Compresso Recsys

Getting Started

  • Installation
    • Install from PyPI
    • Optional Extras
    • Local Development
    • Install from GitHub
    • Build the Documentation Locally
  • Getting Started
    • Basic Python Usage
    • Checkpoint Workflow
    • What Goes Into a Checkpoint
    • Split Modes
    • Retrieval Metrics
  • Quickstart
    • Choose a Split Mode

User Guide

  • Examples
    • Dataset Loader
    • Building Checkpoints
    • Checkpoint Read/Write
    • Evaluate Embeddings From Python
    • Evaluate Model Predictions
    • Train TEASER for Cold Items
    • Train TEASER with Gradient Descent
    • Serve New TEASER Candidates
    • Train and Evaluate ELSA
    • Using item embeddings
      • Precomputed features
      • User-computed features
      • Reading and evaluating features
  • Datasets
    • Reading the default tables
    • MovieLens 1M
    • MovieLens 20M
    • Goodbooks-10k
    • Amazon Reviews 2023
    • Steam
    • Netflix Prize
    • MSD Taste Profile
    • Gowalla
    • DBbook
    • Last.fm-2K
    • Using item embeddings
  • Bringing Your Own Dataset
    • What a dataset has to produce
    • The dataset
    • The dataset class
    • Two defaults that will empty your dataset
    • What is in it
    • Does it work?
    • Cold items, for free
    • What timestamps would have added
    • You do not need a dataset class at all
  • Implementing a Recommender
    • What the base class owns
    • Data
    • Configuration
    • The complete model
    • Low-level prediction
    • Stable-ID recommendation
    • Standalone persistence
    • Persistence inside a data checkpoint
    • Evaluation
    • Executable contract checks
    • Extending the pattern
  • Reproducing SASRec Results on ML1M
    • Configuration and run modes
    • Build the MovieLens 1M checkpoint
    • Inspect the split
    • Train SASRec
    • Score the catalog
    • Published protocol: 100 sampled negatives at 10
    • Package protocol: full catalog at 20
    • Check the epoch budget on validation data
    • Save the trained model
  • Comparing Models Statistically
    • Why the aggregates are not enough
    • Do the comparison
    • A complete worked example
      • Build the split
      • Train both models
      • Evaluate both on identical users
      • Compare
      • Reading it
    • Reading the output
      • The effect
      • How precisely you know it
      • How often the two models tied
      • Whether it could be chance
      • The verdict
    • Five traps
      • p-values have a floor
      • Ties are not missing data
      • The interval and the p-value can disagree
      • confidence_level also moves the significance bar
      • One trained model is not the method
    • Choosing the settings
      • n_resamples
      • test_method
      • test_method: why not a t-test?
      • alternative
      • correction
    • When one user owns several rows
    • What the methods do, formally
    • Reporting the result
      • Methods
      • Results
      • Checklist
    • References
    • See also
  • Citing Compresso Recsys
    • Compresso
    • EASE
    • Neighborhood Models
    • Mult-VAE, Mult-DAE, and AutoRec
    • TEASER
    • ELSA
    • ELSA with sampled output candidates
    • Compressed ELSA
    • SASRec
    • SimpleRNN
    • SimpleGPT and SimpleBidirectionalTransformer
    • Content, Popularity, and Random Baselines
    • Datasets
      • MovieLens 1M and 20M
      • Goodbooks-10k
      • Amazon Reviews 2023
      • Steam
      • Netflix Prize
      • MSD / Taste Profile
      • Gowalla
      • Additional Movie and Book Descriptions
    • Evaluation Protocol
      • Dataset preprocessing
    • Statistical Comparison
    • SWAP multimodal ML-1M, DBbook, and Last.fm-2K
  • Checkpoint CLI Reference
    • Temporary MovieLens download workaround
    • Steam
    • Amazon Reviews 2023
      • Leave-Last-Out Checkpoint
      • Temporal Checkpoint
    • Checkpoint Split Schema
    • Builder Parameters
    • Supported Amazon Reviews 2023 Datasets

API Reference

  • API Reference
    • Core API
      • Datasets
        • SplitBundle
        • RecSysDataset
        • Goodbooks
        • MovieLens1M
        • MovieLens20M
        • AmazonReviews2023
      • Checkpoint Helpers
        • build_recsys_checkpoint()
        • update_checkpoint()
        • read_checkpoint()
        • load_manifest()
        • save_manifest()
        • update_stage_manifest()
        • save_json()
        • load_json()
        • save_recsys_split()
        • load_recsys_split()
        • save_cluster_graph_stage()
        • load_cluster_graph_stage()
    • Datasets API
      • DBbook
        • DBbook.timestamp_precision
        • DBbook.get_official_split()
      • LastFM2K
        • LastFM2K.timestamp_precision
    • Checkpoint API
      • Item embeddings
        • save_item_embeddings()
        • load_item_embeddings()
        • list_item_embeddings()
        • enrich_multimodal_checkpoint()
      • Checkpoint Contexts
      • Manifest and JSON Helpers
      • Split and Cluster Stages
        • Item partitions
    • Fitted Model Persistence
      • Embedding Models in Data Checkpoints
      • Stable Item IDs
      • Optimizer State
      • Checkpoint Format
        • ModelCheckpointWriter
        • ModelCheckpointReader
      • Extending Persistence
      • Warm Catalog Adapters
    • Sequences API
      • Interaction Histories
        • ItemSequences
        • save_item_sequences()
        • load_item_sequences()
      • Which Split Modes Produce Them
    • Models API
      • Recommender Contract
        • Recommender
        • ColdStartRecommender
        • ItemVocabulary
        • SequentialRecommender
      • Production Recommendations
        • IdentifiedRecommender
        • Recommendations
      • Training and Prediction Progress
      • Fitted Model Persistence
        • PersistableRecommender
        • BasePersistableRecommender
      • Implementing New Models
        • BaseIdentifiedRecommender
        • BaseCollaborativeRecommender
        • BaseColdStartRecommender
        • BaseSequentialRecommender
        • The Owned Candidate Catalog
        • Training Interaction Batches
      • Transductive Models in Expanding Catalogs
        • When to Reach for It Outside temporal
      • Collaborative Filtering Models
        • Baselines
        • Neighborhood Models
        • Multinomial Autoencoders
        • EASE
        • ELSA
      • Cold-Start Models
        • ContentRecommender
        • TEASER
        • TEASERGD
      • Sequential Models
        • Tokenizing Histories
        • SimpleRNN
        • SimpleGPT
        • SASRec
        • SimpleBidirectionalTransformer
    • Retrieval API
      • Holdout Builders
        • build_eval_holdout()
        • build_item_cold_holdout()
        • build_leave_last_out_holdout()
        • build_temporal_holdout()
      • Embedding Evaluation
        • evaluate_item_embeddings()
        • evaluate_item_embeddings_with_holdout()
    • Ranking Evaluation API
      • Predictions and Targets
      • Evaluation Results
        • EvaluationResult
      • Custom Metrics
        • evaluate_recommender()
        • evaluate_ranked_predictions()
        • RankingEvaluator
      • Metrics
        • Default Metrics
        • Optional Metrics
    • Statistics API
      • Comparison Functions
        • compare_models()
        • compare_pair()
      • Results
        • PairwiseComparison
        • ComparisonReport
      • Parameter Values
Compresso Recsys
  • Search

© Copyright .

Built with Sphinx using a theme provided by Read the Docs.