community competition on Kaggle! 👉Get started at https://lnkd.in/dkK2NAC9