DeepXDE v1.8.0

A lot of implementations for backend paddle! Feel free to use backend paddle.🎉🎉🎉

Areas of improvement

• dde.nn.FNN supports defining activation functions for each layer
• dde.geometry.PointCloud supports boundary points and normals
• Bug fix

New APIs

• Backend PyTorch: Support dde.nn.DeepONet

DeepXDE v1.7.2

Areas of improvement

• Add dde.icbc.PointSetOperatorBC
• dde.callbacks.OperatorPredictor can be used during training
• Backend PyTorch: dde.icbc.PointSetBC supports multi-component outputs
• Bug fix

DeepXDE v1.7.1

Areas of improvement

• dde.data.TripleCartesianProd and dde.data.QuadrupleCartesianProd support mini-batch for both branch and trunk nets
• Backend TensorFlow 1.x: L-BFGS dumps trainable variables and test loss

DeepXDE v1.7.0

Areas of improvement

• dde.icbc.PointSetBC supports mini-batch
• Bug fix

New APIs

• Backend paddle: Support dde.nn.DeepONet and dde.nn.DeepONetCartesianProd

API changes

• Change dde.callback.PDEResidualResampler to dde.callback.PDEPointResampler

DeepXDE v1.6.2

Areas of improvement

• Set Hammersley as the default point sampling for PINN
• Improve point sampling of dde.geometry.GeometryXTime.random_points
• dde.callback.ModelCheckpoint supports monitoring test loss
• PyTorch backend: dde.nn.PODMIONet and dde.nn.MIONetCartesianProd support multiple merge operations

DeepXDE v1.6.1

Areas of improvement

• Backend TensorFlow 1.x dde.nn.DeepONet supports customized branch
• Fix Triangle.on_boundary for float32
• Bug fix: a few issues of float64
• Many documentation improvements

New APIs

• Backend PyTorch adds dde.nn.PODMIONet
• Backend PyTorch adds dde.nn.MIONetCartesianProd

DeepXDE v1.6.0

Backend JAX supports PDE inverse problems.🎉🎉🎉

Areas of improvement

• Backend PyTorch supports l2 regularization
• Improve XLA jit

New APIs

• Add dde.data.PDEOperatorCartesianProd
• Backend TensorFlow supports dde.nn.PFNN

API changes

• Model.train: epochs is replaced by iterations
• Merge PIDeepONet into DeepONet

DeepXDE v1.5.1

Areas of improvement

• Bug fix: dde.icbc.NeumannBC for dde.geometry.Polygon

New APIs

• Add new geometry dde.geometry.PointCloud

DeepXDE v1.5.0

Areas of improvement

• Support new loss "mean l2 relative error"
• Backend PyTorch supports dde.nn.DeepONetCartesianProd and dde.nn.PODDeepONet
• Backend PyTorch supports "step" LR decay
• dde.callbacks.OperatorPredictor supports backends tensorflow.compat.v1, tensorflow, pytorch, paddle
• Fix dde.config.set_random_seed

New APIs

• Add dde.config.enable_xla_jit and dde.config.disable_xla_jit

API changes

• Change DDEBACKEND to DDE_BACKEND

DeepXDE v1.4.0

DeepXDE supports a new backend PaddlePaddle.🎉🎉🎉

Areas of improvement

• Backend PyTorch supports dde.nn.PFNN
• Backend TensorFlow 1.x enables XLA for GPU