Loss Function Is Returning Nan Tensorflow
I've written a simple tensorflow program here that reads in a feature list and tries to predict the class. with tf.Session() as sess: sess.run(tf.initialize_all_variables()
Solution 1:
- the inputs have nan's, fix it by
X[np.isnan(X)] = 0
. the inputs are not scaled, use sklearn's
StandardScaler
to normalize your inputs.Set the weights to a small initial value use stddev in random_normal.
- Fix the bug in calculation of output:
output = tf.add(tf.matmul(l3, output_layer['weights']),output_layer['biases'] )
.
Post a Comment for "Loss Function Is Returning Nan Tensorflow"