I m using Tenserflow 1.x by google colab but why i m getting error in that

Question

Code:

weights1 = tf.get_variable("weights1",shape=[12,80],initializer = tf.contrib.layers.xavier_initializer())
biases1 = tf.get_variable("biases1",shape = [80],initializer = tf.zeros_initializer)
layer1out = tf.nn.relu(tf.matmul(X,weights1)+biases1)

weights2 = tf.get_variable("weights2",shape=[80,50],initializer = tf.contrib.layers.xavier_initializer())
biases2 = tf.get_variable("biases2",shape = [50],initializer = tf.zeros_initializer)
layer2out = tf.nn.relu(tf.matmul(layer1out,weights2)+biases2)

weights3 = tf.get_variable("weights3",shape=[50,3],initializer = tf.contrib.layers.xavier_initializer())
biases3 = tf.get_variable("biases3",shape = [3],initializer = tf.zeros_initializer)
prediction =tf.matmul(layer2out,weights3)+biases3strong text

Error:

value error:Variable weights1 already exists, disallowed. Did you mean to set reuse=True or reuse=tf.AUTO_REUSE in VarScope? Originally defined at: