Skip to content

Instantly share code, notes, and snippets.

@rpitonak

rpitonak/resize_constants.py

Last active July 12, 2022 07:48
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save rpitonak/68f3ee992131dba632776a3dab233b40 to your computer and use it in GitHub Desktop.
Save rpitonak/68f3ee992131dba632776a3dab233b40 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
resize_constants.py
resizer_node_output_name = "Resize_Y"
Y_resize = onnx.helper.make_tensor_value_info(resizer_node_output_name,
onnx.TensorProto.FLOAT,
[None, 224, 224, 3])
roi = onnx.helper.make_node("Constant", inputs=[], outputs=["roi"], name="roi-constant",
value=onnx.helper.make_tensor(name="roi-values",
data_type=onnx.TensorProto.FLOAT,
dims=np.array([]).shape,
vals=np.array([]).flatten()
))
scales = onnx.helper.make_node("Constant", inputs=[], outputs=["scales"], name="scales-constant",
value=onnx.helper.make_tensor(name="scales-values",
data_type=onnx.TensorProto.FLOAT,
dims=np.array([]).shape,
vals=np.array([]).flatten()
))
output_size = onnx.helper.make_node("Constant", inputs=[], outputs=["output_size"], name="output_size-constant",
value=onnx.helper.make_tensor(name="output_size-values",
data_type=onnx.TensorProto.INT64,
dims=np.array([1, 224, 224, 3]).shape,
vals=np.array([1, 224, 224, 3]).flatten()
))
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment