deeplearning/modulus/modulus-sym-v130/_modules/modulus/sym/node.html

Sym v1.3.0

Source code for modulus.sym.node

# Copyright (c) 2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

""" Modulus nodes
"""
from sympy import Add
import torch
from .constants import diff_str
from .key import Key


[docs]class Node: """ Base class for all nodes used to unroll computational graph in Modulus. Parameters ---------- inputs : List[Union[str, Key]] Names of inputs to node. For example, `inputs=['x', 'y']`. outputs : List[Union[str, Key]] Names of outputs to node. For example, `inputs=['u', 'v', 'p']`. evaluate : Pytorch Function A pytorch function that takes in a dictionary of tensors whose keys are the above `inputs`. name : str Name of node for print statements and debugging. optimize : bool If true then any trainable parameters contained in the node will be optimized by the `Trainer`. """ def __init__(self, inputs, outputs, evaluate, name="Node", optimize=False): super().__init__() self._inputs = Key.convert_list([x for x in inputs if diff_str not in str(x)]) self._outputs = Key.convert_list(outputs) self._derivatives = Key.convert_list([x for x in inputs if diff_str in str(x)]) self.evaluate = evaluate self._name = name self._optimize = optimize # set evaluate saveable to false if doesn't exist if not hasattr(self.evaluate, "saveable"): self.evaluate.saveable = False # check that model has name if optimizable if self._optimize: assert hasattr( self.evaluate, "name" ), "Optimizable nodes require model to have unique name"
[docs] @classmethod def from_sympy(cls, eq, out_name, freeze_terms=[], detach_names=[]): """ generates a Modulus Node from a SymPy equation Parameters ---------- eq : Sympy Symbol/Exp the equation to convert to a Modulus Node. The inputs to this node consist of all Symbols, Functions, and derivatives of Functions. For example, `f(x,y) + f(x,y).diff(x) + k` will be converted to a node whose input is [`f,f__x,k`]. out_name : str This will be the name of the output for the node. freeze_terms : List[int] The terms that need to be frozen detach_names : List[str] This will detach the inputs of the resulting node. Returns ------- node : Node """ from modulus.sym.utils.sympy.torch_printer import ( torch_lambdify, _subs_derivatives, SympyToTorch, ) # sub all functions and derivatives with symbols sub_eq = _subs_derivatives(eq) # construct Modulus node if bool(freeze_terms): print( "the terms " + str(freeze_terms) + " will be frozen in the equation " + str(out_name) + ": " + str(Add.make_args(sub_eq)) ) print("Verify before proceeding!") else: pass evaluate = SympyToTorch(sub_eq, out_name, freeze_terms, detach_names) inputs = Key.convert_list(evaluate.keys) outputs = Key.convert_list([out_name]) node = cls(inputs, outputs, evaluate, name="Sympy Node: " + out_name) return node

@property def name(self): return self._name @property def outputs(self): """ Returns ------- outputs : List[str] Outputs of node. """ return self._outputs @property def inputs(self): """ Returns ------- inputs : List[str] Inputs of node. """ return self._inputs @property def derivatives(self): """ Returns ------- derivatives : List[str] Derivative inputs of node. """ return self._derivatives @property def optimize(self): return self._optimize def __str__(self): return ( "node: " + self.name + "\n" + "evaluate: " + str(self.evaluate.__class__.__name__) + "\n" + "inputs: " + str(self.inputs) + "\n" + "derivatives: " + str(self.derivatives) + "\n" + "outputs: " + str(self.outputs) + "\n" + "optimize: " + str(self.optimize) )

© Copyright 2023, NVIDIA Modulus Team. Last updated on Jan 25, 2024.