From 39283948bea68a6abfde166e957d31267ae086c2 Mon Sep 17 00:00:00 2001
From: Yaman Umuroglu <maltanar@gmail.com>
Date: Mon, 17 Feb 2020 20:30:35 +0100
Subject: [PATCH] Revert "[Refactor] move StreamingDataflowPartition exec to
own file"
This reverts commit 0105584ba3fc7275f8a95a0c5c504ec06b65978e.
---
src/finn/core/onnx_exec.py | 78 +++++++++++--------
.../custom_op/streamingdataflowpartition.py | 8 +-
2 files changed, 49 insertions(+), 37 deletions(-)
diff --git a/src/finn/core/onnx_exec.py b/src/finn/core/onnx_exec.py
index 57c229663..9786c55c0 100644
--- a/src/finn/core/onnx_exec.py
+++ b/src/finn/core/onnx_exec.py
@@ -31,7 +31,9 @@ import onnx.helper as helper
import onnxruntime as rt
import finn.core.execute_custom_node as ex_cu_node
+from finn.core.modelwrapper import ModelWrapper
from finn.core.remote_exec import remote_exec
+from finn.custom_op.registry import getCustomOp
def execute_node(node, context, graph):
@@ -39,42 +41,54 @@ def execute_node(node, context, graph):
if dataflow partition by using remote execution or rtlsim.
Input/output provided via context."""
- if node.domain == "finn":
- ex_cu_node.execute_custom_node(node, context, graph)
+ if node.op_type == "StreamingDataflowPartition":
+ sdp_node = getCustomOp(node)
+ model = ModelWrapper(sdp_node.get_nodeattr("model"))
+ execute_onnx(model, context)
else:
- # onnxruntime unfortunately does not implement run_node as defined by ONNX,
- # it can only execute entire models -- so we create a model which solely
- # consists of our current node.
- node_inputs = list(filter(lambda x: x.name in node.input, graph.input))
- node_inputs += list(filter(lambda x: x.name in node.input, graph.value_info))
- node_outputs = list(filter(lambda x: x.name in node.output, graph.output))
- node_outputs += list(filter(lambda x: x.name in node.output, graph.value_info))
- node_graph = helper.make_graph(
- nodes=[node],
- name="single-node-exec",
- inputs=node_inputs,
- outputs=node_outputs,
- )
- node_model = helper.make_model(node_graph)
- input_dict = dict()
- for inp in node.input:
- input_dict[inp] = context[inp]
+ if node.domain == "finn":
- sess = rt.InferenceSession(node_model.SerializeToString())
- output_list = sess.run(None, input_dict)
+ ex_cu_node.execute_custom_node(node, context, graph)
- for output_ind in range(len(node.output)):
- outp = node.output[output_ind]
- if output_list[output_ind].shape != context[outp].shape:
- raise Exception(
- """Output shapes disagree after node execution:
- found %s vs expected %s"""
- % (
- str(output_list[output_ind].shape.shape),
- str(context[outp].shape),
+ else:
+
+ # onnxruntime unfortunately does not implement run_node as defined by ONNX,
+ # it can only execute entire models -- so we create a model which solely
+ # consists of our current node.
+ node_inputs = list(filter(lambda x: x.name in node.input, graph.input))
+ node_inputs += list(
+ filter(lambda x: x.name in node.input, graph.value_info)
+ )
+ node_outputs = list(filter(lambda x: x.name in node.output, graph.output))
+ node_outputs += list(
+ filter(lambda x: x.name in node.output, graph.value_info)
+ )
+ node_graph = helper.make_graph(
+ nodes=[node],
+ name="single-node-exec",
+ inputs=node_inputs,
+ outputs=node_outputs,
+ )
+ node_model = helper.make_model(node_graph)
+ input_dict = dict()
+ for inp in node.input:
+ input_dict[inp] = context[inp]
+
+ sess = rt.InferenceSession(node_model.SerializeToString())
+ output_list = sess.run(None, input_dict)
+
+ for output_ind in range(len(node.output)):
+ outp = node.output[output_ind]
+ if output_list[output_ind].shape != context[outp].shape:
+ raise Exception(
+ """Output shapes disagree after node execution:
+ found %s vs expected %s"""
+ % (
+ str(output_list[output_ind].shape.shape),
+ str(context[outp].shape),
+ )
)
- )
- context[outp] = output_list[output_ind]
+ context[outp] = output_list[output_ind]
def execute_onnx(model, input_dict, return_full_exec_context=False):
diff --git a/src/finn/custom_op/streamingdataflowpartition.py b/src/finn/custom_op/streamingdataflowpartition.py
index b9bb0a02f..856ba6d56 100644
--- a/src/finn/custom_op/streamingdataflowpartition.py
+++ b/src/finn/custom_op/streamingdataflowpartition.py
@@ -1,5 +1,3 @@
-from finn.core.modelwrapper import ModelWrapper
-from finn.core.onnx_exec import execute_onnx
from finn.custom_op import CustomOp
# note that the StreamingDataflowPartition node is only a meta/container node,
@@ -21,9 +19,9 @@ class StreamingDataflowPartition(CustomOp):
pass
def execute_node(self, context, graph):
- # retrieve linked "child" dataflow model and execute
- dataflow_model = ModelWrapper(self.get_nodeattr("model"))
- execute_onnx(dataflow_model, context)
+ # TODO add RPC execution with synthesized bitfile?
+ # whole-design rtlsim with PyVerilator may also be an alternative
+ pass
def verify_node(self):
info_messages = []
--
GitLab