- 68
- 93 125
Unit8
Switzerland
Приєднався 21 жов 2019
Unit8 is a Swiss data & AI services company focusing on turning data into value for non-digital native industries.
We are committed to closing the AI knowledge gap and we'll continue to share our best practices / lessons learnt from having delivered 50+ AI/ML/data science projects.
We're based out of Lausanne (Switzerland) with offices in Zurich and Krakow (Poland)
#unit8 #data #AI #ml #datascience
We are committed to closing the AI knowledge gap and we'll continue to share our best practices / lessons learnt from having delivered 50+ AI/ML/data science projects.
We're based out of Lausanne (Switzerland) with offices in Zurich and Krakow (Poland)
#unit8 #data #AI #ml #datascience
Harnessing the Power of Forecasting: Best Practices and Use Cases with Darts
Join Dennis Bader, Unit8 Senior Data Scientist and Darts Lead, at the Swiss Data Science Conference 2024 as he delves into the power of forecasting.
Learn about essential best practices, use cases in various sectors like energy and retail, and how our open-source Python library facilitates time series forecasting and anomaly detection.
Dennis also shares insights from real-world projects and reveals how even simple models can be highly effective, along with the potential of neural networks for efficient and accurate forecasting. Don't miss the opportunity to enhance your forecasting skills and understand its business impact.
Thank you to #SDS2024 for the video!
And if you want to learn more about Darts - visit our website: unit8.com/darts-open-source/
Learn about essential best practices, use cases in various sectors like energy and retail, and how our open-source Python library facilitates time series forecasting and anomaly detection.
Dennis also shares insights from real-world projects and reveals how even simple models can be highly effective, along with the potential of neural networks for efficient and accurate forecasting. Don't miss the opportunity to enhance your forecasting skills and understand its business impact.
Thank you to #SDS2024 for the video!
And if you want to learn more about Darts - visit our website: unit8.com/darts-open-source/
Переглядів: 346
Відео
Azure OpenAI service made easy by Unit8 - Part 5: Query Translator
Переглядів 1086 місяців тому
Modernizing SQL Queries with Generative AI: PySpark Translation Demo In this video, we demonstrate an effective approach to modernizing old SQL database queries by translating them into new platforms using generative AI. The demo showcases the conversion of typical and complex SQL queries into PySpark code. The process involves leveraging human expertise and well-engineered prompts to ensure th...
Unit8 GenAI Accelerator
Переглядів 7928 місяців тому
Setting the right foundation for generative AI use-Cases is essential for quickly iterating and delivering value. Unit8 GenAI Accelerator deploys the infra and code on Azure that powers your GenAI use-cases. The solution runs fully within your Azure subscription providing full control and customization options: - Configurable OpenAI or Open-source models, - Secured environment following best-pr...
Leveraging Time Series Analysis for Arrhythmia Detection in Healthcare using Darts
Переглядів 445Рік тому
Explore the vital role of time series analysis in healthcare for arrhythmia detection using ECG signals, and learn how Darts can optimize this process. In the digital era, healthcare is rapidly transforming with the integration of advanced analytical techniques. Among these, time series analysis stands out as a key player-its applications range from resource allocation and diagnostics to the li...
Unit8 - Get to know us!
Переглядів 487Рік тому
For those of you who don't know us, we are a leading Swiss data, analytics & AI consulting and services company. We are 110 engineers, data scientists, and executives of more than 15 nationalities. Join us on our journey to help non-digital native companies turn data into value. Learn more: unit8.com/
Getting started with your first Generative AI Use Case - Mostafa Ajallooeian at Intelligent Health
Переглядів 242Рік тому
Drastically reducing R&D time is GenAI's main footprint in healthcare now, but what about other, smaller yet impactful use cases? In this session, we'll briefly explore GenAI and its growing applications in the healthcare sector. Next, we'll walk through a live demo where we leverage ChatGPT to extract, summarize and reference insights from medical journals. We'll conclude with a healthcare cli...
