nn_quantizer.py in the ML-CIFAR10 examples got error KeyError: 'accuracy'

Offline rafakath over 1 year ago

I am executing ML-examples ->cmsisnn-cifar10. While converting trained caffe model into cmsis-nn by below command

python nn_quantizer.py --model models/cifar10_m7_train_test.prototxt --weights models/cifar10_m7_iter_300000.caffemodel.h5 --save models/cifar10_m7.pkl

getting error message.

Traceback (most recent call last):
File "nn_quantizer.py", line 614, in <module>
my_model.get_graph_connectivity()
File "nn_quantizer.py", line 231, in get_graph_connectivity
num_branch = len(self.bottom_blob[current_layer])
KeyError: 'accuracy'

Kindly provide any refernce for solving above error.

Parents

0 Offline rafakath over 1 year ago

There was something wrong in prototxt file. The above error was solve. Thanks
Cancel
Up 0 Down

Reply

Accept answer

Cancel

0 Offline Sumant Sakhalkar 29 days ago in reply to rafakath

Hey rafakath, could you tell me what went wrong in your prototxt? I'm having the same issue but I can't seem to find anything wrong with my prototxt.

name: "VGGNet_xavier"
layer {
  name: "data"
  type: "Data"
  include {
    phase: TRAIN
  }
 transform_param {
    crop_size: 224
    mean_value: 104
    mean_value: 117
    mean_value: 123
    mirror: true
 }
 data_param {
    source: "/home/sumant/Project/ML-examples/cmsisnn-cifar10/data_set/train_lmdb"
    batch_size: 128
    backend: LMDB
  }
  top: "data"
  top: "label"
}
layer {
  name: "data"
  type: "Data"
  include {
    phase: TEST
  }
 transform_param {
    crop_size: 224
    mean_value: 104
    mean_value: 117
    mean_value: 123
    mirror: false
 }
 data_param {
    source: "/home/sumant/Project/ML-examples/cmsisnn-cifar10/data_set/val_lmdb"
    batch_size: 32
    backend: LMDB
  }
  top: "data"
  top: "label"
}
layer {
  name: "conv1_1"
  type: "Convolution"
  bottom: "data"
  top: "conv1_1"
  convolution_param {
    num_output: 64
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 8
    fl_layer_out: 6
    fl_params: 7
  }
}
layer {
  name: "relu1_1"
  type: "ReLU"
  bottom: "conv1_1"
  top: "conv1_1"
}
layer {
  name: "conv1_2"
  type: "Convolution"
  bottom: "conv1_1"
  top: "conv1_2"
  convolution_param {
    num_output: 64
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 6
    fl_layer_out: 3
    fl_params: 8
  }
}
layer {
  name: "relu1_2"
  type: "ReLU"
  bottom: "conv1_2"
  top: "conv1_2"
}
layer {
  name: "pool1"
  type: "Pooling"
  bottom: "conv1_2"
  top: "pool1"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "conv2_1"
  type: "Convolution"
  bottom: "pool1"
  top: "conv2_1"
  convolution_param {
    num_output: 128
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 3
    fl_layer_out: 2
    fl_params: 8
  }
}
layer {
  name: "relu2_1"
  type: "ReLU"
  bottom: "conv2_1"
  top: "conv2_1"
}
layer {
  name: "conv2_2"
  type: "Convolution"
  bottom: "conv2_1"
  top: "conv2_2"
  convolution_param {
    num_output: 128
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 2
    fl_layer_out: 2
    fl_params: 8
  }
}
layer {
  name: "relu2_2"
  type: "ReLU"
  bottom: "conv2_2"
  top: "conv2_2"
}
layer {
  name: "pool2"
  type: "Pooling"
  bottom: "conv2_2"
  top: "pool2"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "conv3_1"
  type: "Convolution"
  bottom: "pool2"
  top: "conv3_1"
  convolution_param {
    num_output: 256
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 2
    fl_layer_out: 1
    fl_params: 7
  }
}
layer {
  name: "relu3_1"
  type: "ReLU"
  bottom: "conv3_1"
  top: "conv3_1"
}
layer {
  name: "conv3_2"
  type: "Convolution"
  bottom: "conv3_1"
  top: "conv3_2"
  convolution_param {
    num_output: 256
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 1
    fl_layer_out: 1
    fl_params: 8
  }
}
layer {
  name: "relu3_2"
  type: "ReLU"
  bottom: "conv3_2"
  top: "conv3_2"
}
layer {
  name: "conv3_3"
  type: "Convolution"
  bottom: "conv3_2"
