ML - Level 1 - Modele 1 - Uyen

  1. An overview of things learned

Technical Area:

  • Popular Machine Learning (ML) applications and algorithms in general and particularly in neural network
  • An advanced Recommender System: Content-based, its measure and evaluation metrics
  • Data Mining by Web crawling and scraping with Python librabies: Scrapy, urllib and BeautifulSoup.
  • Git basic command lines and how Git content-flow strategical organized in branches.
  • NLP concept and different techniques: Vanilla Neural Network, RNNS, LSTM, GPT2, Attention, Word Embedding and how tokens trained in BERT Tools:
  • Some IDE for ML: VSCode, Colab
  • Data mining: Scrapy, urllib, BeautifulSoup, Selenium, Git and Github
  • NPL: Spacy, Pytorch, TensorFlow
  • Project management: Trello

Soft skills

  • Ethical and legal issue with web crawling.
  • Project management framework SCRUM and tool Trello as well as useful tips to practice.
  • Useful technical communities and data sources.
  1. Achievements highlighting
  • Differentiating different library for Web scraping and experimenting data mining from several websites to get used to functions.
  • Understanding BERT transforming architecture and creating training, testing, evaluating loop
  • Getting familiar with NLP concepts and working mechanisms, together with Deep learning library Pytorch and Tensorflow.
  1. Challenges
  • Being familiar with Web scraping and NLP libraries.
  • Understanding the structure of HTML tags on web script
  • Troubleshooting the code error