MCPcopy
hub / github.com/NVIDIA/TensorRT-LLM / get_tensor

Function get_tensor

examples/quantization/quantize_mixed_precision_moe.py:185–190  ·  view source on GitHub ↗
(name)

Source from the content-addressed store, hash-verified

183 end_layer = min(num_layer, args.rank * part_layer + part_layer)
184
185 def get_tensor(name):
186 if name not in weight_map:
187 return None
188 ff = weight_map[name]
189 safetensors_loader = processed_files[ff]
190 return safetensors_loader.get_tensor(name).cuda()
191
192 def get_file_name(layer):
193 rank = layer // part_layer

Callers 1

mainFunction · 0.85

Calls 1

get_tensorMethod · 0.80

Tested by

no test coverage detected