{"course":{"productid":34489,"modality":6,"active":true,"language":"fr","title":"Enhancing Data Science Outcomes With Efficient Workflow","productcode":"EDSOEW","vendorcode":"NV","vendorname":"Nvidia","fullproductcode":"NV-EDSOEW","courseware":{"has_ekit":false,"has_printkit":true,"language":""},"url":"https:\/\/portal.flane.ch\/course\/nvidia-edsoew","objective":"<ul>\n<li>Develop and deploy an accelerated end-to-end data processing pipeline for large datasets<\/li><li>Scale data science workflows using distributed computing<\/li><li>Perform DataFrame transformations that take advantage of hardware acceleration and avoid hidden slowdowns<\/li><li>Enhance machine learning solutions through feature engineering and rapid experimentation<\/li><li>Improve data processing pipeline performance by optimizing memory management and hardware utilization<\/li><\/ul>","essentials":"<ul>\n<li>Basic knowledge of a standard data science workflow on tabular data. To gain an adequate understanding, we recommend this article.<\/li><li>Knowledge of distributed computing using Dask. To gain an adequate understanding, we recommend the &ldquo;Get Started&rdquo; guide from Dask.<\/li><li>Completion of the DLI&rsquo;s Fundamentals of Accelerated Data Science course or an ability to manipulate data using cuDF and some experience building machine learning models using cuML.<\/li><\/ul>","outline":"<p><strong>Introduction<\/strong>\t\n<\/p>\n<ul>\n<li>Meet the instructor.<\/li><li>Create an account at courses.nvidia.com\/join<\/li><\/ul><p><strong>Advanced Extract, Transform, and Load (ETL)<\/strong>\t\n<\/p>\n<ul>\n<li>Learn to process large volumes of data efficiently for downstream analysis:<ul>\n<li>Discuss current challenges of growing data sizes.<\/li><li>Perform ETL efficiently on large datasets.<\/li><li>Discuss hidden slowdowns and perform DataFrame transformations properly.<\/li><li>Discuss diagnostic tools to monitor and optimize hardware utilization.<\/li><li>Persist data in a way that&rsquo;s conducive for downstream analytics.<\/li><\/ul><\/li><\/ul><p><strong>Training on Multiple GPUs With PyTorch Distributed Data Parallel (DDP)<\/strong>\t\n<\/p>\n<ul>\n<li>Learn how to improve data analysis on large datasets:<ul>\n<li>Build and compare classification models.<\/li><li>Perform feature selection based on predictive power of new and existing features.<\/li><li>Perform hyperparameter tuning.<\/li><li>Create embeddings using deep learning and clustering on embeddings.<\/li><\/ul><\/li><\/ul><p><strong>Deployment<\/strong>\t\n<\/p>\n<ul>\n<li>Learn how to deploy and measure the performance of an accelerated data processing pipeline:<\/li><li>Deploy a data processing pipeline with Triton Inference Server.<\/li><li>Discuss various tuning parameters to optimize performance.<\/li><\/ul><p><strong>Assessment and Q&amp;A<\/strong><\/p>","summary":"<p>Learn how to create an end-to-end, hardware-accelerated machine learning pipeline for large datasets. Throughout the development process, you&rsquo;ll use diagnostic tools to identify delays and learn to mitigate common pitfalls.<\/p>\n<p><em>Please note that once a booking has been confirmed, it is non-refundable. This means that after you have confirmed your seat for an event, it cannot be cancelled and no refund will be issued, regardless of attendance.<\/em><\/p>","objective_plain":"- Develop and deploy an accelerated end-to-end data processing pipeline for large datasets\n- Scale data science workflows using distributed computing\n- Perform DataFrame transformations that take advantage of hardware acceleration and avoid hidden slowdowns\n- Enhance machine learning solutions through feature engineering and rapid experimentation\n- Improve data processing pipeline performance by optimizing memory management and hardware utilization","essentials_plain":"- Basic knowledge of a standard data science workflow on tabular data. To gain an adequate understanding, we recommend this article.\n- Knowledge of distributed computing using Dask. To gain an adequate understanding, we recommend the \u201cGet Started\u201d guide from Dask.\n- Completion of the DLI\u2019s Fundamentals of Accelerated Data Science course or an ability to manipulate data using cuDF and some experience building machine learning models using cuML.","outline_plain":"Introduction\t\n\n\n\n- Meet the instructor.\n- Create an account at courses.nvidia.com\/join\nAdvanced Extract, Transform, and Load (ETL)\t\n\n\n\n- Learn to process large volumes of data efficiently for downstream analysis:\n- Discuss current challenges of growing data sizes.\n- Perform ETL efficiently on large datasets.\n- Discuss hidden slowdowns and perform DataFrame transformations properly.\n- Discuss diagnostic tools to monitor and optimize hardware utilization.\n- Persist data in a way that\u2019s conducive for downstream analytics.\nTraining on Multiple GPUs With PyTorch Distributed Data Parallel (DDP)\t\n\n\n\n- Learn how to improve data analysis on large datasets:\n- Build and compare classification models.\n- Perform feature selection based on predictive power of new and existing features.\n- Perform hyperparameter tuning.\n- Create embeddings using deep learning and clustering on embeddings.\nDeployment\t\n\n\n\n- Learn how to deploy and measure the performance of an accelerated data processing pipeline:\n- Deploy a data processing pipeline with Triton Inference Server.\n- Discuss various tuning parameters to optimize performance.\nAssessment and Q&A","summary_plain":"Learn how to create an end-to-end, hardware-accelerated machine learning pipeline for large datasets. Throughout the development process, you\u2019ll use diagnostic tools to identify delays and learn to mitigate common pitfalls.\n\nPlease note that once a booking has been confirmed, it is non-refundable. This means that after you have confirmed your seat for an event, it cannot be cancelled and no refund will be issued, regardless of attendance.","skill_level":"Beginner","version":"1.0","duration":{"unit":"d","value":0.5,"formatted":"0,5 jours"},"pricelist":{"List Price":{"US":{"country":"US","currency":"USD","taxrate":null,"price":500},"DE":{"country":"DE","currency":"EUR","taxrate":19,"price":500},"AT":{"country":"AT","currency":"EUR","taxrate":20,"price":500},"SE":{"country":"SE","currency":"EUR","taxrate":25,"price":500},"SI":{"country":"SI","currency":"EUR","taxrate":20,"price":500},"GB":{"country":"GB","currency":"GBP","taxrate":20,"price":420},"IT":{"country":"IT","currency":"EUR","taxrate":20,"price":500},"CA":{"country":"CA","currency":"CAD","taxrate":null,"price":690}}},"lastchanged":"2025-07-29T12:18:27+02:00","parenturl":"https:\/\/portal.flane.ch\/swisscom\/fr\/json-courses","nexturl_course_schedule":"https:\/\/portal.flane.ch\/swisscom\/fr\/json-course-schedule\/34489","source_lang":"fr","source":"https:\/\/portal.flane.ch\/swisscom\/fr\/json-course\/nvidia-edsoew"}}