search

anibali / margipose

3D monocular human pose estimation
Apache License 2.0
98 stars 20 forks source link

Onnx conversion giving error #11

Closed jaggernaut007 closed 4 years ago

jaggernaut007 commented 4 years ago

ValueError: Dimensions must be equal, but are 32 and 33 for '{{node Add_135}} = Add[T=DT_FLOAT](Relu_132, BatchNormalization_134/add_1)' with input shapes: [1,128,32,32], [1,128,33,33].

Stack trace

Traceback (most recent call last): File "/home/shreyas/.local/lib/python3.8/site-packages/tensorflow/python/framework/ops.py", line 1654, in _create_c_op c_op = pywrap_tf_session.TF_FinishOperation(op_desc) tensorflow.python.framework.errors_impl.InvalidArgumentError: Dimensions must be equal, but are 32 and 33 for '{{node Add_135}} = Add[T=DT_FLOAT](Relu_132, BatchNormalization_134/add_1)' with input shapes: [1,128,32,32], [1,128,33,33].

During handling of the above exception, another exception occurred:

Traceback (most recent call last): File "example/onnx_to_tf.py", line 6, in tf_rep = prepare(onnx_model) # prepare tf representation File "/home/shreyas/onnx/onnx-tensorflow/onnx_tf/backend.py", line 66, in prepare return cls.onnx_model_to_tensorflow_rep(model, strict) File "/home/shreyas/onnx/onnx-tensorflow/onnx_tf/backend.py", line 86, in onnx_model_to_tensorflow_rep return cls._onnx_graph_to_tensorflow_rep(model.graph, opset_import, strict) File "/home/shreyas/onnx/onnx-tensorflow/onnx_tf/backend.py", line 143, in _onnx_graph_to_tensorflow_rep output_ops = cls._onnx_node_to_tensorflow_op(onnx_node, File "/home/shreyas/onnx/onnx-tensorflow/onnx_tf/backend.py", line 249, in _onnx_node_to_tensorflow_op return handler.handle(node, tensor_dict=tensor_dict, strict=strict) File "/home/shreyas/onnx/onnx-tensorflow/onnx_tf/handlers/handler.py", line 60, in handle return ver_handle(node, kwargs) File "/home/shreyas/onnx/onnx-tensorflow/onnx_tf/handlers/backend/add.py", line 23, in version_7 return [cls.make_tensor_from_onnx_node(node, kwargs)] File "/home/shreyas/onnx/onnx-tensorflow/onnx_tf/handlers/backend_handler.py", line 111, in make_tensor_from_onnx_node return cls._run_tf_func(tf_func, inputs, attrs) File "/home/shreyas/onnx/onnx-tensorflow/onnx_tf/handlers/backend_handler.py", line 181, in _run_tf_func return tf_func(*inputs, File "/home/shreyas/.local/lib/python3.8/site-packages/tensorflow/python/ops/gen_mathops.py", line 346, in add , _, _op, _outputs = _op_def_library._apply_op_helper( File "/home/shreyas/.local/lib/python3.8/site-packages/tensorflow/python/framework/op_def_library.py", line 742, in _apply_op_helper op = g._create_op_internal(op_type_name, inputs, dtypes=None, File "/home/shreyas/.local/lib/python3.8/site-packages/tensorflow/python/framework/ops.py", line 3319, in _create_op_internal ret = Operation( File "/home/shreyas/.local/lib/python3.8/site-packages/tensorflow/python/framework/ops.py", line 1816, in init self._c_op = _create_c_op(self._graph, node_def, inputs, File "/home/shreyas/.local/lib/python3.8/site-packages/tensorflow/python/framework/ops.py", line 1657, in _create_c_op raise ValueError(str(e)) ValueError: Dimensions must be equal, but are 32 and 33 for '{{node Add_135}} = Add[T=DT_FLOAT](Relu_132, BatchNormalization_134/add_1)' with input shapes: [1,128,32,32], [1,128,33,33].

Code to replicate issue

import onnx

from onnx_tf.backend import prepare

onnx_model = onnx.load("../margipose/pretrained/margipose2.onnx")  # load onnx model
tf_rep = prepare(onnx_model)  # prepare tf representation
tf_rep.export_graph("output_path")  # export the model
jaggernaut007 commented 4 years ago

issue fixed in onnx-tensorflow library.