Machine learning. Professional - free course from Otus, training 5 months, Date: December 2, 2023.
Miscellaneous / / December 05, 2023
You will consistently master modern data analysis tools and be able to create machine learning models at a professional level. To consolidate your skills with each algorithm, you will carry out a full pipeline of work from preparing the dataset to analyzing the results and preparing for production. The practice and knowledge you will receive will be enough to independently solve classical ML problems and apply for Junior+ and Middle Data Scientist positions.
Portfolio projects
During the course, you will complete several portfolio projects and learn how to competently present the results of your work in order to pass interviews. For your final project, you can take one of the options proposed by the teacher or implement your own idea.
Who is this course for?
For beginning analysts and Data Scientists. The course will help you systematize and deepen your knowledge. You will be able to experiment with approaches, analyze working cases and receive high-quality feedback from experts.
For developers and specialists in other areas who want to change their profession and develop in the field of Data Science. The course will give you the opportunity to build a strong portfolio and immerse yourself in the atmosphere of real-life tasks as a data scientist.
To learn, you will need Python experience at the level of writing your own functions, as well as knowledge of mathematical analysis, linear algebra, probability theory and mathematics. statistics.
Course Features
Best Practices and Trends. Each launch, the program is updated to reflect rapidly changing trends in Data Science. After training, you will be able to immediately start working on real projects.
Important secondary skills. The course includes topics that are usually overlooked, but are necessary for a specialist in everyday tasks and are highly valued by employers:
— building systems for automatically searching for anomalies;
— forecasting time series using machine learning;
— end-to-end pipelines for working with data, ready for implementation in production.
Creative atmosphere and conditions close to real work processes. The entire course is built as a simulator of the everyday working life of a data scientist, where you will have to cope with “dirty” data, calculate your actions in advance, experiment with solutions and prepare models in production In this case, you will need curiosity, perseverance and a thirst for new experiences.
3
courseWorks as a data analyst in the AGI NLP team in Sberbank. Works on neural network language models and their application in real-life problems. Believes that working in the field of Data Science provides a unique...
Works as a data analyst in the AGI NLP team in Sberbank. Works on neural network language models and their application in real-life problems. He believes that working in the field of Data Science provides a unique opportunity to do crazy cool things on the edge of science that are changing the world here and now. Teaches subjects in data analysis, machine learning and data science at the Higher School of Economics. Maria graduated from the Faculty of Mechanics and Mathematics of Moscow State University and the Yandex School of Data Analysis. Maria is currently a graduate student at the Higher School of Economics at the Faculty of Computer Science. Her research interests include data science areas such as natural language processing and topic modeling. Program Manager
3
coursePracticing machine learning and data analysis since 2012. Currently working as Head of R&D at WeatherWell. Has experience in the practical application of machine learning in game development, banking and...
Practicing machine learning and data analysis since 2012. Currently working as Head of R&D at WeatherWell. Has experience in the practical application of machine learning in game development, banking and Health Tech. He taught machine learning and data analysis at the Center for Mathematical Finance of Moscow State University, and was a guest lecturer at the Faculty of Computer Science of the National Research University Higher School of Economics and various summer schools. Education: Economics-mathematics REU im. Plekhanov, Central Faculty of Mathematics and Mathematics of Moscow State University, advanced professional training of the Faculty of Computer Science of the Higher School of Economics "Practical data analysis and machine learning", MSc Computer Science Aalto University Stack/Interests: Python, Machine Learning, Time Series, Anomaly Detection, Open Data, ML for social good
Advanced Machine Learning Techniques
-Topic 1. Introductory lesson. Revisit basic machine learning concepts with a practical example
-Topic 2.Decision trees
-Theme 3.Python for ML: pipelines, pandas acceleration, multiprocessing
-Topic 4.Model ensembles
-Topic 5.Gradient boosting
-Topic 6.Support vector machine
-Topic 7.Dimensionality reduction methods
-Topic 8. Learning without a teacher. K-means, EM algorithm
-Topic 9. Learning without a teacher. Hierarchical clustering. DB-Scan
-Topic 10. Finding anomalies in data
-Topic 11. Practical lesson - Construction of end-to-end pipelines and serialization of models
-Topic 12.Algorithms on graphs
Data collection. Analysis of text data.
-Topic 13.Data collection
-Topic 14.Analysis of text data. Part 1: Preprocessing and tokenization
-Topic 15.Analysis of text data. Part 2: Vector representations of words, working with pre-trained embeddings
-Topic 16.Analysis of text data. Part 3: Named Entity Recognition
-Topic 17.Analysis of text data. Part 4: Topic Modeling
-Topic 18.Q&A
Time Series Analysis
-Topic 19. Time series analysis. Part 1: Statement of the problem, simplest methods. ARIMA model
-Topic 20. Time series analysis. Part 2: Feature extraction and application of machine learning models. Automatic forecasting
-Topic 21. Time series analysis Part 3: Clustering time series (looking for related stock quotes)
Recommender systems
-Topic 22. Recommender systems. Part 1: Statement of the problem, quality metrics. Collaborative filtering. Cold start
-Topic 23. Recommender systems. Part 2: Content filtering, hybrid approaches. Association rules
-Topic 24. Recommender systems. Part 3: Implicit feedback
-Topic 25. Practical lesson on recommender systems. Surprise
-Topic 26.Q&A
Additional topics
-Topic 27.Kaggle ML training No. 1
-Topic 28.Kaggle ML training No. 2
-Topic 29.ML in Apache Spark
-Topic 30.Searching for Data Science jobs
Project work
-Topic 31. Selection of topic and organization of project work
-Topic 32. Consultation on projects and homework
-Topic 33.Protection of design work