SageMaker Built-in Algorithm Cheat Sheet

  • BlazingText
    • Word2vec
    • Text classification
  • Deep AR
    • Forecasting with RNN
  • Factorization Machine
    • Regression
    • Classification
  • Image Classification
    • Image multi-classification with CNN (ResNet)
  • IP Insights
    • Unsupervised IP usage pattern (i.e. how IP connects to user ID, account numbers, etc.)
  • K-means
    • Unsupervised clustering
  • K-NN
    • Regression
    • Classification
  • Latent Dirichlet Allocation (LDA)
    • Unsupervised text clustering
  • Linear Learner
    • Regression
    • Classification
  • Neural Topic Model (NTM)
    • Unsupervised text clustering
    • Used for classification, summarization and recommendation
  • Object2Vec
    • ?
  • Object Detection
    • Supervised multi-object detection in images
    • Yes or no as answer (no object location information)
  • PCA
    • Unsupervised reduce dimensionality of data
    • Make features more concise and prominent to improve training efficiency
  • Random Cut Forest
    • Unsupervised data anomaly detection
  • Semantic Segmentation
    • Multi-object detection and segmentation in images
    • Will detect and create a pixel-by-pixel mask of objects in image
  • seq2seq
    • Supervised machine translation, summarization
    • Input = a sequence of tokens; output = another sequence of tokens
  • XGBoost
    • Supervised prediction

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s