date_range 18/09/2023 01:12

“Machine Learning Design Patterns” authored by Valliappa Lakshmanan, Sara Robinson, and Michael Munn, from Google, is a highly recommended read for individuals engaged in machine learning and data science. This book serves as an invaluable resource, offering a comprehensive overview of best practices and design patterns for machine learning development. Within its pages, we can discover practical solutions to the common hurdles encountered in ML projects, encompassing critical areas such as data preparation, model training, and deployment. The book’s extensive coverage spans essential topics, including data preparation, model selection and training, evaluation and validation, as well as deployment and monitoring.

date_range 26/08/2023 22:15

My first hackthon in a company, organized by Walmart and Google, was an exciting venture into the world of innovation and technology. Taking place over four days, starting from Monday and ending on Thursday, it was a whirlwind of creativity and collaboration.

date_range 05/08/2023 22:30

As a data science practitioner eager to harness the potential of artificial intelligence, I embarked on an enlightening journey with renowned AI expert, Andrew Ng and OpenAI techincal memeber Isa Fulford. Their courses, “ChatGPT Prompt Engineering for Developers” and “Building Systems with the ChatGPT API,” offered on DeepLearning.AI, opened new horizons in the world of large language model (LLM) of ChatGPT. Join me as I share my experiences and express my gratitude for the invaluable knowledge gained from these transformative courses.

date_range 11/03/2023 21:48

I have been working in the machine learning and data science (ML/DS) engineering area in industry for a year. Also, I had worked in the software engineering (SE) area for two years before this. Some of the SE experience could be shifted to ML/DS, but most of them are totally not transferrable. Here are my personal persepctives.

date_range 04/03/2023 22:30

Recently, we have done a project with xgboost model for classification. With the increasing of large amouts of data, we need to use XGBoost distributed training to replace the current pandas XGBoost training solution in Spark.