local changes

  def __init__(self, max_local_steps=10, max_mutations=10):

    self.max_local_steps = max_local_steps

    self.max_mutations = max_mutations

    self.graph, self.input_placeholders, self.output_placeholder = self.construct_graph()

    self.graph, input_placeholders, self.label_placeholder, self.loss_op, self.train_op = (

        self.construct_graph())

    (self.protein_coords_placeholder, self.protein_Z_placeholder,

     self.ligand_coords_placeholder, self.ligand_Z_placeholder) = input_placeholders

    self.sess = tf.Session(graph=self.graph)

  def construct_graph(self, N_protein=1000, N_ligand=100, M=50, ndim=3, k=5, nbr_cutoff=6):

      loss = 0.5 * (energy - label_placeholder)**2

    return (graph, (protein_coords_placeholder, protein_Z_placeholder,

                    ligand_coords_placeholder, ligand_Z_placeholder), label_placeholder)

      train = self.optimizer.minimize(loss)

  def fit(self, dataset):

    return (graph,

            (protein_coords_placeholder, protein_Z_placeholder,

             ligand_coords_placeholder, ligand_Z_placeholder),

            label_placeholder,

            loss,

            train)

  def construct_feed_dict(self, X_protein, Z_protein, X_ligand, Z_ligand, y):

    """Create the feed dictionary for the fit() method."""

    feed_dict = {self.protein_coords_placeholder: X_protein,

                 self.protein_Z_placeholder: Z_protein,

                 self.ligand_coords_placeholder: X_ligand,

                 self.ligand_Z_placeholder: Z_ligand,

                 self.label_placeholder: y}

    return feed_dict

  def fit(self, X_protein, Z_protein, X_ligand, Z_ligand, y):

    """Fit to actual data."""

    # TODO(rbharath): Add an actual fit method.

    return

    feed_dict = self.construct_feed_dict(X_protein, Z_protein, X_ligand, Z_ligand, y)

    self.sess.run(self.train_op, feed_dict=feed_dict)

  def mutate_conformer(protein, ligand):

    """Performs a mutation on the ligand position."""