  top: "conv3_3"
  convolution_param {
    num_output: 256
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 1
    fl_layer_out: 1
    fl_params: 8
  }
}
layer {
  name: "relu3_3"
  type: "ReLU"
  bottom: "conv3_3"
  top: "conv3_3"
}
layer {
  name: "pool3"
  type: "Pooling"
  bottom: "conv3_3"
  top: "pool3"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "conv4_1"
  type: "Convolution"
  bottom: "pool3"
  top: "conv4_1"
  convolution_param {
    num_output: 512
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 1
    fl_layer_out: 1
    fl_params: 8
  }
}
layer {
  name: "relu4_1"
  type: "ReLU"
  bottom: "conv4_1"
  top: "conv4_1"
}
layer {
  name: "conv4_2"
  type: "Convolution"
  bottom: "conv4_1"
  top: "conv4_2"
  convolution_param {
    num_output: 512
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 1
    fl_layer_out: 2
    fl_params: 8
  }
}
layer {
  name: "relu4_2"
  type: "ReLU"
  bottom: "conv4_2"
  top: "conv4_2"
}
layer {
  name: "conv4_3"
  type: "Convolution"
  bottom: "conv4_2"
  top: "conv4_3"
  convolution_param {
    num_output: 512
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 2
    fl_layer_out: 3
    fl_params: 8
  }
}
layer {
  name: "relu4_3"
  type: "ReLU"
  bottom: "conv4_3"
  top: "conv4_3"
}
layer {
  name: "pool4"
  type: "Pooling"
  bottom: "conv4_3"
  top: "pool4"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "conv5_1"
  type: "Convolution"
  bottom: "pool4"
  top: "conv5_1"
  convolution_param {
    num_output: 512
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 3
    fl_layer_out: 4
    fl_params: 9
  }
}
layer {
  name: "relu5_1"
  type: "ReLU"
  bottom: "conv5_1"
  top: "conv5_1"
}
layer {
  name: "conv5_2"
  type: "Convolution"
  bottom: "conv5_1"
  top: "conv5_2"
  convolution_param {
    num_output: 512
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 4
    fl_layer_out: 5
    fl_params: 9
  }
}
layer {
  name: "relu5_2"
  type: "ReLU"
  bottom: "conv5_2"
  top: "conv5_2"
}
layer {
  name: "conv5_3"
  type: "Convolution"
  bottom: "conv5_2"
  top: "conv5_3"
  convolution_param {
    num_output: 512
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 5
    fl_layer_out: 6
    fl_params: 8
  }
}
layer {
  name: "relu5_3"
  type: "ReLU"
  bottom: "conv5_3"
  top: "conv5_3"
}
layer {
  name: "pool5"
  type: "Pooling"
  bottom: "conv5_3"
  top: "pool5"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "fc6"
  type: "InnerProduct"
  bottom: "pool5"
  top: "fc6"
  inner_product_param {
    num_output: 4096
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 4
    fl_layer_in: 6
    fl_layer_out: 8
    fl_params: 8
  }
}
layer {
  name: "relu6"
  type: "ReLU"
  bottom: "fc6"
  top: "fc6"
}
layer {
  name: "drop6"
  type: "Dropout"
  bottom: "fc6"
  top: "fc6"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layer {
  name: "fc7"
  type: "InnerProduct"
  bottom: "fc6"
  top: "fc7"
  inner_product_param {
    num_output: 4096
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 4
    fl_layer_in: 8
    fl_layer_out: 10
    fl_params: 7
  }
}
layer {
  name: "relu7"
  type: "ReLU"
  bottom: "fc7"
  top: "fc7"
}
layer {
  name: "drop7"
  type: "Dropout"
  bottom: "fc7"
  top: "fc7"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layer {
  name: "fc8"
  type: "InnerProduct"
  bottom: "fc7"
  top: "fc8"
  inner_product_param {
    num_output: 1000
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 4
    fl_layer_in: 10
    fl_layer_out: 10
    fl_params: 7
  }
}
layer {
  bottom: "fc8"
  top: "prob"
  name: "prob"
  type: "Softmax"
}
layer {
  name: "accuracy/top1"
  type: "Accuracy"
  bottom: "fc8"
  bottom: "label"
  top: "accuracy@1"
  include: { phase: TEST }
  accuracy_param {
    top_k: 1
  }
}
layer {
  name: "accuracy/top5"
  type: "Accuracy"
  bottom: "fc8"
  bottom: "label"
  top: "accuracy@5"
  include: { phase: TEST }
  accuracy_param {
    top_k: 5
  }
}

