5.2. PopTorch Geometric
torchversion from 1.13.1 to 2.0.1.
Added support for the following PyTorch Geometric operators: *
SplineWeightingto support PyTorch Geometric
Added support for PyTorch Geometric
HeteroLinearand normalization layers.
max_pool_xfunctions to avoid reading IPU tensor.
PopTorch now uses ONNX 11 by default.
Added support for a
FixedSizeOptionsobject which can be initialised from an existing dataset or data loader.
Added support for different fixed-size strategies. Currently supports pad-to-max and stream packing strategies.
Added support for custom samplers.
Added support for all arguments used by
torch_geometric.DataLoader, which this class is a drop-in replacement for.
Added support for
See the supported operations chapter in the PopTorch Geometric documentation for full details of the supported operations.
PowerMeanAggregationlayer by changing power layer gradient calculations.
Fixed support for
torch_geometric.Batchin PopTorch which required
Fixed error in node/edge stores of
HeteroDataobject when using fixed-size loader.
SequentialSampleras the default
sampler. Previously the default was
Initial release of PopTorch Geometric. See PyTorch Geometric for the IPU: User Guide for more information.
This release of PopTorch Geometric is experimental. Not all features of PyTorch Geometric are supported, and some functions may not work as expected. The implementation may change without warning in future releases in ways that are not backwards compatible.
Examples demonstrating different ways of using PopTorch Geometric are available in the Graphcore examples repository on GitHub.
There are also easy-to-follow tutorials in the form of notebooks showing how to use PopTorch Geometric to fully utilize the IPU systems available in the PyTorch Geometric tutorials directory in the Graphcore GitHub examples repository.