Incorrect shape size for NN input data

[RK900]

Tensorflow

cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=prediction,labels=y))

Logits and labels not of same size.

TFLearn

net = tflearn.input_data(shape=[None, 4716])

Shape size is incorrect for input matrix.

[RK900]

Reshape

np.reshape([-1,24])

[RK900]

YES IT WORKS!!!!

Fixes:

• n_classes = 2 to reflect moveset encoding
• data placeholder reshaped to [-1,16] because 2D shape needed, and x placeholder changed accordingly
• y placeholder should be [None,2] for [batch size, length]

n_classes = 2
batch_size = 100 # load 100 features at a time

x = tf.placeholder('float',[None,16]) 
y = tf.placeholder('float',[None,2])

enc = enc0.reshape([-1,16])
ms = ms0