Reply

0 Offline Sumant Sakhalkar 29 days ago in reply to rafakath

Hey rafakath, could you tell me what went wrong in your prototxt? I'm having the same issue but I can't seem to find anything wrong with my prototxt.

name: "VGGNet_xavier"
layer {
  name: "data"
  type: "Data"
  include {
    phase: TRAIN
  }
 transform_param {
    crop_size: 224
    mean_value: 104
    mean_value: 117
    mean_value: 123
    mirror: true
 }
 data_param {
    source: "/home/sumant/Project/ML-examples/cmsisnn-cifar10/data_set/train_lmdb"
    batch_size: 128
    backend: LMDB
  }
  top: "data"
  top: "label"
}
layer {
  name: "data"
  type: "Data"
  include {
    phase: TEST
  }
 transform_param {
    crop_size: 224
    mean_value: 104
    mean_value: 117
    mean_value: 123
    mirror: false
 }
 data_param {
    source: "/home/sumant/Project/ML-examples/cmsisnn-cifar10/data_set/val_lmdb"
    batch_size: 32
    backend: LMDB
  }
  top: "data"
  top: "label"
}
layer {
  name: "conv1_1"
  type: "Convolution"
  bottom: "data"
  top: "conv1_1"
  convolution_param {
    num_output: 64
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 8
    fl_layer_out: 6
    fl_params: 7
  }
}
layer {
  name: "relu1_1"
  type: "ReLU"
  bottom: "conv1_1"
  top: "conv1_1"
}
layer {
  name: "conv1_2"
  type: "Convolution"
  bottom: "conv1_1"
  top: "conv1_2"
  convolution_param {
    num_output: 64
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 6
    fl_layer_out: 3
    fl_params: 8
  }
}
layer {
  name: "relu1_2"
  type: "ReLU"
  bottom: "conv1_2"
  top: "conv1_2"
}
layer {
  name: "pool1"
  type: "Pooling"
  bottom: "conv1_2"
  top: "pool1"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "conv2_1"
  type: "Convolution"
  bottom: "pool1"
  top: "conv2_1"
  convolution_param {
    num_output: 128
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 3
    fl_layer_out: 2
    fl_params: 8
  }
}
layer {
  name: "relu2_1"
  type: "ReLU"
  bottom: "conv2_1"
  top: "conv2_1"
}
layer {
  name: "conv2_2"
  type: "Convolution"
  bottom: "conv2_1"
  top: "conv2_2"
  convolution_param {
    num_output: 128
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 2
    fl_layer_out: 2
    fl_params: 8
  }
}
layer {
  name: "relu2_2"
  type: "ReLU"
  bottom: "conv2_2"
  top: "conv2_2"
}
layer {
  name: "pool2"
  type: "Pooling"
  bottom: "conv2_2"
  top: "pool2"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "conv3_1"
  type: "Convolution"
  bottom: "pool2"
  top: "conv3_1"
  convolution_param {
    num_output: 256
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 2
    fl_layer_out: 1
    fl_params: 7
  }
}
layer {
  name: "relu3_1"
  type: "ReLU"
  bottom: "conv3_1"
  top: "conv3_1"
}
layer {
  name: "conv3_2"
  type: "Convolution"
  bottom: "conv3_1"
  top: "conv3_2"
  convolution_param {
    num_output: 256
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 1
    fl_layer_out: 1
    fl_params: 8
  }
}
layer {
  name: "relu3_2"
  type: "ReLU"
  bottom: "conv3_2"
  top: "conv3_2"
}
layer {
  name: "conv3_3"
  type: "Convolution"
  bottom: "conv3_2"
  top: "conv3_3"
  convolution_param {
    num_output: 256
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 1
    fl_layer_out: 1
    fl_params: 8
  }
}
layer {
  name: "relu3_3"
  type: "ReLU"
  bottom: "conv3_3"
  top: "conv3_3"
}
layer {
  name: "pool3"
  type: "Pooling"
  bottom: "conv3_3"
  top: "pool3"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "conv4_1"
  type: "Convolution"
  bottom: "pool3"
  top: "conv4_1"
  convolution_param {
    num_output: 512
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 1
    fl_layer_out: 1
    fl_params: 8
  }
}
layer {
  name: "relu4_1"
  type: "ReLU"
  bottom: "conv4_1"
  top: "conv4_1"
}
layer {
  name: "conv4_2"
  type: "Convolution"
  bottom: "conv4_1"
