leandron commented on a change in pull request #7055:
URL: https://github.com/apache/tvm/pull/7055#discussion_r540287151
##########
File path: python/tvm/driver/tvmc/frontends.py
##########
@@ -161,14 +161,11 @@ def load(self, path):
# pylint: disable=E1101
model = onnx.load(path)
+ shape_dict = {}
# pylint: disable=E1101
- name = model.graph.input[0].name
-
- # pylint: disable=E1101
- proto_shape = model.graph.input[0].type.tensor_type.shape.dim
- shape = [d.dim_value for d in proto_shape]
-
- shape_dict = {name: shape}
+ for i in model.graph.input:
+ # pylint: disable=E1101
+ shape_dict[i.name] = [dim.dim_value for dim in
i.type.tensor_type.shape.dim]
Review comment:
My understanding is that we all agree two things here:
1) moving the current logic to to `from_onnx`, to be done in the current PR
(@dlexplorer )
2) a new feature to include `--input-shape` (in `tvmc compile` and `tvmc
tune` I think), to be submitted in a separate PR (@lixiaoquan)
Is that correct?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
