c8f075268f7e5645a77eef21591a62a07e7e8baa,hypergan/trainers/sgd_trainer.py,,create,#Any#Any#Any#Any#,27

Before Change

    d_lr = np.float32(config.discriminator_learn_rate)

    gan.graph.d_vars = d_vars
    g_optimizer = tf.train.GradientDescentOptimizer(g_lr).minimize(g_loss, var_list=g_vars)
    d_optimizer = tf.train.GradientDescentOptimizer(d_lr).minimize(d_loss, var_list=d_vars)

    return g_optimizer, d_optimizer

After Change

    d_optimizer = tf.train.GradientDescentOptimizer(d_lr)

    if(config.clipped_gradients):
        g_optimizer = capped_optimizer(g_optimizer, config.clipped_gradients, g_loss, g_vars)
        d_optimizer = capped_optimizer(d_optimizer, config.clipped_gradients, d_loss, d_vars)
    else:
        g_optimizer = g_optimizer.minimize(g_loss, var_list=g_vars)
        d_optimizer = d_optimizer.minimize(d_loss, var_list=d_vars)

    gan.graph.clip = [tf.assign(d,tf.clip_by_value(d, -config.d_clipped_weights, config.d_clipped_weights))  for d in d_vars]

    return g_optimizer, d_optimizer

Instances