Issues regarding classification instead of regression using deep learning

Question

I have a question. I have a network which works fine I want to do the regression. However, when I am trying to use it for classification(after maing supposedly appropriate changes) I am facing some issues. I have 9 classes, but the issue is that the network is outputting me in a way that is not clear to me. It outputs me a 9x1 vector for each object which is fine but the values inside are not probabilities. I have tried converting the softmax output to probabilities(exp(1)/(exp(1)+..+exp(n))) but to no affect . I am using caffe an matcaffe . What I want is given the input the network tells me to which class it belongs. Basically in the output I want a single value which represents my class. I am attaching my prototxt file.`

name: "Zeiler_conv5"

input: "data"
input_dim: 1
input_dim: 3
input_dim: 224
input_dim: 224

input: "rois"
input_dim: 1 # to be changed on-the-fly to num ROIs
input_dim: 5 # [batch ind, x1, y1, x2, y2] zero-based indexing
input_dim: 1
input_dim: 1

input: "labels"
input_dim: 1 # to be changed on-the-fly to match num ROIs
input_dim: 1
input_dim: 1
input_dim: 1

input: "bbox_targets"
input_dim: 1  # to be changed on-the-fly to match num ROIs
input_dim: 84 # 4 * (K+1) (=21) classes
input_dim: 1
input_dim: 1

input: "bbox_loss_weights"
input_dim: 1  # to be changed on-the-fly to match num ROIs
input_dim: 84 # 4 * (K+1) (=21) classes
input_dim: 1
input_dim: 1

input: "angle_head"
input_dim: 1  # to be changed on-the-fly to match num ROIs
input_dim: 9 # 9 (-180:45:180) classes
input_dim: 1
input_dim: 1

input: "angle_head_weight"
input_dim: 1  # to be changed on-the-fly to match num ROIs
input_dim: 9 # 9 (-180:45:180) classes
input_dim: 1
input_dim: 1


layer {
    name: "conv1"
    type: "Convolution"
    bottom: "data"
    top: "conv1"
    param {
        lr_mult: 0.0
    }
    param {
        lr_mult: 0.0
    }
    convolution_param {
        num_output: 96
        kernel_size: 7
        pad: 3
        stride: 2
        weight_filler {
            type: "gaussian"
            std: 0.01
        }
        bias_filler {
            type: "constant"
            value: 0
        }
    }
}

layer {
    name: "relu1"
    type: "ReLU"
    bottom: "conv1"
    top: "conv1"
}

layer {
    name: "norm1"
    type: "LRN"
    bottom: "conv1"
    top: "norm1"
    lrn_param {
        local_size: 3
        alpha: 0.00005
        beta: 0.75
        norm_region: WITHIN_CHANNEL
    }
}

layer {
    name: "pool1"
    type: "Pooling"
    bottom: "norm1"
    top: "pool1"
    pooling_param {
        kernel_size: 3
        stride: 2
        pad: 1
        pool: MAX
    }
}

layer {
    name: "conv2"
    type: "Convolution"
    bottom: "pool1"
    top: "conv2"
    param {
        lr_mult: 0.0
    }
    param {
        lr_mult: 0.0
    }
    convolution_param {
        num_output: 256
        kernel_size: 5
        pad: 2
        stride: 2
        weight_filler {
            type: "gaussian"
            std: 0.01
        }
        bias_filler {
            type: "constant"
            value: 1
        }
    }
}

layer {
    name: "relu2"
    type: "ReLU"
    bottom: "conv2"
    top: "conv2"
}

layer {
    name: "norm2"
    type: "LRN"
    bottom: "conv2"
    top: "norm2"
    lrn_param {
        local_size: 3
        alpha: 0.00005
        beta: 0.75
        norm_region: WITHIN_CHANNEL
    }
}

layer {
    name: "pool2"
    type: "Pooling"
    bottom: "norm2"
    top: "pool2"
    pooling_param {
        kernel_size: 3
        stride: 2
        pad: 1
        pool: MAX
    }
}

layer {
    name: "conv3"
    type: "Convolution"
    bottom: "pool2"
    top: "conv3"
    param {
        lr_mult: 0.0
    }
    param {
        lr_mult: 0.0
    }
    convolution_param {
        num_output: 384
        kernel_size: 3
        pad: 1
        stride: 1
        weight_filler {
            type: "gaussian"
            std: 0.01
        }
        bias_filler {
            type: "constant"
            value: 0
        }
    }
}

layer {
    name: "relu3"
    type: "ReLU"
    bottom: "conv3"
    top: "conv3"
}

layer {
    name: "conv4"
    type: "Convolution"
    bottom: "conv3"
    top: "conv4"
    param {
        lr_mult: 0.0
    }
    param {
        lr_mult: 0.0
    }
    convolution_param {
        num_output: 384
        kernel_size: 3
        pad: 1
        stride: 1
        weight_filler {
            type: "gaussian"
            std: 0.01
        }
        bias_filler {
            type: "constant"
            value: 1
        }
    }
}

layer {
    name: "relu4"
    type: "ReLU"
    bottom: "conv4"
    top: "conv4"
}

layer {
    name: "conv5"
    type: "Convolution"
    bottom: "conv4"
    top: "conv5"
    param {
        lr_mult: 0.0
    }
    param {
        lr_mult: 0.0
    }
    convolution_param {
        num_output: 256
        kernel_size: 3
        pad: 1
        stride: 1
        weight_filler {
            type: "gaussian"
            std: 0.01
        }
        bias_filler {
            type: "constant"
            value: 1
        }
    }
}

