3-Tier Architecture

Below APIs are for using the Oracle Graph Server (3-tier usage). For APIs to execute PGQL against Oracle Database (2-tier usage), please see 2-Tier Architecture.

• Connecting to a graph server
• Using Autonomous Database Graph Client
• Accessing Properties and Labels on Vertices and Edges
  • Setting and Getting Property Values
  • Ordered Iteration over the values of a Property
  • Getting Label Values
• Loading a Subgraph from Property Graph Views
  • PGQL Based Subgraph Loading
  • Prepared PGQL Queries
  • Providing Database Connection Credentials
  • Dynamically Expanding a Subgraph
• Graph Mutation and Subgraphs
  • Graph Mutation & Subgraphs
    • Copying and Simplifying Graphs
    • Transposing Graphs
    • Undirecting Graphs
    • Advanced Multi-Edge Handling
      • Picking
      • StrategyBuilder in General
      • Creating Subgraphs
    • Creating a Bipartite Subgraph
    • Creating a Sparsified Subgraph
  • Creating Subgraphs from Loaded Graphs
    • Recap: load a graph into memory
    • An introduction to filter expressions
    • Creating a subgraph based on a filter in PGX
      • Example for creating and using an edge filter
      • Example for creating and using a vertex filter
    • Create a subgraph based on a more complex filter
    • Combining filters
      • Creating a union of two filters
      • Creating an intersection of two filters
    • Creating a set of vertices or edges using a filter
      • Example for creating a vertex set
    • Create a bipartite subgraph based on a vertex list
  • Loading Subgraphs
    • Recap: Graph loading and subgraph creation
      • Loading a graph
      • Creating a subgraph from a loaded graph
    • Combining loading and subgraph creation
• Graph Synchronization
• Managing Transient Properties and Collections
  • Transient Properties
  • Collections and Scalars
• Versioning a Graph
  • Graph Versioning
    • Configuring the snapshots source
    • Snapshot creation
      • Snapshot creation via Refreshing
      • Snapshot creation via ChangeSet
    • Checking out the latest snapshots of a graph
    • Checking out different snapshots of a graph
    • Directly loading a specific snapshot of a graph
  • Auto-refreshing graphs
    • Configuring the PGX server
    • Configuring basic auto-refresh
    • Reading the graph using the PGX Shell or a java application
    • Checking out the newest version of the graph
    • Auto-refresh with delta update (“delta-refresh”)
      • Differences to normal auto-refresh
      • Delta-refresh requirements
      • Example configuration
  • In-place delta graph refresh
• Running Built-in Algorithms Using the Analyst API
  • Obtaining an Analyst instance
  • Running Algorithms and Browsing Results
  • Optional arguments
  • Reuse existing data structures
  • Disposing an Analyst Instance
• Graph Pattern Matching (PGQL)
  • Running Graph Pattern Matching Queries (PGQL)
    • Submitting a Query and Getting a Result Set
      • Examples
    • Getting Result Element Information from a Result Set
    • Print a Result Set
    • Creating vertex/edge sets out of a PGQL result set
    • Prepared Statements
    • Explain the query execution plan
    • ON clause
  • Graph Pattern Matching Examples
    • The Datasets
    • Read the Graphs
    • Submit Queries
      • Enemy of My Enemy is My Friend
      • Top 10 Most Collaborative People
      • Transitive Connectivity
• PgxFrame (Tabular Data-Structure)
  • Overview
  • Functionalities
    • Loading a PgxFrame (with multiple data types) from some specified path
    • Loading a PgxFrame from client-side data
    • Printing the content of a PgxFrame
    • Destroying a PgxFrame
    • Storing a PgxFrame to some specified path
    • Flattening vector properties
    • Union of PGX Frames
    • Joining PGX Frames
    • PgxFrame helpers
    • PgxFrame-PgqlResultSet conversions
    • Creating a graph from multiple PgxFrame instances
• PgxML: Machine Learning library for Graphs
  • DeepWalk
    • Overview of the algorithm
    • Functionalities
      • Loading a graph
