diff --git a/code/learned_reconstruction.ipynb b/code/learned_reconstruction.ipynb
index a2437bc..8469147 100644
--- a/code/learned_reconstruction.ipynb
+++ b/code/learned_reconstruction.ipynb
@@ -36,20 +36,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "metadata": {
     "scrolled": true
    },
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/home/olivierv/anaconda3/lib/python3.6/importlib/_bootstrap.py:205: RuntimeWarning: compiletime version 3.5 of module 'tensorflow.python.framework.fast_tensor_util' does not match runtime version 3.6\n",
-      "  return f(*args, **kwds)\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "import tensorflow as tf\n",
     "import numpy as np\n",
@@ -62,7 +53,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 19,
+   "execution_count": null,
    "metadata": {
     "collapsed": true,
     "scrolled": true
@@ -87,7 +78,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "metadata": {
     "collapsed": true,
     "scrolled": true
@@ -105,7 +96,7 @@
     "# Create tensorflow wrappers\n",
     "tf_op = odl.contrib.tensorflow.as_tensorflow_layer(operator)\n",
     "tf_op_adj = odl.contrib.tensorflow.as_tensorflow_layer(operator.adjoint)\n",
-    "tf_fbp_op = odl.contrib.tensorflow.as_tensorflow_layer(fbp_op)"
+    "tf_fbp_op = odl.contrib.tensorflow.as_tensorflow_layer(fbp_op, name=\"FBP\")"
    ]
   },
   {
@@ -121,22 +112,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "metadata": {
     "scrolled": true
    },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Extracting MNIST_data/train-images-idx3-ubyte.gz\n",
-      "Extracting MNIST_data/train-labels-idx1-ubyte.gz\n",
-      "Extracting MNIST_data/t10k-images-idx3-ubyte.gz\n",
-      "Extracting MNIST_data/t10k-labels-idx1-ubyte.gz\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# Get MNIST data\n",
     "mnist = input_data.read_data_sets('MNIST_data')"
@@ -144,7 +124,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": null,
    "metadata": {
     "collapsed": true,
     "scrolled": true
@@ -171,7 +151,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": null,
    "metadata": {
     "collapsed": true,
     "scrolled": true
@@ -211,7 +191,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 20,
+   "execution_count": null,
    "metadata": {
     "collapsed": true
    },
@@ -229,20 +209,36 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": null,
+   "metadata": {
+    "collapsed": true
+   },
+   "outputs": [],
+   "source": [
+    "def get_tensor_name(tensor, default_name=\"Truth\"):\n",
+    "    try:\n",
+    "        name = tensor.name.split('/')[0]\n",
+    "    except AttributeError:\n",
+    "        name = default_name\n",
+    "    return name"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
    "metadata": {
     "collapsed": true,
     "scrolled": true
    },
    "outputs": [],
    "source": [
-    "def visualize(result_tensor, data_placeholder, indices=None):\n",
+    "def visualize(result_tensors, data_placeholder, indices=None):\n",
     "    \"\"\"Visualize the result of a reconstruction.\n",
     "    \n",
     "    Parameters\n",
     "    ----------\n",
-    "    result_tensor : `tf.Tensor`, shape (None, 28, 28, 1)\n",
-    "        The tensorflow tensor containing the result of the reonstruction\n",
+    "    result_tensors : list of `tf.Tensor`, shape (None, 28, 28, 1)\n",
+    "        The tensorflow tensor containing the result of the reconstruction\n",
     "        operator.\n",
     "    data_placeholder : `tf.Tensor`, shape (None, 5, 41, 1)\n",
     "        The tensorflow tensor containing the input to the reconstruction\n",
@@ -251,28 +247,29 @@
     "    if indices is None:\n",
     "        indices = [0]\n",
     "        \n",
-    "    result = result_tensor.eval(\n",
-    "        feed_dict={data_placeholder: test_data[indices]})\n",
+    "    results = [result_tensor.eval(\n",
+    "        feed_dict={data_placeholder: test_data[indices]}) for result_tensor in result_tensors]\n",
     "    \n",
-    "    size = 2\n",
-    "    fig, rows = plt.subplots(len(indices), 2, sharex=True, sharey=True, figsize=(2*size, size*len(indices)))\n",
+    "    results_ = [test_images[indices]] + results\n",
+    "    names = [get_tensor_name(tensor) for tensor in [None]+result_tensors]\n",
+    "    \n",
+    "    figsize = 2\n",
+    "    fig, rows = plt.subplots(len(indices), len(results_), sharex=True, sharey=True, figsize=(len(results_)*figsize, figsize*len(indices)))\n",
     "    # stupid matplotlib:\n",