layer {
    name: "relu5"
    type: "ReLU"
    bottom: "conv5"
    top: "conv5"
}

layer {
    bottom: "conv5"
    bottom: "rois"
    top: "pool5"
    name: "roi_pool5"
    type: "ROIPooling"
    roi_pooling_param {
        pooled_w: 6
        pooled_h: 6
        spatial_scale: 0.0625  # (1/16)
    }
}

layer {
    bottom: "pool5"
    top: "fc6"
    name: "fc6"
    param {
        lr_mult: 1.0
    }
    param {
        lr_mult: 2.0
    }
    type: "InnerProduct"
    inner_product_param {
        num_output: 4096
    }
}

layer {
    bottom: "fc6"
    top: "fc6"
    name: "relu6"
    type: "ReLU"
}

layer {
    bottom: "fc6"
    top: "fc6"
    name: "drop6"
    type: "Dropout"
    dropout_param {
        dropout_ratio: 0.5
        scale_train: false
    }
}

layer {
    bottom: "fc6"
    top: "fc7"
    name: "fc7"
    param {
        lr_mult: 1.0
    }
    param {
        lr_mult: 2.0
    }
    type: "InnerProduct"
    inner_product_param {
        num_output: 4096
    }
}

layer {
    bottom: "fc7"
    top: "fc7"
    name: "relu7"
    type: "ReLU"
}

layer {
    bottom: "fc7"
    top: "fc7"
    name: "drop7"
    type: "Dropout"
    dropout_param {
        dropout_ratio: 0.5
        scale_train: false
    }
}

layer {
    bottom: "fc7"
    top: "cls_score"
    name: "cls_score"
    param {
        lr_mult: 1.0
    }
    param {
        lr_mult: 2.0
    }
    type: "InnerProduct"
    inner_product_param {
        num_output: 21
        weight_filler {
            type: "gaussian"
            std: 0.01
        }
        bias_filler {
            type: "constant"
            value: 0
        }
    }
}

layer {
    bottom: "fc7"
    top: "angle_pred"
    name: "angle_pred"
    type: "InnerProduct"
    param {
        lr_mult: 1.0
    }
    param {
        lr_mult: 2.0
    }
    inner_product_param {
        num_output: 9
        weight_filler {
            type: "gaussian"
            std: 0.001
        }
        bias_filler {
            type: "constant"
            value: 0
        }
    }
}

layer {
    bottom: "fc7"
    top: "bbox_pred"
    name: "bbox_pred"
    type: "InnerProduct"
    param {
        lr_mult: 1.0
    }
    param {
        lr_mult: 2.0
    }
    inner_product_param {
        num_output: 84
        weight_filler {
            type: "gaussian"
            std: 0.001
        }
        bias_filler {
            type: "constant"
            value: 0
        }
    }
}

layer {
    name: "loss"
    type: "SoftmaxWithLoss"
    bottom: "cls_score"
    bottom: "labels"
    top: "loss_cls"
    loss_weight: 0
}

layer {
    name: "accuarcy"
    type: "Accuracy"
    bottom: "cls_score"
    bottom: "labels"
    top: "accuarcy"
}

layer {
    name: "loss_angle"
    type: "SmoothL1Loss"
    bottom: "angle_pred"
    bottom: "angle_head"
    bottom: "angle_head_weight"
    top: "loss_angle"
    loss_weight: 1
}

layer {
    name: "loss_bbox"
    type: "SmoothL1Loss"
    bottom: "bbox_pred"
    bottom: "bbox_targets"
    bottom: "bbox_loss_weights"
    top: "loss_bbox"
    loss_weight: 0
}

`

In particular I am speaking about "angle_pred" – lora loper Oct 23 '16 at 18:42 — lora loper, Oct 23 '16 at 18:42

score 0 · Answer 1 · answered Oct 24 '16 at 01:09

0

You have used softmax function in the output layer to convert the scores generated by the neural network model to probability. Now you need to consider the max probability and the class associated with the max probability is your answer. By the way what did you mean by that, softmax function has no effect? Softmax function will also give you a vector, instead of a single value. You can decide from the probabilities what is the final class predicted by your classifier.

answered Oct 24 '16 at 01:09

Wasi Ahmad

35,739
32
114
161

I did mention in my problem statement, "It outputs me a 9x1 vector for each object which is fine but the values inside are not probabilities" The values are ranging from -180 to some positive integer. Whereas, probabilities should oscillate between [0 1]. – lora loper Oct 24 '16 at 07:10
Please check your softmax implementation then. I guess there is some mistake because its not possible that softmax is returning you a vector of unbounded values. The values should be in between 0-1 because the expression of softmax function includes normalization. – Wasi Ahmad Oct 24 '16 at 07:31
Ok let me debug. Thanks anyways. – lora loper Oct 24 '16 at 10:17

score 0 · Accepted Answer · answered Oct 25 '16 at 07:51

0

I think the file you have uploaded is the training prototxt file. You have used SoftmaxWithLoss layer. This layer will not give you the probabilities. Replace it with SoftMax layer during deployment to get the probability of each class.

answered Oct 25 '16 at 07:51

Krishna Kishore Andhavarapu

828
5
12

Just to confirm by "deployment" you mean TEST ? – lora loper Oct 25 '16 at 11:50
yeah.. test prototxt – Krishna Kishore Andhavarapu Oct 25 '16 at 11:53

Issues regarding classification instead of regression using deep learning

2 Answers2