AM5749: Cannot quantize Crop layer in TIDL

Hi,

I'm trying to quantize a model with a CropLayer. According to the TIDL documentation it should be supported.

Here's my deploy.prototxt:

name: "graph_deploy"
input: "data"
input_shape {
  dim: 1
  dim: 3
  dim: 32
  dim: 32
}
layer {
  name: "data/bias"
  type: "Bias"
  bottom: "data"
  top: "data/bias"
  param {
    lr_mult: 0.0
    decay_mult: 0.0
  }
  bias_param {
    filler {
      type: "constant"
      value: 0.0
    }
  }
}
layer {
  name: "conv1"
  type: "Convolution"
  bottom: "data/bias"
  top: "conv1"
  param {
    lr_mult: 1.0
    decay_mult: 1.0
  }
  param {
    lr_mult: 2.0
    decay_mult: 0.0
  }
  convolution_param {
    num_output: 3
    bias_term: true
    pad: 3
    kernel_size: 7
    group: 1
    stride: 1
    weight_filler {
      type: "msra"
    }
    bias_filler {
      type: "constant"
      value: 0.0
    }
    dilation: 1
  }
}
layer {
  name: "conv1/relu"
  type: "ReLU"
  bottom: "conv1"
  top: "conv1"
}
layer {
  name: "crop_dummy_data"
  type: "DummyData"
  top: "crop_dummy_data"
  dummy_data_param {
    data_filler {
      type: "constant"
      value: 0.0
    }
    shape {
      dim: 1
      dim: 3
      dim: 10
      dim: 10
    }
  }
}
layer {
  name: "crop1"
  type: "Crop"
  bottom: "conv1"
  bottom: "crop_dummy_data"
  top: "crop1"
  crop_param {
    axis: 2
  }
}
layer {
  name: "flatten1"
  type: "Flatten"
  bottom: "crop1"
  top: "flatten1"
  flatten_param {
    axis: 1
  }
}
layer {
  name: "fc10"
  type: "InnerProduct"
  bottom: "flatten1"
  top: "fc10"
  param {
    lr_mult: 1.0
    decay_mult: 1.0
  }
  param {
    lr_mult: 2.0
    decay_mult: 0.0
  }
  inner_product_param {
    num_output: 10
    weight_filler {
      type: "msra"
    }
    bias_filler {
      type: "constant"
      value: 0.0
    }
  }
}

And here's the result:

Caffe Network File : models/cj1718/deploy.prototxt
Caffe Model File : models/cj1718/snapshot.caffemodel
TIDL Network File : models/cj1718/tidl_net.bin
TIDL Model File : models/cj1718/tidl_params.bin
Name of the Network : graph_deploy
Num Inputs : 1
[libprotobuf FATAL /oe/bld/build-CORTEX_1/arago-tmp-external-linaro-toolchain/work/x86_64-nativesdk-arago-linux/nativesdk-tidl-import/01.00.00.00-r1/recipe-sysroot-native/usr/include/google/protobuf/repeated_field.h:1478] CHECK failed: (index) < (current_size_):
terminate called after throwing an instance of 'google::protobuf::FatalException'
what(): CHECK failed: (index) < (current_size_):

What am I doing wrong? Is using a DummyDataLayer as the second blob the supposed way to do this?

Thanks in advance,

Gabriele