submissions

your homeworks

homework 1

  • a first look at language processing
  • assigned: january 9
  • due date: january 16

homework 2

  • neural network fundamentals
  • assigned: january 16
  • due date: january 30

homework 3

  • autocorrection in practice
  • assigned: january 30
  • due date: february 6

homework 4

  • autocomplete with topical information
  • assigned: february 6
  • due date: february 20

homework 5

  • word embeddings
  • assigned: february 20
  • due date: march 13

homework 6

  • designing recurrent networks
  • assigned: march 13
  • due date: march 20

homework 7

  • attention neural networks
  • assigned: march 20
  • due date: april 3


your project



in-class colabs

week 1 colabs

  • In-Class
    Introduction to Google Cloud Platform
    Industry Practice of Cloud Computing

week 2 colabs

  • In-Class
    Introducing Docker and Docker Containers
    Industry Practice of Containerization

week 3 colabs

  • In-Class
    The Complete Naïve Bayes Algorithm
    Predicting Sentiment in Tweets with Naïve Bayes

week 4 colabs

  • In-Class
    Autocorrect - Vocabulary and Candidates
    Processing Text for Autocorrect Candidates

week 5, part 1 - colabs

  • In-Class
    N-Grams - Processing Data

week 5, part 2 - colabs

  • In-Class
    N-Grams - Out of Vocabulary Words
    Using the vocabulary to identify and represent out of vocabulary words

week 5, part 3 - colabs

  • In-Class
    N-Grams - Building N-Grams Models
    Preparing the data for training and inference

week 6 colabs

  • In-Class
    JupyterLabs with a GPU
    Using Google's Vertex AI Workbench

week 7, part 1 - colabs

  • In-Class
    Training a CBOW Word Embedding Model
    Preparing the data for training and inference

week 7, part 2 - colabs

  • In-Class
    word2vec in C (optional)
    Trying Mikolov's Original Code

week 8, part 1 - colabs

  • In-Class
    Preprocessing Parts of Speech
    First Steps- Working with text files, Creating a Vocabulary and Handling Unknown Words

week 8, part 2 - colabs

  • In-Class
    PoS Processing with Numpy
    Parts-of-Speech Tagging - Working with tags and Numpy

week 10 - colabs

  • In-Class
    Recurrent Neural Networks
    Implementing your first RNN with Numpy and Tensorflow

week 11, part 1 - colabs

  • In-Class
    The Attention Mechanism
    Implementing attention in Google Colab

week 11, part 2 - colabs

  • In-Class
    Masking
    Incorporating padding and lookahead masks into attention modeling

week 11, part 3 - colabs

  • In-Class
    Positional Encoding
    Encode position with sinusoidal functions

week 12 colabs

  • In-Class
    Serving LLMs on Google Cloud
    Using the Ollama library with Streamlit frontend to serve the latest models on GPU

week 12 colabs

  • In-Class
    Fine Tuning Large Language Models (optional)
    Fine tune an LLM for dialogue summarization using FLAN-T5

week 13 colabs

  • In-Class
    Detoxifying LLMs using RLHF (optional)
    Fine tune FLAN-T5 with reinforcement learning (PPO) to generate less toxic summaries