      • Building a DeepWalk Model (minimal)
      • Building a DeepWalk Model (customized)
      • Training the DeepWalk model
      • Getting Loss value
      • Computing the similar vertices
      • Computing the similars (for a vertex batch)
      • Getting all trained vertex vectors
      • Storing a trained model
      • Loading a pre-trained model
      • Destroying a model
  • SupervisedGraphWise
    • Overview
    • Model Structure
    • Functionalities
      • Loading a graph
      • Building a GraphWise Model (minimal)
      • Advanced hyperparameter customization
      • Property types supported
      • Classification vs Regression models
      • Setting a custom Loss Function and Batch Generator (for Anomaly Detection)
      • Training the SupervisedGraphWiseModel
      • Getting Loss value
      • Inferring vertex labels
      • Evaluating model performance
      • Inferring embeddings
      • Storing a trained model
      • Loading a pre-trained model
      • Explaining a Prediction
      • Destroying a model
  • UnsupervisedGraphWise
    • Overview
    • Model Structure
    • Functionalities
      • Loading a graph
      • Building an UnsupervisedGraphWise Model (minimal)
      • Advanced hyperparameter customization
      • Property types supported
      • Training the UnsupervisedGraphWiseModel
      • Getting Loss value
      • Inferring embeddings
      • Storing a trained model
      • Loading a pre-trained model
      • Explaining a Prediction
      • Destroying a model
  • UnsupervisedAnomalyDetectionGraphWise
    • Overview
    • Model Structure
    • Functionalities
      • Loading a graph
      • Building an UnsupervisedGraphWise Model (minimal)
      • Advanced hyperparameter customization
      • Training the UnsupervisedAnomalyDetectionGraphWiseModel
      • Getting Loss value
      • Inferring embeddings
      • Inferring anomalies
      • Storing a trained model
      • Loading a pre-trained model
      • Destroying a model
  • SupervisedEdgeWise
    • Overview
    • Model Structure
    • Functionalities
      • Loading a graph
      • Example: predicting ratings on the Movielens Dataset
      • Building an EdgeWise Model (minimal)
      • Advanced hyperparameter customization
      • Property types supported
      • Classification vs Regression models
      • Setting a custom Loss Function and Batch Generator (for Anomaly Detection)
      • Setting the edge embedding production method
      • Training the SupervisedEdgeWiseModel
      • Getting Loss value
      • Inferring edge labels
      • Evaluating model performance
      • Inferring embeddings
      • Storing a trained model
      • Loading a pre-trained model
      • Destroying a model
  • UnsupervisedEdgeWise
    • Overview
    • Model Structure
    • Functionalities
      • Loading a graph
      • Example: computing edge embeddings on the Movielens Dataset
      • Building an EdgeWise Model (minimal)
      • Advanced hyperparameter customization
      • Property types supported
      • Setting the edge embedding production method
      • Training the UnsupervisedEdgeWiseModel
      • Getting Loss value
      • Inferring embeddings
      • Storing a trained model
      • Loading a pre-trained model
      • Destroying a model
  • Pg2vec
    • Overview of the algorithm
    • Functionalities
      • Loading a graph
      • Building a Pg2vec Model (minimal)
      • Building a Pg2vec Model (customized)
      • Training the Pg2vec model
      • Getting the loss value
      • Computing the similar graphlets
      • Computing the similars (for a graphlet batch)
      • Inferring a graphlet vector
      • Inferring vectors (for a graphlet batch)
      • Getting all trained graphlet vectors
      • Storing a trained model
      • Loading a pre-trained model
      • Destroying a model
  • Model repository and model stores
    • Overview
    • DB backed model repository
    • File backed model repository
• Running Custom Graph Algorithms
  • Compiling Custom Algorithms
  • Running Algorithms
• User-Defined Functions
  • Using UDFs in PGX
    • UDFs in PGQL