Caffe Data Layer

Data through the data layer into the Caffe network: The data layer is at the bottom of the network, the data can be read from a high-efficiency database (such as: LevelDB, LMDB), can be read directly from memory, if the requirements for reading and writing efficiency is not high can also be read from the hard disk hdft files or ordinary picture files.

The data comes from the database:

Layer Type: Data

Parameters that must be set:

Source: The directory name that contains the database, such as Examples/mnist/mnist_train_lmdb

Batch_size: The number of data processed each time, such as 64

Optional Parameters:

Rand_skip: At the beginning, passing in the input of some data. This is often useful for asynchronous SGD.

Backend: Leveldb,lmdb can be used, default is LEVELDB.

Example:

Layer {
  name: "Mnist"
  Type: "Data"
  Top: "Data"
  Top: "label"
  include {
    phase:train
  }
  Transform_param {
    scale:0.00390625
  }
  data_param {
    Source: "Examples/mnist/mnist_train_lmdb"
    batch_size:64
    backend:lmdb
  }
}

The data comes from memory:

Layer Type: Memorydata

Parameters that must be set:

Batch_size: The number of data processed each time, such as 2

Channels: Number of channels

Height: High

Width: Wide

Example:

Layer {
  top: "Data"
  Top: "Label"
  Name: "Memory_data"
  type: "Memorydata"
  memory_data_param{
    Batch_size:2
    height:100
    width:100
    channels:1
  }
  transform_param {
    scale:0.0078125
    mean_file: "Mean.proto"
    mirror:false
  }
}

The data comes from HDF5:

Layer Type: Hdf5data

Parameters that must be set:

Source: Read file name

Batch_size: Number of data processed each time

Example:

Layer {
  Name: "Data"
  type: "Hdf5data"
  Top: "Data"
  Top: "Label"
  hdf5_data_param {
    Source: " Examples/hdf5_classification/data/train.txt "
    batch_size:10
  }
}

The data comes from the picture:

Layer Type: ImageData

Parameters that must be set:

Source: The name of a text file, each line given a picture file names and labels (label)

Batch_size: Number of data processed each time, that is, pictures

Optional Parameters:

Rand_skip: At the beginning, passing in the input of some data. This is often useful for asynchronous SGD.

Shuffle: Random shuffle Order, default value False

New_height,new_width: If set, the picture is resize

Example:

Layer {
  Name: "Data"
  type: "ImageData"
  Top: "Data"
  Top: "Label"
  transform_param {
    Mirror: False
    crop_size:227
    mean_file: "Data/ilsvrc12/imagenet_mean.binaryproto"
  }
  Image_data_param {
    Source: "Examples/_temp/file_list.txt"
    batch_size:50
    new_height:256
    new_width:256
  }
}

The data comes from Windows:

Layer Type: Windowdata

Parameters that must be set:

Source: The name of a text file

Batch_size: Number of data processed each time, that is, pictures

Example:

Layer {
  Name: "Data"
  type: "Windowdata"
  Top: "Data"
  Top: "label"
  include {
    phase:train
  }
  Transform_param {
    mirror:true
    crop_size:227
    mean_file: "data/ilsvrc12/imagenet_ Mean.binaryproto "
  }
  Window_data_param {
    Source:" Examples/finetune_pascal_detection/window_file_ 2007_trainval.txt "
    batch_size:128
    fg_threshold:0.5
    bg_threshold:0.5
    fg_fraction:0.25
    Context_pad:16
    Crop_mode: "Warp"
  }
}

The data comes from dummy: Dummydata is mainly used for development and commissioning.