  top: "conv4_2"
  convolution_param {
    num_output: 512
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 1
    fl_layer_out: 2
    fl_params: 8
  }
}
layer {
  name: "relu4_2"
  type: "ReLU"
  bottom: "conv4_2"
  top: "conv4_2"
}
layer {
  name: "conv4_3"
  type: "Convolution"
  bottom: "conv4_2"
  top: "conv4_3"
  convolution_param {
    num_output: 512
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 2
    fl_layer_out: 3
    fl_params: 8
  }
}
layer {
  name: "relu4_3"
  type: "ReLU"
  bottom: "conv4_3"
  top: "conv4_3"
}
layer {
  name: "pool4"
  type: "Pooling"
  bottom: "conv4_3"
  top: "pool4"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "conv5_1"
  type: "Convolution"
  bottom: "pool4"
  top: "conv5_1"
  convolution_param {
    num_output: 512
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 3
    fl_layer_out: 4
    fl_params: 9
  }
}
layer {
  name: "relu5_1"
  type: "ReLU"
  bottom: "conv5_1"
  top: "conv5_1"
}
layer {
  name: "conv5_2"
  type: "Convolution"
  bottom: "conv5_1"
  top: "conv5_2"
  convolution_param {
    num_output: 512
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 4
    fl_layer_out: 5
    fl_params: 9
  }
}
layer {
  name: "relu5_2"
  type: "ReLU"
  bottom: "conv5_2"
  top: "conv5_2"
}
layer {
  name: "conv5_3"
  type: "Convolution"
  bottom: "conv5_2"
  top: "conv5_3"
  convolution_param {
    num_output: 512
    pad: 1
    kernel_size: 3
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 8
    fl_layer_in: 5
    fl_layer_out: 6
    fl_params: 8
  }
}
layer {
  name: "relu5_3"
  type: "ReLU"
  bottom: "conv5_3"
  top: "conv5_3"
}
layer {
  name: "pool5"
  type: "Pooling"
  bottom: "conv5_3"
  top: "pool5"
  pooling_param {
    pool: MAX
    kernel_size: 2
    stride: 2
  }
}
layer {
  name: "fc6"
  type: "InnerProduct"
  bottom: "pool5"
  top: "fc6"
  inner_product_param {
    num_output: 4096
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 4
    fl_layer_in: 6
    fl_layer_out: 8
    fl_params: 8
  }
}
layer {
  name: "relu6"
  type: "ReLU"
  bottom: "fc6"
  top: "fc6"
}
layer {
  name: "drop6"
  type: "Dropout"
  bottom: "fc6"
  top: "fc6"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layer {
  name: "fc7"
  type: "InnerProduct"
  bottom: "fc6"
  top: "fc7"
  inner_product_param {
    num_output: 4096
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 4
    fl_layer_in: 8
    fl_layer_out: 10
    fl_params: 7
  }
}
layer {
  name: "relu7"
  type: "ReLU"
  bottom: "fc7"
  top: "fc7"
}
layer {
  name: "drop7"
  type: "Dropout"
  bottom: "fc7"
  top: "fc7"
  dropout_param {
    dropout_ratio: 0.5
  }
}
layer {
  name: "fc8"
  type: "InnerProduct"
  bottom: "fc7"
  top: "fc8"
  inner_product_param {
    num_output: 1000
    weight_filler {
      type: "xavier"
    }
    bias_filler {
      type: "constant"
      value: 0.1
    }
  }
  quantization_param {
    bw_layer_in: 16
    bw_layer_out: 16
    bw_params: 4
    fl_layer_in: 10
    fl_layer_out: 10
    fl_params: 7
  }
}
layer {
  bottom: "fc8"
  top: "prob"
  name: "prob"
  type: "Softmax"
}
layer {
  name: "accuracy/top1"
  type: "Accuracy"
  bottom: "fc8"
  bottom: "label"
  top: "accuracy@1"
  include: { phase: TEST }
  accuracy_param {
    top_k: 1
  }
}
layer {
  name: "accuracy/top5"
  type: "Accuracy"
  bottom: "fc8"
  bottom: "label"
  top: "accuracy@5"
  include: { phase: TEST }
  accuracy_param {
    top_k: 5
  }
}

Children

No data

nn_quantizer.py in the ML-CIFAR10 examples got error KeyError: 'accuracy'

Forum FAQs 0

Are there any machine learning libraries in C#? I'm after something like WEKA. 0

Great websites on AI and ML services 0

Cannot make inference with PyArmnn on custom quantized model trained with TF2 0

Need help with learning resource for ML on ARM based devices 0