python - Neural network for more than one class not working -

i trying use neural network classification problem. have 6 possible classes , same input may in more 1 class.

the problem when try train 1 nn each class, set output_num_units = 1 , on train, pass first column of y, y[:,0]. following output , error:

## layer information  #  name      size ---  ------  ------   0  input       32   1  dense0      32   2  output       1  indexerror: index 1 out of bounds axis 1 size 1 apply node caused error: crossentropycategorical1hot(elemwise{composite{scalar_sigmoid((i0 + i1))}}[(0, 0)].0, y_batch) inputs types: [tensortype(float32, matrix), tensortype(int32, vector)] inputs shapes: [(128, 1), (128,)] inputs strides: [(4, 4), (4,)] inputs values: ['not shown', 'not shown'] 

if try use output_num_units=num_class (6) , full y (all 6 fields), first error of kstratifiedfold, because seems not expect y have multiple rows. if set eval_size=none, following error:

typeerror: ('bad input argument theano function name "/usr/local/lib/python2.7/site-packages/nolearn-0.6a0.dev0-py2.7.egg/nolearn/lasagne/base.py:311"   @ index 1(0-based)', 'wrong number of dimensions: expected 1, got 2 shape (128, 6).') 

the configuration working setting more 1 output unit , passing 1 column y. trains nn, not seem right giving me 2 output layers, , have 1 y compare to.

what doing wrong? why can't use 1 output? should convert y classes vector of 6 columns vector of 1 column number?

i use following code (extract):

# load data data,labels = prepare_data_train('../input/train/subj1_series1_data.csv')  # x_train (119496, 32) <type 'numpy.ndarray'> x_train = data_preprocess_train(data) #print x_train.shape, type(x_train)  # y (119496, 6) <type 'numpy.ndarray'> y = labels.values.astype(np.int32) print y.shape, type(y)  # net config num_features = x_train.shape[1] num_classes = labels.shape[1]  # train neural net layers0 = [('input', inputlayer),            ('dense0', denselayer),            ('output', denselayer)]  net1 = neuralnet(     layers=layers0,      # layer parameters:     input_shape=(none, num_features),  # 32 input     dense0_num_units = 32,  # number of units in hidden layer     output_nonlinearity=sigmoid,  # sigmoid function has 1 class     output_num_units=2 ,  # if try 1, not work      # optimization method:     update=nesterov_momentum,     update_learning_rate=0.01,     update_momentum=0.9,      max_epochs=50,  # want train many epochs     verbose=1,     eval_size=0.2     )   net1.fit(x_train,  y[:,0]) 

i wanted use cnns in lasagne, didn't work same way, predictions 0... recommend @ the mnist example. find 1 better use , extend, old code snippets didn't work due api changes on time. i've amended mnist example, target vector has labels 0 or 1 , create output layer nn way:

# finally, we'll add fully-connected output layer, of 2 softmax units: l_out = lasagne.layers.denselayer(         l_hid2_drop, num_units=2,         nonlinearity=lasagne.nonlinearities.softmax) 

and cnn:

    layer = lasagne.layers.denselayer(         lasagne.layers.dropout(layer, p=.5),         num_units=2,         nonlinearity=lasagne.nonlinearities.softmax)