     "    if len(indices) == 1:\n",
     "        rows = [rows]\n",
-    "    for i, (img, res, ax) in enumerate(zip(test_images[indices], result, rows)):\n",
-    "        if i == 0:\n",
-    "            ax[0].set_title('truth')\n",
-    "            ax[1].set_title('reconstruction')\n",
-    "        ax[0].imshow(img.squeeze(), clim=[0, 1], cmap='bone')\n",
-    "        ax[0].set_axis_off()\n",
-    "        ax[1].imshow(res.squeeze(), clim=[0, 1], cmap='bone')\n",
-    "        ax[1].set_axis_off()\n",
+    "    for i, row in enumerate(rows):\n",
+    "        for name, res, ax in zip(names, results_, row):\n",
+    "            if i == 0:\n",
+    "                ax.set_title(name)\n",
+    "            ax.imshow(res[i].squeeze(), clim=[0,1], cmap=\"bone\")\n",
+    "            ax.set_axis_off()\n",
     "    plt.show()"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": null,
    "metadata": {
     "collapsed": true
    },
@@ -281,51 +278,45 @@
     "rand_indices = np.random.randint(low=0, high=len(test_data), size=8)"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Create placeholders. Placeholders are needed in tensorflow since tensorflow is a lazy language,\n",
+    "and hence we first define the computational graph with placeholders as input, and later we evaluate it.\n"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": null,
    "metadata": {
     "collapsed": true,
     "scrolled": true
    },
    "outputs": [],
    "source": [
-    "# Create placeholders. Placeholders are needed in tensorflow since tensorflow is a lazy language,\n",
-    "# and hence we first define the computational graph with placeholders as input, and later we evaluate it.\n",
     "with tf.name_scope('placeholders'):\n",
     "    x_true = tf.placeholder(tf.float32, shape=[None, 28, 28, 1], name=\"x_true\")\n",
     "    y = tf.placeholder(tf.float32, shape=[None, *operator.range.shape, 1], name=\"y\")"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Use the evaluator on the FBP reconstruction as a baseline"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": null,
    "metadata": {
     "scrolled": false
    },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "FBP Mean squared error: 0.15122871100902557\n"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQQAAAOVCAYAAACPrZ7FAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
-      "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f83ac1766a0>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
-    "# Use the evaluator on the FBP reconstruction as a baseline\n",
     "print('FBP Mean squared error: {}'.format(evaluate(tf_fbp_op(y), y)))\n",
-    "visualize(tf_fbp_op(y), y, indices=rand_indices)"
+    "visualize([tf_fbp_op(y)], y, indices=rand_indices)"
    ]
   },
   {
@@ -339,39 +330,37 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
-   "metadata": {
-    "scrolled": false
-   },
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "'9900/10000 Error: 0.02479'"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": [
-    "with tf.name_scope('fully_learned_reconstruction'):\n",
-    "    x = tf.contrib.layers.flatten(y)\n",
-    "\n",
-    "    x = tf.contrib.layers.fully_connected(x, num_outputs=1024)\n",
-    "    x = tf.contrib.layers.fully_connected(x, num_outputs=1024)\n",
-    "    x = tf.contrib.layers.fully_connected(x, num_outputs=28 * 28,\n",
-    "                                          activation_fn=None)\n",
+    "with tf.variable_scope('flr', reuse=tf.AUTO_REUSE):\n",
+    "    with tf.name_scope('fully_learned_reconstruction'):\n",
+    "        x = tf.contrib.layers.flatten(y)\n",
     "\n",
-    "    x_result_fully = tf.reshape(x, [-1, 28, 28, 1])\n",
+    "        x = tf.contrib.layers.fully_connected(x, num_outputs=1024)\n",
+    "        x = tf.contrib.layers.fully_connected(x, num_outputs=1024)\n",
+    "        x = tf.contrib.layers.fully_connected(x, num_outputs=28 * 28,\n",
+    "                                              activation_fn=None)\n",
     "\n",
-    "with tf.name_scope('optimizer'):\n",
-    "    loss = tf.reduce_mean((x_result_fully - x_true) ** 2)\n",
-    "    optimizer = tf.train.AdamOptimizer().minimize(loss)\n",
-    "\n",
-    "# Initialize all TF variables\n",
-    "session.run(tf.global_variables_initializer())\n",
+    "        x_result_fully = tf.reshape(x, [-1, 28, 28, 1])\n",
     "\n",
+    "    with tf.name_scope('optimizer_fully'):\n",
+    "        loss = tf.reduce_mean((x_result_fully - x_true) ** 2)\n",
