Closed RK900 closed 7 years ago
Reshape
np.reshape([-1,24])
YES IT WORKS!!!!
Fixes:
n_classes = 2
to reflect moveset encoding[-1,16]
because 2D shape needed, and x placeholder changed accordingly[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
Tensorflow
Logits and labels not of same size.
TFLearn
Shape size is incorrect for input matrix.