Foundry Dev Tips #4 - OBJECT Type
Переглядів 163Рік тому
Hey there, Foundry developers! 🎉 We're back with another insightful episode of “ Unit8 Foundry Dev Tips." In this video, Elsemiek is diving deep into the world of object types creation in Foundry and why it's a game-changer for you. Have you ever transformed a dataset and thought, "How can I make this more interactive for users in applications?" 🤔 Well, we've got the answers for you! In this vi...
Azure OpenAI Service made easy by Unit8 - Part 4: Large File Seeker (Multi-Source Answers Demo)
Переглядів 53Рік тому
Sometimes, finding an answer requires compiling information from various sources. Watch as we pose a question about Roche and Nestle's greenhouse gas emission reduction plans. Azure seamlessly retrieves relevant chunks from both companies' documents and delivers a comprehensive summary with fact-checked details. For more info on Generative AI and its capabilities, visit our website: unit8.com/s...
Azure OpenAI Service made easy by Unit8 - Part 4: Large File Seeker (Annual Report demo)
Переглядів 45Рік тому
Using annual reports from major Swiss companies, we illustrate how Azure Cognitive Search efficiently handles detailed and complex questions. Witness how we inquire about Credit Suisse's asset status in 2022 compared to 2021, and Azure provides us with a well-substantiated answer that empowers you with the relevant financial data. For more info on Generative AI and its capabilities, visit our w...
Azure OpenAI Service made easy by Unit8 - Part 4: Large File Seeker (Device Manual Demo)
Переглядів 63Рік тому
Have you ever been stuck with a car problem and dreaded sifting through a lengthy car manual? With our demo, we demonstrate how you can easily upload and index your car's user manual into Azure Cognitive Search. Simply ask your question in natural language, like "Why won't my Prius start in emergency mode?" and see how Azure promptly delivers a validated response, saving you precious time and e...
Swiss Economic Forum Awards - Unit8 pitch by Dr. Marcin Pietrzyk, Co-founder & CEO
Переглядів 246Рік тому
The Swiss Economic Award is a renowned accolade that recognizes companies that demonstrate exceptional innovation, sustainability, and #growth potential. It is a privilege, that Unit8 was nominated as a finalist in the "Services" category, alongside other outstanding companies. Because of that our CEO, Dr. Marcin Pietrzyk had the chance to present the company's mission to the esteemed crowd gat...
Azure OpenAI Service made easy by Unit8 - Part 4: Large File Seeker
Переглядів 155Рік тому
Discover How Azure Cognitive Search Optimizes Information Retrieval from Large Files! In this video, Unit8 experts showcase Azure Cognitive Search capabilities in finding answers within extensive documents. Say goodbye to time-consuming searches and manual data extraction! The video includes: - real-life examples, - how this solution can empower your decisions with accurate data, - a demo of th...
