"Python" not accepted as layer type in the network protoxt

[Mohamed Ezz]

I'm trying to create a python data layer, and this is how I define the layer in the prototxt

layer {

  name: "data"

  type: "Python"

  top: "data"

  top: "label"

  python_param {

    module: "nifti_data_layer"

    layer: "NumpyTrainDataLayer"

  }

  include: { phase: TRAIN }

}

but when I load the prototxt it gives this error :

I0511 16:34:41.107846  8992 layer_factory.hpp:77] Creating layer data

F0511 16:34:41.107899  8992 layer_factory.hpp:81] Check failed: registry.count(type) == 1 (0 vs. 1) Unknown layer type: Python (known types: AbsVal, Accuracy, ArgMax, BNLL, BatchNorm, BatchReindex, Bias, Concat, ContrastiveLoss, Convolution, Crop, Data, Deconvolution, Dropout, DummyData, ELU, Eltwise, Embed, EuclideanLoss, Exp, Filter, Flatten, HDF5Data, HDF5Output, HingeLoss, Im2col, ImageData, InfogainLoss, InnerProduct, Input, LRN, Log, MVN, MemoryData, MultinomialLogisticLoss, PReLU, Parameter, Pooling, Power, ReLU, Reduction, Reshape, SPP, Scale, Sigmoid, SigmoidCrossEntropyLoss, Silence, Slice, Softmax, SoftmaxWithLoss, Split, TanH, Threshold, Tile, WindowData)

The python layer is intended to read images out of 3D .npy files. I'm using the HEAD of BVLC/Caffe.

I'm following the example prototxt here and code here as a reference. Any ideas why it does not work ?

[Mohamed Ezz]

It turns out that I had to re-build caffe with

WITH_PYTHON_LAYER = 1

in the Makefile.config