v1.3.4

[1.3.4] - 2023-04-13

Release of pyTigerGraph version 1.3.4.

Fixed:

• The fetch() function of various dataloaders was previously ignoring a custom delimiter, now resolved.
• Improved scalability of NodePiece.

v1.3.3

[1.3.3] - 2023-04-03

Release of pyTigerGraph version 1.3.3.

Fixed:

• Made the output of batched dataloaders a consistent batch size, rather than using approximate methods.
• Change getVertexCount() endpoint to more scalable solution.

Added:

• customizeHeader() function to change default timeout and response size values.

v1.3.2

[1.3.2] - 2023-03-13

Release of pyTigerGraph version 1.3.2.

Fixed:

• Logic of shuffling vertices in dataloaders when filter_by attribute was used
• SSL certificate handling when custom certificate is used
• Remove ANSI escape characters from output of .gsql() calls

v1.3.1

[1.3.1] - 2023-02-16

Release of pyTigerGraph version 1.3.1.

Fixed:

• Support for multi-edges in the upsertEdges() function
• Changed the default delimiter used in the dataloaders from , to | in order to support MAP attributes with no user change.

v1.3

[1.3] - 2023-02-09

Release of pyTigerGraph version 1.3.

Added:

• Two new dataloaders: HGTLoader and NodePiece
  • The HGTLoader is a dataloader that performs stratified neighbor sampling Heterogeneous Graph Transformer.
  • NodePiece is a scalable graph embedding approach introduced in NodePiece: Compositional and Parameter-Efficient Representations of Large Knowledge Graphs
• Callback functions to all dataloaders: users can write functions to process the batch in a background thread before it is passed into the training loop.
• A delimiter parameter to all dataloaders: users can choose what delimiter they want to separate attributes they are loading from the graph.

Changed:

• Added better error messaging when REST requests are incorrect.
• Template query support in the featurizer (requires TigerGraph Database version 3.9+)
• Data splitters automatically perform a schema change if needed to add attributes to the database.

Fixed:

• Fixed how custom SSL certificates were handled when instantiating the GSQL client.

v1.2.6

[1.2.6] - 2022-11-30

Release of pyTigerGraph version 1.2.6.

Fixed:

• Fix dependency error importing tqdm for datasets functionality. Will be automatically downloaded if installing the GDS variant of pyTigerGraph, else it will try and throw a warning if tqdm is not available.

v1.2.5

[1.2.5] - 2022-11-15

Release of pyTigerGraph version 1.2.5.

Fixed:

• Fix featurizer install of FastRP on versions DB versions 3.8+ with global schema elements.

v1.2.4

[1.2.4] - 2022-11-14

Release of pyTigerGraph version 1.2.4.

Changed:

• Improve data ingestion print out
• Skip downloading if dataset exists on disk

Fixed:

• Fix metrics issue
• Fix unit test issue in featurizer

v1.2.3

[1.2.3] - 2022-11-13

Release of pyTigerGraph version 1.2.3.

Fixed:

• NeighborLoader fetch() functionality restored.
• runInstalledQuery() defaulting back to GET REST requests.
  • If you wish to pass empty sets, set the usePost = True parameter.

v1.2.2

[1.2.2] - 2022-11-11

Release of pyTigerGraph version 1.2.2.

Fixed:

• The way empty sets are serialized into JSON for installed queries.