Using Convolution Neural Net with Lasagne in Python error -

i have used frame work provided daniel nouri on eponymous website. here code used.it looks fine change made change output_nonlinearity=lasagne.nonlinearities.softmax , regression false.otherwise looks pretty straight forward

from lasagne import layers import theano lasagne.updates import sgd,nesterov_momentum nolearn.lasagne import neuralnet sklearn.metrics import classification_report import lasagne import cv2 import numpy np sklearn.cross_validation import train_test_split sklearn.datasets import fetch_mldata import sys  mnist = fetch_mldata('mnist original') x = np.asarray(mnist.data, dtype='float32') y = np.asarray(mnist.target, dtype='int32')  (trainx, testx, trainy, testy) = train_test_split(x,y,test_size =0.3,random_state=42) trainx = trainx.reshape(-1, 1, 28, 28) testx = testx.reshape(-1, 1, 28, 28)  clf = neuralnet(     layers=[     ('input', layers.inputlayer),     ('conv1', layers.conv2dlayer),     ('pool1', layers.maxpool2dlayer),     ('dropout1', layers.dropoutlayer),  # !     ('conv2', layers.conv2dlayer),     ('pool2', layers.maxpool2dlayer),     ('dropout2', layers.dropoutlayer),  # !     ('hidden4', layers.denselayer),     ('dropout4', layers.dropoutlayer),  # !     ('hidden5', layers.denselayer),     ('output', layers.denselayer),     ],  input_shape=(none,1, 28, 28),  conv1_num_filters=20, conv1_filter_size=(3, 3), pool1_pool_size=(2, 2),  dropout1_p=0.1,  # !  conv2_num_filters=50, conv2_filter_size=(3, 3), pool2_pool_size=(2, 2),  dropout2_p=0.2,  # !  hidden4_num_units=500,  dropout4_p=0.5,  # !  hidden5_num_units=500,   output_num_units=10,   output_nonlinearity=lasagne.nonlinearities.softmax,   update=nesterov_momentum,   update_learning_rate=theano.shared(float32(0.03)),  update_momentum=theano.shared(float32(0.9)),   regression=false,  max_epochs=3000,  verbose=1,  )  clf.fit(trainx,trainy) 

however on running nan

input               (none, 1, 28, 28)       produces     784 outputs conv1               (none, 20, 26, 26)      produces   13520 outputs pool1               (none, 20, 13, 13)      produces    3380 outputs dropout1            (none, 20, 13, 13)      produces    3380 outputs conv2               (none, 50, 11, 11)      produces    6050 outputs pool2               (none, 50, 6, 6)        produces    1800 outputs dropout2            (none, 50, 6, 6)        produces    1800 outputs hidden4             (none, 500)             produces     500 outputs dropout4            (none, 500)             produces     500 outputs hidden5             (none, 500)             produces     500 outputs output              (none, 10)              produces      10 outputs epoch    train loss    valid loss    train/val    valid acc  dur -------  ------------  ------------  -----------  -----------  ------   1           nan           nan          nan      0.09923  16.18s   2           nan           nan          nan      0.09923  16.45s 

thanks in advance.

i'm late game, finds answer useful!

in experience, there number of things going wrong here. i'll write out steps debugging kind of problem in nolearn/lasagne:

1. using theano's fast_compile optimizer can lead underflow issues, result in nan output (this ultimate problem in case)

2. when output starts nan values, or if nan values start appearing after training starts, learning rate may high. if 0.01, try , make 0.001.

3. the input or output values may close 1 another, , may want try scaling them. standard approach scale input subtracting mean , dividing standard deviation.

4. make sure using regression=true when using nolearn regression problem

5. try using linear output instead of softmax. other nonlinearities help, in experience not often.

6. if fails, try , isolate whether issue network or data. if feed in random values within expected range , still nan output, it's not specific dataset training on.

hope helps!