Forums - DLC Conversion Error from Tensorflow

1 post / 0 new
DLC Conversion Error from Tensorflow
pallavisaha321
Join Date: 8 Jun 20
Posts: 1
Posted: Mon, 2020-06-15 22:09

I am trying to convert a skeleton detector tensorflow based model to DLC.  I am getting the below error during conversion.

 

     [[Node: MobilenetV1/Conv2D = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1, 1], explicit_paddings=[], padding="SAME", strides=[1, 2, 2, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_image_0_0, MobilenetV1/Conv2d_0/weights)]]
2020-06-16 10:29:33,584 - 166 - ERROR - %
Traceback (most recent call last):
  File "/home/pallavi/snpe-sdk/lib/python/qti/aisw/converters/tensorflow/tf_to_ir.py", line 525, in _create_layer
    layer_builder.build_layer(self.graph, self._context, descriptor, inputs, outputs)
  File "/home/pallavi/snpe-sdk/lib/python/qti/aisw/converters/tensorflow/layers/convolution.py", line 472, in build_layer
    descriptor.dilationX])
  File "/home/pallavi/snpe-sdk/lib/python/qti/aisw/converters/tensorflow/layers/convolution.py", line 502, in calculate_padding_size
    pad_y = max(((output_size[0] - 1) * strides[0] + filter_h - input_size[0]), 0)
IndexError: list index out of range
2020-06-16 10:29:33,589 - 166 - ERROR - Conversion failed: build_layer failed with Exception list index out of range in layer ConvolutionLayerBuilder.

 

Could you please suggest me the next steps.

  • 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.