compara-deep-learning/model.py

dim=30

import tensorflow as tf

def create_model():
    tf.reset_default_graph()
    g=tf.Graph()
    with g.as_default():
        synm=tf.placeholder(dtype=tf.float64,shape=(None,5,5,2),name="Synteny_matrix_placeholder")
        bls=tf.placeholder(dtype=tf.float64,shape=(None,21),name="Species_Branch_Length_Placeholder")
        blhs=tf.placeholder(dtype=tf.float64,shape=(None,21),name="Homology_Species_Branch_Length_Placeholder")
        gl=tf.placeholder(dtype=tf.float64,shape=(None,1),name="Mean_gene_length")
        dps=tf.placeholder(dtype=tf.float64,shape=(None,1),name="mca_species_distance")
        dphs=tf.placeholder(dtype=tf.float64,shape=(None,1),name="mca_homology_species_distance")
        dis=tf.placeholder(dtype=tf.float64,shape=(None,1),name="total_distance")
        lr=tf.placeholder_with_default(0.1,(),"learning_rate")
        y=tf.placeholder(dtype=tf.int32,shape=(None),name="labels")

        #reshape the variables
        #gl=tf.expand_dims(gl,-1)
        #dps=tf.expand_dims(dps,-1)
        #dphs=tf.expand_dims(dphs,-1)
        #dis=tf.expand_dims(dis,-1)

        x=tf.concat([gl,dps,dphs,dis],1,name="Create_train_vector")

        print(synm,"\n",bls,"\n",blhs,"\n",gl,"\n",dps,"\n",dphs,"\n",dis,"\n",x)

        reg=tf.contrib.layers.l2_regularizer(0.01)

        def get_variable_by_shape(shape,name):
            f=tf.get_variable(name,shape=shape,initializer=tf.glorot_uniform_initializer(),dtype=tf.float64,regularizer=reg)
            return f

        with tf.variable_scope("Synteny_Aligner",reuse=tf.AUTO_REUSE):

            fconv=get_variable_by_shape((2,2,2,dim),"fconv")
            conv=tf.nn.conv2d(synm,fconv,(1,1,1,1),padding="VALID",name="Conv_aligner")

            fconv_1=get_variable_by_shape((2,2,dim,dim*2),"fconv_1")
            conv_1=tf.nn.conv2d(conv,fconv_1,(1,1,1,1),padding="VALID",name="Conv_aligner_1")

            print(conv)
            print(conv_1)
            #fxconv=get_variable_by_shape((1,5,2,dim),"fxconv")
            fxconv=get_variable_by_shape((5,2,dim*2),"fxconv")
            x_conv=tf.reshape(synm,(-1,25,2))
            x_conv=tf.nn.conv1d(x_conv,fxconv,stride=5,padding="SAME",name="row_aligner")
            print(x_conv)
            #fyconv=get_variable_by_shape((5,1,2,dim),"fyconv")
            #y_conv=tf.nn.conv2d(synm,fyconv,(1,1,1,1),padding="SAME",name="column_aligner")
            y_conv=tf.reshape(tf.transpose(synm,(0,2,1,3)),(-1,25,2))
            print(y_conv)
            fyconv=get_variable_by_shape((5,2,dim*2),"fyconv")
            y_conv=tf.nn.conv1d(y_conv,fyconv,stride=5,padding="SAME",name="column_aligner")

            print(y_conv)
            conv_1=tf.reshape(conv_1,(-1,9,dim*2))
            print(conv)
            x_conv=tf.reshape(x_conv,(-1,5,dim*2))
            print(x_conv)
            y_conv=tf.reshape(y_conv,(-1,5,dim*2))
            print(y_conv)
            conv_final=tf.concat([conv_1,x_conv,y_conv],1,name="Concatenate_All_Alignments")
            print(conv_final)

            W=get_variable_by_shape((5,5,2),"W")
            print(W)
            W=W*synm
            print(W)
            #W=tf.reduce_sum(W,[1,2,3])
            #W=tf.reshape(W,(-1,1))
            #print(W)


        with tf.variable_scope("Combine_Renormalize",reuse=tf.AUTO_REUSE):
            bl=tf.concat([bls,blhs],1)
            theta_bl=get_variable_by_shape((42,1),"theta_bl")
            theta_bl=tf.matmul(bl,theta_bl)
            x=tf.concat([x,theta_bl],1)
            theta=get_variable_by_shape((5,19),"theta")
            bias=get_variable_by_shape((1,19),"b")
            theta_2=tf.matmul(x,theta)+bias
            print(theta_2)
            theta_2=tf.reshape(theta_2,(-1,19,1))
            theta_2=tf.tile(theta_2,[1,1,dim*2])
            print(theta_2)
            conv_final=conv_final*theta_2
            print(conv_final)

        flat=tf.layers.flatten(conv_final)
        flat_w=tf.layers.flatten(W)
        flat=tf.concat([flat,flat_w],1)
        print(flat)
        dense=tf.layers.dense(flat,2048,kernel_regularizer=reg,bias_regularizer=reg)
        dense_2=tf.layers.dense(dense,1024,kernel_regularizer=reg,bias_regularizer=reg)
        dense_3=tf.layers.dense(dense_2,512,kernel_regularizer=reg,bias_regularizer=reg)
        logits_pred=tf.layers.dense(dense_3,5,name="Predictions")
        entropy=tf.nn.sparse_softmax_cross_entropy_with_logits(logits=logits_pred,labels=y)
        print(entropy)
        reg_losses = tf.get_collection(tf.GraphKeys.REGULARIZATION_LOSSES)
        reg_constant = 0.0001
        loss=tf.reduce_mean(entropy)+reg_constant * sum(reg_losses)
        optimizer=tf.train.RMSPropOptimizer(lr)
        t_op=optimizer.minimize(loss)
        acc=tf.math.in_top_k(tf.cast(logits_pred,tf.float32),y,1)
        accuracy=tf.reduce_mean(tf.cast(acc,tf.float32))
        init=tf.global_variables_initializer()
        saver=tf.train.Saver()

    for node in (synm,bls,blhs,gl,dps,dphs,dis,lr,y):
        g.add_to_collection("input_nodes",node)

    for node in (loss,t_op,accuracy,init):
        g.add_to_collection("output_nodes",node)

    return g,saver