v0.6.0

This is the release note of v0.6.0. See here for the complete list of solved issues and merged PRs.

Changes that break compatibility

• [Big Change] Unify arguments in model, update and readout (#368)

• return svg text instead of ipython SVG object in MolVisualizer and SmilesVisualizer (#388)

• Support ChainerX (#376)

• fix typo (#358, Thank you @shllln)

Refactoring:

• Separate links to modules (#270, #301, #305)
  • This is for users to define their own network easily.

New features

Model, Link

• Add Message Passing Neural Network(MPNN) (#312)
• Add SparseGGNN (#342, Thank you @ir5)
• Add GNN-FiLM (#386)
• Add GraphWarpModule(GWM) (#321, #340, #368, #372)
• Add GWMGraphConvModel
• Add Graph Isomorphism Network(GIN) (#321, #337, #345, #371, Thank you @TaizoAyase)

Scaler

• FlowScaler (#318)
• Support more than 3-dim array by StandardScaler (#348)
• MinMaxScaler (#351)
• MaxAbsScaler (#350)

Utils

• GraphConvPredictor (#317)
• set_up_predictor (#336)
• AutoPrintReport (#373)
• Train Utils (#382)

Docker

• Add official dockerfile (#364, #365, #374, #381, Thank you @cks-coil, @zaltoprofen)

Development Utils

• typing module (#331)
• document checker (#332)

Bug-fix

• addHs in SchNet preprocess (#390)
• weave preprocessor (#360)
• fix the type of return value from R2ScoreEvaluator (#339)

Tests

• Use FlexCI (#378, #381)
• GWMPreprocessor tests (#383)
• Fix some tests (#377)