MCPcopy
hub / github.com/NVIDIA/Isaac-GR00T / set_runtime_tensor_shape

Method set_runtime_tensor_shape

scripts/deployment/trt_torch.py:131–133  ·  view source on GitHub ↗

Set runtime input shape for dynamic dimensions.

(self, name, shape)

Source from the content-addressed store, hash-verified

129 raise KeyError(f"Input tensor '{tensor_name}' not found in engine.")
130
131 def set_runtime_tensor_shape(self, name, shape):
132 """Set runtime input shape for dynamic dimensions."""
133 self.execution_context.set_input_shape(name, shape)
134
135 def forward(self, *args, **kwargs):
136 """Execute TRT inference with the given input tensors.

Callers 6

_qwen3_vit_and_scatterFunction · 0.80
mainFunction · 0.80

Calls

no outgoing calls

Tested by

no test coverage detected