[Coursera] Practical Data Science Specialization
Become a cloud data science expert. Develop and scale your data science projects into the cloud using Amazon SageMaker
WHAT YOU WILL LEARN
Prepare data, detect statistical data biases, perform feature engineering at scale to train models, & train, evaluate, & tune models with AutoML
Store & manage ML features using a feature store, & debug, profile, tune, & evaluate models while tracking data lineage and model artifacts
Build, deploy, monitor, & operationalize end-to-end machine learning pipelines
Build data labeling and human-in-the-loop pipelines to improve model performance with human intelligence
SKILLS YOU WILL GAIN
- Natural Language Processing with BERT
- ML Pipelines and ML Operations (MLOps)
- A/B Testing and Model Deployment
- Data Labeling at Scale
- Automated Machine Learning (AutoML)
- Statistical Data Bias Detection
- Multi-class Classification with FastText and BlazingText
- Data ingestion
- Exploratory Data Analysis
- ML Pipelines and MLOps
- Model Training and Deployment with BERT
- Model Debugging and Evaluation
Description
Practical Data Science Specialization is an applied science training package published by the Coursera Academy. These two trainings are organized by DeepLearning.AI and mazon Web Services foundations, and during the training process, you will get acquainted with the process of developing, scaling, and implementing data science projects on the Amazon SageMaker cloud server platform. Development environments are very different from the final product production and implementation environment, and require fewer prerequisites and considerations to develop decades. The transfer of data-driven and machine-based projects from the conceptualization and design stage to the production of the final product requires scattered sets of skills that not every developer possesses. The overall architecture and structure of your project should be such that you provide the best performance with the least resources and the process of development and use is easy.
Science is a vast interdisciplinary industry that requires skills in mathematics, statistics, statistics, and programming. This training package is completely dedicated to developers, analysts and knowledge that is designed to deal with data on a daily basis and its applicants are expected to be designed in the programming language of Python, SQL and systems . Master database management.
What you will learn in Practical Data Science Specialization
- Initial data collection and preparation
- Detection of biases and defects of raw statistical data
- Practice, evaluate and optimize different models using AutoML
- Design, implement, monitor and manage machine learning pipeline operations
- Natural language processing with the BERT library
- A / B testing of different machine learning models
- Automatic machine learning
- Multiple classification with FastText and BlazingText libraries
- Forbidden data
- Exploratory data analysis
- Evaluate and troubleshoot different machine learning models
Course Specifications
Publisher: Coursera
Instructors: Antje Barth, Shelbee Eigenbrode, Sireesha Muppala and Chris Fregly
Language: English
Level: Advanced
institution/university: DeepLearning.AI and Amazon Web Services
Number of Courses: 3
Duration: Approximately 3 months to complete – Suggested pace of 5 hours/week
Size: 561 MB
https://www.coursera.org/specializations/practical-data-science.