Muhammad Qori Ramadhan Nasution’s Post

Why Use MLflow? 🚀 MLflow simplifies tracking, managing, and sharing machine learning experiments. 🌟 Track Machine Learning Model with MLflow! 🌟 Discover how to efficiently track your model training processes directly within Jupyter Notebooks using MLflow. Check out this comprehensive guide from Microsoft to get started: Track Model Training in Jupyter Notebooks with MLflow 💻📊 For knowing more click here [https://lnkd.in/dAnbn7GU]

In the past week of my Digital Skola program, I've gained a tremendous amount. It's been a whirlwind of learning and growth, especially regarding the basics of Airflow. After a brief introduction in the prior week, I delved into important concepts like Airflow connections, variables, XComs, the REST API, data backfill, and advanced operations. A major highlight was the practical experience of deploying Airflow with PyPI and integrating it with Google BigQuery. This hands-on application solidified my grasp of Airflow's potential in handling intricate data workflows. Furthermore, in our final Airflow session, we participated in a practical lab where we incorporated Airflow with Spark using Docker. This provided valuable knowledge on managing and running data workflows within a containerized environment. These experiences have significantly improved my skillset and influenced how I approach problem-solving in both my professional and personal life. Don't forget to look at the overview slides my colleagues and I created for a comprehensive look at our recent learning experiences. We're eager to share our journey with you and keep you updated in the future. Stay connected! #StudiIndependenBersertifikat #SIBDigitalSkola #DigitalSkola #DataEngineer

Just finishing cohorts 5 from Machine Learning Zoomcamp, even though I joined late and this is my first cohorts 😁 In this cohorts, I learn about loading machine learning model using picke and serve it using flask. I also learn how to deploy the flask app to the docker and request prediction from the app that running in the docker. If you want to try to solve the cohorts also, visit the machine learning zoomcamp repository https://lnkd.in/gT8f5CWV #mlzoomcamp

🎯 Experimenting with Mage to build a Machine Learning workflow 🛠 During the last few days, I've been testing and playing around with Mage, building a machine learning workflow, reusing code blocks from previous experiments, and fitting the pieces together to create a reliable and efficient flow. You can use data loaders to load your dataset from many sources, the local disk, a S3 or GCP bucket, from a database, a streaming source, and many more. If you already had a data load block from a previous project, that is compatible with this new task, you can simply reuse it in your new pipeline. 🔄 Subsequently, you can generate one or more transformer blocks to prepare your data and generate the necessary features. Once the data is ready, you can save it in a remote storage to share it with another team and reuse it in the next pipeline such as the model training pipeline. 🚀 By generating a new pipeline for the training phase, we can build blocks with the training code, which we can use in similar projects. In addition, you can rely on the dynamic blocks to perform the hyperparameter tuning stage. These blocks allow you to run the same code on different input parameters. In this way, you could generate a training for model A and X parameters, another for model B with Y parameters, and so on. 🌈 There are still more features and parts to assemble, but without a doubt, Mage allows you to speed up these tasks with little effort. You can also join their discord and seek information and support to solve your doubts. 🙌 Thanks to DataTalksClub, we continue exploring the keys to a reliable MLOps process. #mlopszoomcamp #mlops #machinelearning #mage

Day 13 of 100 Days of Learning with I4G One thing that stood out in my course of learning today was been able to manipulate the pivot table. Few days back, when I started learning about pivot table, it was hard for me to carry out exercises without so much error but today, the error code was less. I executed my tasks with less number of trial and that stood out for me. #DataAnalysis #DataCamp #IngressiveForGood #100DaysOfLearningWithI4

Hey Data Learners! I'm excited to update you on Week 6 of my Data Engineering course at Digital Skola! This week was all about diving into Spark and Airflow, two crucial tools for data processing and workflow management. Here are some of the key points we covered: - Getting Started with Spark - Understanding Spark Architecture - Working with Structured Operations, Schemas, and the Structured API - Introduction to Airflow - Exploring Airflow Architecture - Navigating the Airflow UI - Understanding DAG Elements in Airflow - Scheduling with Airflow - Managing Connections and Variables in Airflow - Using XComs - Exploring the Airflow REST API - Backfilling Data A big thank you to our class representative, Vinitiara Ningrum, for the incredible support and guidance throughout this learning journey. Thanks for following along! If you're interested in data and want to keep up with my journey, connect with me for more updates on my path to becoming a Data Engineer. 😊🔗 #DigitalSkola #LearningJourney #DataEngineering

Just finished the course “Machine Learning Fundamentals for Healthcare”! #machinelearning

🚀 Learn Machine Learning with DataFlair! 🚀 Curious about Machine Learning? DataFlair is the perfect place to start! Visit :-https://lnkd.in/dMcNG_22 💡 Why DataFlair? ▶ Expert Instructors: Learn from professionals. ▶ Practical Projects: Get hands-on experience. ▶ Flexible Learning: Study at your own pace. ▶ Community Support: Connect with fellow learners.