diff options
Diffstat (limited to 'ethosu/vela/npu_serialisation.py')
-rw-r--r-- | ethosu/vela/npu_serialisation.py | 8 |
1 files changed, 5 insertions, 3 deletions
diff --git a/ethosu/vela/npu_serialisation.py b/ethosu/vela/npu_serialisation.py index 0cb40ed0..08dc0d38 100644 --- a/ethosu/vela/npu_serialisation.py +++ b/ethosu/vela/npu_serialisation.py @@ -46,11 +46,13 @@ def copy_compressed_values_to_memory_tensor(memory_tensor, src_tensor): memory_tensor.values[start_addr:end_addr] = compressed_values start_addr = end_addr + def copy_ifm_values_to_memory_tensor(memory_tensor, src_tensor): start_addr = src_tensor.address end_addr = start_addr + src_tensor.quant_values.size memory_tensor.values[start_addr:end_addr] = src_tensor.quant_values + def serialise_npu_subgraph_into_tensors(nng, sg, arch, scratch_tens, flash_tens): if sg.placement != PassPlacement.Npu: return scratch_tens, flash_tens @@ -95,7 +97,7 @@ def serialise_npu_subgraph_into_tensors(nng, sg, arch, scratch_tens, flash_tens) for cps in sg.cascaded_passes: for ps in cps.passes: if ps.placement == PassPlacement.Npu: - if ps.weight_tensor != None: + if ps.weight_tensor is not None: # For DMA ops, ps.weight_tensor is referring to the SRAM weight tensor and therefore the address # is pointing at the destination address of where the weights should be placed in SRAM. # This ensures that the Flash weight tensor is used instead and thus gets the correct address. @@ -106,9 +108,9 @@ def serialise_npu_subgraph_into_tensors(nng, sg, arch, scratch_tens, flash_tens) copy_compressed_values_to_memory_tensor(sg.flash_tensor, ps.scale_tensor) - if ps.ifm_tensor != None and ps.ifm_tensor.mem_area != MemArea.Sram: + if ps.ifm_tensor is not None and ps.ifm_tensor.mem_area != MemArea.Sram: copy_ifm_values_to_memory_tensor(sg.flash_tensor, ps.ifm_tensor) - if ps.ifm2_tensor != None and ps.ifm2_tensor.mem_area != MemArea.Sram: + if ps.ifm2_tensor is not None and ps.ifm2_tensor.mem_area != MemArea.Sram: copy_ifm_values_to_memory_tensor(sg.flash_tensor, ps.ifm2_tensor) sg.command_stream_tensor = make_memory_tensor( |