Azure OpenAI Service made easy by Unit8 - Part 3: Realtime Conversation Insights
Переглядів 280Рік тому
Azure OpenAI Service made easy by Unit8 - Part 3: Realtime Conversation Insights
Azure OpenAI Service made easy by Unit8 - Part 2: Field Intelligence Assistant
Переглядів 432Рік тому
Azure OpenAI Service made easy by Unit8 - Part 2: Field Intelligence Assistant
Azure OpenAI service made easy by Unit8 - Part 1: Genuine Search
Переглядів 951Рік тому
Azure OpenAI service made easy by Unit8 - Part 1: Genuine Search
Understand the Data Vault Modelling
Переглядів 1,8 тис.2 роки тому
Understand the Data Vault Modelling
Understand the Data Vault Architecture
Переглядів 9 тис.2 роки тому
Understand the Data Vault Architecture
Optimizing resource consumption & reducing the cost of cloud data platforms - Unit8 Talks #25
Переглядів 2182 роки тому
Optimizing resource consumption & reducing the cost of cloud data platforms - Unit8 Talks #25
AI in Manufacturing: Expectation vs Reality - Unit8 Talks #18
Переглядів 3102 роки тому
AI in Manufacturing: Expectation vs Reality - Unit8 Talks #18
Darts for Time Series Forecasting - Julien Herzen, Francesco Lässig at PyData Global 2021
Переглядів 13 тис.2 роки тому
Darts for Time Series Forecasting - Julien Herzen, Francesco Lässig at PyData Global 2021
Full Data Journey #Hackathon in #Zurich
Переглядів 2,9 тис.2 роки тому
Full Data Journey #Hackathon in #Zurich
Advanced Analytics Trends for 2022, Panel Discussion with Merck and Swiss Re - Unit8 Talks #17
Переглядів 5463 роки тому
Advanced Analytics Trends for 2022, Panel Discussion with Merck and Swiss Re - Unit8 Talks #17
AI in Health and Pharma, a Data & AI Service Provider Perspective - Hidden Champions
Переглядів 1413 роки тому
AI in Health and Pharma, a Data & AI Service Provider Perspective - Hidden Champions
Product Line Optimization - Unit8 Impacts
Переглядів 2443 роки тому
Product Line Optimization - Unit8 Impacts
Darts: Time series made easy in Python
Переглядів 11 тис.3 роки тому
Darts: Time series made easy in Python
AI & Data Marriage Fundamentals by Dr. Marcin Pietrzyk
Переглядів 1973 роки тому
AI & Data Marriage Fundamentals by Dr. Marcin Pietrzyk
thanks a lot !! no way we would have done our hackathon without you
I want to go part way being a bot a one to get a pretty face perfect body nice teeth hair nails.grest body I want I want great breasts and turn into a rabbit in bed lmao think you could be part a one and half of me what iam now I'd like the bots vokab and subjects to spark conversations without scaring them away.i tend to get needy at times prob cause it's been long time since my mom and sister even like me loves yes but wtf is love anyways everything l it's hwavwn a o the list lonely.tommorow get servicecard. 1:09 t
I want to go part way being a bot a one to get a pretty face perfect body nice teeth hair nails.grest body I want I want great breasts and turn into a rabbit in bed lmao think you could be part a one and half of me what iam now I'd like the bots vokab and subjects to spark conversations without scaring them away.i tend to get needy at times prob cause it's been long time since my mom and sister even like me loves yes but wtf is love anyways everything l it's hwavwn a o the list lonely.tommorow get servicecard. 1:09 t
Gents, this is an amazing package. I'm diving in today!
Easy to learn, hard to customize.
Thank you
Interesting intro video, but compromised by poor audio.
What is require feature for anomaly detection?
where exactly is this notebook uploaded?
Thank you for the best webinar. do you provide online AI course in chemistry? waiting for your kind reply.
Can you please publish the notebook? :)
I am a Sr Consultant to a popular Supply Chain firm and, I am happy to say we are building our ML forecast model in our product using Darts
thanks for the video and intrudction, really helpful for beginer
How to undo onehotencoding and add shap values
I am in shock. How those young guys make something complicated so easy to understand and in a interesting way. I really enjoyed the video. Thank you!!!
Gute Erklärung, aber ihr müsst bei euren Videos eine höhere Bitrate verwenden. Es gibt dazu eine Chart auf UA-cam für die Bitrate je Auflösung. Bei 2k sind es z.B. 20Mbits
Can it also be said the anomaly detection model can used to label data that can be used in supervised learning model that can be used to for fraud detection?
Many people believe that darts is for beginners, but I think that it's a really good way for speeding up the workflow for data scientists. Thank you for sharing you hard work.
Please share the code
Great talks! I would like to ask if it is possible to use anomaly detection to detect fraud in ATM transactions with the following features: CardNo, branch-code, AtmID, Trans-date, Amount, Trans-type, Trans-status How can the customer's regular transaction patterns be used to detect anomalies (suspicious fraud)?
