Forums - memory error while converting quantized ssd model

1 post / 0 new
or Register
memory error while converting quantized ssd model
neeraj.partha
Join Date: 1 Aug 19
Posts: 4
Posted: Sun, 2020-09-27 01:55
Top
2020-09-27 12:04:46,880 - 172 - INFO - INFO_ALL_BUILDING_NETWORK: 
==============================================================
Building Network
==============================================================
Followed the exact same procedure as given in this link
ssd_mobilenet_v2_quantized_300x300_coco_2019_01_03
 
2020-09-27 14:06:52,146 - 167 - ERROR - Conversion FAILED!
Traceback (most recent call last):
  File "/home/partha/Downloads/snpe-1.42/bin/x86_64-linux-clang/snpe-tensorflow-to-dlc", line 52, in <module>
    main()
  File "/home/partha/Downloads/snpe-1.42/bin/x86_64-linux-clang/snpe-tensorflow-to-dlc", line 48, in main
    raise e
  File "/home/partha/Downloads/snpe-1.42/bin/x86_64-linux-clang/snpe-tensorflow-to-dlc", line 38, in main
    ir_graph = converter.convert()
  File "/home/partha/Downloads/snpe-1.42/lib/python/qti/aisw/converters/tensorflow/tf_to_ir.py", line 331, in convert
    self._convert_layers()
  File "/home/partha/Downloads/snpe-1.42/lib/python/qti/aisw/converters/tensorflow/tf_to_ir.py", line 369, in _convert_layers
    descriptors = self._resolve_descriptors_from_nodes(graph_ops)
  File "/home/partha/Downloads/snpe-1.42/lib/python/qti/aisw/converters/tensorflow/tf_to_ir.py", line 507, in _resolve_descriptors_from_nodes
    resolved_descriptors = resolver.resolve_layer(graph_matcher, self._graph_helper)
  File "/home/partha/Downloads/snpe-1.42/lib/python/qti/aisw/converters/tensorflow/layers/crop.py", line 46, in resolve_layer
    size, size_shape, size_consumed_nodes = graph_helper.get_static_data_info(size_tensor)
  File "/home/partha/Downloads/snpe-1.42/lib/python/qti/aisw/converters/tensorflow/util.py", line 457, in get_static_data_info
    queue.extend(head.op.inputs)
MemoryError
 
  • Up0
  • Down0

Opinions expressed in the content posted here are the personal opinions of the original authors, and do not necessarily reflect those of Qualcomm Incorporated or its subsidiaries (“Qualcomm”). The content is provided for informational purposes only and is not meant to be an endorsement or representation by Qualcomm or any other party. This site may also provide links or references to non-Qualcomm sites and resources. Qualcomm makes no representations, warranties, or other commitments whatsoever about any non-Qualcomm sites or third-party resources that may be referenced, accessible from, or linked to this site.