.torch_interpreter.investigate_helper¶
- experimental_experiment.torch_interpreter.investigate_helper.prepare_args_kwargs(torch_results: Dict[str, Any], node: Node) Tuple[Tuple[Any, ...], Dict[str, Any]][source]¶
Prepares args and kwargs before executing a fx node.
- Parameters:
torch_results – existing results
node – node to execute
- Returns:
new args and kwargs
- experimental_experiment.torch_interpreter.investigate_helper.run_aligned(ep: ExportedProgram, onx: ModelProto | FunctionProto, args: Tuple[Tensor, ...], check_conversion_cls: Dict[str, Any] | type, kwargs: Dict[str, Any] | None = None, verbose: int = 0) Iterator[Tuple[Any, ...]][source]¶
Runs in parallel both the exported program and the onnx proto and looks for discrepancies. The function does match on result names so it assumes the exported program and the onnx model have the same names fro equivalent results.
- Parameters:
ep – exported program
onx – model or function proto
args – input args
check_conversion_cls – defines the runtime to use for this task
kwargs – input kwargs
verbose – verbosity level
- Returns:
a list of tuples containing the results, they come in tuple,
Example:
<<<
import pprint import pandas import torch from experimental_experiment.reference import ( # This can be replace by any runtime taking NodeProto as an input. ExtendedReferenceEvaluator as ReferenceEvaluator, ) from experimental_experiment.torch_interpreter import to_onnx from experimental_experiment.torch_interpreter.investigate_helper import run_aligned class Model(torch.nn.Module): def forward(self, x): ry = x.abs() rz = ry.exp() rw = rz + 1 ru = rw.log() + rw return ru def post_process(obs): dobs = dict(zip(["ep_id_node", "onnx_id_node", "ep_name", "onnx_name"], obs)) dobs["err_abs"] = obs[-1]["abs"] dobs["err_rel"] = obs[-1]["rel"] return dobs x = torch.randn((5, 4)) Model()(x) # to make sure the model is running ep = torch.export.export( Model(), (x,), dynamic_shapes=({0: torch.export.Dim("batch")},) ) onx = to_onnx(ep) results = list( map( post_process, run_aligned( ep, onx, (x,), check_conversion_cls=dict(cls=ReferenceEvaluator, atol=1e-5, rtol=1e-5), verbose=1, ), ), ) print("------------") print("final results") df = pandas.DataFrame(results) print(df)
>>>
[run_aligned] +torch-cst: p_init1_s_::RSh1: T1s1[1.0,1.0:A1.0] [run_aligned] +onnx-init: init1_s_::RSh1: A1s1[1.0,1.0:A1.0] [run_aligned] +onnx-init: p_init1_s_::RSh1: A1s1[1.0,1.0:A1.0] [run_aligned] +onnx-input: x: T1s5x4[-1.8834538459777832,2.7419533729553223:A0.37898920737206937] [run_aligned] run ep.graph.nodes[0]: placeholder -> 'x' [run_aligned] +torch x=T1s5x4[-1.8834538459777832,2.7419533729553223:A0.37898920737206937] [run_aligned] run ep.graph.nodes[1]: call_function[aten.abs.default] -> 'abs_1' [run_aligned] +torch abs_1=T1s5x4[0.07908470928668976,2.7419533729553223:A0.8167600449174642] [run_aligned] run onx.graph.node[0]: Abs(x) -> abs_1 [run_aligned] +onnx abs_1=A1s5x4[0.07908470928668976,2.7419533729553223:A0.8167600449174642] [run_aligned] =common results abs_1: abs=0.0, rel=0.0,amax=0,0 [run_aligned] run ep.graph.nodes[2]: call_function[aten.exp.default] -> 'exp' [run_aligned] +torch exp=T1s5x4[1.0822960138320923,15.517266273498535:A3.076218467950821] [run_aligned] run onx.graph.node[1]: Exp(abs_1) -> exp [run_aligned] +onnx exp=A1s5x4[1.0822960138320923,15.517266273498535:A3.076218432188034] [run_aligned] =common results exp: abs=4.76837158203125e-07, rel=1.0148514777933538e-07, n=20.0,amax=2,2 [run_aligned] run ep.graph.nodes[3]: call_function[aten.add.Tensor] -> 'add' [run_aligned] +torch add=T1s5x4[2.0822958946228027,16.51726531982422:A4.076218438148499] [run_aligned] run onx.graph.node[2]: Add(exp, init1_s_::RSh1) -> add [run_aligned] +onnx add=A1s5x4[2.0822958946228027,16.51726531982422:A4.0762184143066404] [run_aligned] =common results add: abs=4.76837158203125e-07, rel=1.0963549886519673e-07, n=20.0,amax=2,2 [run_aligned] run ep.graph.nodes[4]: call_function[aten.log.default] -> 'log' [run_aligned] +torch log=T1s5x4[0.7334710955619812,2.80440616607666:A1.229018098115921] [run_aligned] run ep.graph.nodes[5]: call_function[aten.add.Tensor] -> 'add_1' [run_aligned] +torch add_1=T1s5x4[2.8157670497894287,19.321670532226562:A5.305236494541168] [run_aligned] run ep.graph.nodes[6]: output -> 'output' [run_aligned] +torch output=(T1s5x4[2.8157670497894287,19.321670532226562:A5.305236494541168],) [run_aligned] run onx.graph.node[3]: Log(add) -> _onx_log_add [run_aligned] +onnx _onx_log_add=A1s5x4[0.7334710955619812,2.80440616607666:A1.2290180951356888] [run_aligned] run onx.graph.node[4]: Add(_onx_log_add, add) -> output_0 [run_aligned] +onnx output_0=A1s5x4[2.8157670497894287,19.321670532226562:A5.305236458778381] [run_aligned] =common results* add_1/output_0: abs=9.5367431640625e-07, rel=1.0826541979963732e-07, n=20.0,amax=2,2 ------------ final results ep_id_node onnx_id_node ep_name onnx_name err_abs err_rel 0 1 0 abs_1 abs_1 0.000000e+00 0.000000e+00 1 2 1 exp exp 4.768372e-07 1.014851e-07 2 3 2 add add 4.768372e-07 1.096355e-07 3 6 4 output_0 add_1 9.536743e-07 1.082654e-07
- experimental_experiment.torch_interpreter.investigate_helper.run_fx_node(node: Node, args: Tuple[Any, ...], kwargs: Dict[str, Any] | None = None) Tuple[Any, ...][source]¶
Executes a node
- Parameters:
node – runs a node
args – unnamed inputs to the node
kwargs – named inputs to the node
- Returns:
results