Hi @hakeemojulari7392, I dont know if you got an answer to this but its possible. You might want to look into Deep learning architectures called Autoencoders. In theory, if you give an autoencoder the regular customer's data, you can train it to reconstruct every single datapoint. What hapens as a result is that it learns patterns regarding normal transaction behavior, and the reconstruction loss gets smaller and smaller. You can then use the reconstruction loss of the autoencoder to detect suspicious transactions. Intuitively, if a transaction is so new that the autoencoder has never seen before, the reconstruction loss will be higher that a transaction that it has seen before therefore flagging it.
Great, it is scikit learn for ts! Finally
Can you share the code?
It would be great if you could improve your documentation and show a real world project, saving weights, processing real world data and just having some basic tutorials to follow will save hours of people's work.
Great library! Just one question, I know the trained deep learning models can be saved with "my_model.save_model(path)" command, but is there anyway to save it as h5 or pickle file? The reason for this question is that I was using it on jupyterlab and the session expires every 24 hours, so the "my_model" gets lost when I do my_model.load_model(path) in the new session. Thanks!
I love this library. I have some ideas for it
Hello, thank you very much for the video. can you please give out an example of RandomForest with gridsearch and with Covariate. It would help me a lot. thank you.
Darts 💯👍
Great content, and a amazing project!
Hi , Thank for great video. I have a question. Do you think about predicting stock price and how much its confidence? Now I just know using LSTM or XGBOOST to predict stock price
is it working well for predictions ?
Can we participate remotely?
Does Nbeats work for multivariate analysis?? The documentation says yes but there is no example for a multivariate analysis... Can someone help??
Where can I channel my doubts regarding using darts? is there a slack channel or something?
Yes there is a Discord server: discord.gg/Um3jBTYFsA Otherwise in general the best entry point is here: github.com/unit8co/darts
Its indeed a great work, Thankyou. I want to learn how to implement forecasting model for Multivariate Time series. Can anyone give me reference?
This is a great achievement to the community. However, is it possible to use Darts for Multivariate multistep LSTM modelling?
hi i am looking for the same, please tell if you got anything
great !!! it was very helpful
Thanks for sharing these techniques.
Our pleasure Luca!
For those of you who enjoyed the session and want more, we just release a new article about Probabilistic Forecasting in Darts here: bit.ly/3cIdpZr
Thank you for the article! Very informative and interesting! The link for the example notebook in the article is broken. Do you mind pasting the link here? Thank you!
Thank you everyone for joining us! It was a real pleasure. For our next session, we are preparing a special roundtable edition about Advanced Analytics Trends for 2022. Book your seat here: ua-cam.com/video/-nKDJLNGtxg/v-deo.html
Thank you for sharing.
It's a pleasure, hope you enjoyed it!
Huge contribution to the community unit8. Great job. Very well executed. 👊
Thank you, Mark, we will keep sharing the good content! By the way, we are inviting you for our forecasting webinar tomorrow ua-cam.com/video/fWFbFqp8gzY/v-deo.html
Thank you!!
THe past_covariates is giving error for the model. Please make a video on multivaritate forecasting using past covaritates
Can you please show how to do multivariate time series forecasting with exogenous variable using dart
Hi does the Gridsearchcv within darts support multiple time series input? I see multiple time series works fine when I am using Nbeats algorithm object but not with gridsearch
Thanks, more similar videos required for the hands on experience covering different use cases for the time series forecasting
Can we use this "DARTS" for multivariate time series forecasting using GNNs?
yes
Great library. Please tell us how to save the trained model for using dart?
Thanks Léo and Francesco. I've been looking for this kind of tool for a while.
Thank you Mahery! We're enjoying working on darts! Stayed tuned about the new releases/features - github.com/unit8co/darts
great, thank you !!