+    "        optimizer = tf.train.AdamOptimizer().minimize(loss)\n",
+    "        \n",
+    "# Initialize current variables\n",
+    "session.run([v.initializer for v in tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES, scope='flr')]);"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [],
+   "source": [
     "max_iter = 10000\n",
     "for i in range(max_iter):\n",
     "    batch = mnist.train.next_batch(5)\n",
@@ -385,24 +374,13 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": null,
    "metadata": {
-    "scrolled": false
+    "scrolled": true
    },
-   "outputs": [
-    {
-     "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQQAAAOVCAYAAACPrZ7FAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
-      "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f83980a2da0>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
-    "visualize(x_result_fully, y, rand_indices)"
+    "visualize([x_result_fully], y, rand_indices)"
    ]
   },
   {
@@ -416,39 +394,37 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
-   "metadata": {
-    "scrolled": false
-   },
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "'9500/10000 Error: 0.01956'"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": [
-    "with tf.name_scope('learned_denoiser'):\n",
-    "    x = tf_fbp_op(y)\n",
-    "\n",
-    "    x = tf.contrib.layers.conv2d(x, num_outputs=32, kernel_size=3)\n",
-    "    x = tf.contrib.layers.conv2d(x, num_outputs=32, kernel_size=3)\n",
-    "    x = tf.contrib.layers.conv2d(x, num_outputs=1, kernel_size=1,\n",
-    "                                 activation_fn=None)\n",
+    "with tf.variable_scope('denoise', reuse=tf.AUTO_REUSE):\n",
+    "    with tf.name_scope('learned_denoiser'):\n",
+    "        x = tf_fbp_op(y)\n",
     "\n",
-    "    x_result_denoise = x\n",
+    "        x = tf.contrib.layers.conv2d(x, num_outputs=32, kernel_size=3)\n",
+    "        x = tf.contrib.layers.conv2d(x, num_outputs=32, kernel_size=3)\n",
+    "        x = tf.contrib.layers.conv2d(x, num_outputs=1, kernel_size=1,\n",
+    "                                     activation_fn=None)\n",
     "\n",
-    "with tf.name_scope('optimizer'):\n",
-    "    loss = tf.reduce_mean((x_result_denoise - x_true) ** 2)\n",
-    "    optimizer = tf.train.AdamOptimizer().minimize(loss)\n",
+    "        x_result_denoise = x\n",
     "\n",
-    "# Initialize all TF variables\n",
-    "session.run(tf.global_variables_initializer())\n",
+    "    with tf.name_scope('optimizer_denoiser'):\n",
+    "        loss = tf.reduce_mean((x_result_denoise - x_true) ** 2)\n",
+    "        optimizer = tf.train.AdamOptimizer().minimize(loss)\n",
     "\n",
+    "# Initialize all denoising variables\n",
+    "session.run([v.initializer for v in tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES, scope='denoise')]);"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "scrolled": false
+   },
+   "outputs": [],
+   "source": [
     "max_iter = 10000\n",
     "for i in range(10000):\n",
     "    batch = mnist.train.next_batch(5)\n",
@@ -462,24 +438,13 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 24,
+   "execution_count": null,
    "metadata": {
     "scrolled": false
    },
-   "outputs": [
-    {
-     "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQQAAAOVCAYAAACPrZ7FAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
-      "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f8398240b70>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
-    "visualize(x_result_denoise, y, rand_indices)"
+    "visualize([tf_fbp_op(y), x_result_denoise], y, rand_indices)"
    ]
   },
   {
@@ -493,48 +458,48 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": null,
    "metadata": {
-    "scrolled": true
+    "collapsed": true
    },
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "'9800/10000 Error: 0.01562'"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
-    "with tf.name_scope('learned_primal_dual'):\n",
-    "    primal = tf_fbp_op(y)\n",
-    "    dual = tf.identity(y)\n",
-    "    \n",
-    "    for i in range(1):\n",
-    "        dy = tf.concat([dual, tf_op(primal)], axis=-1)\n",
-    "        dy = tf.contrib.layers.conv2d(dy, num_outputs=32, kernel_size=3)\n",
-    "        dy = tf.contrib.layers.conv2d(dy, num_outputs=1, kernel_size=3,\n",
-    "                                      activation_fn=None)\n",
-    "        dual = dual + dy\n",
-    "\n",
-    "        dx = tf.concat([primal, tf_op_adj(dual)], axis=-1)\n",
-    "        dx = tf.contrib.layers.conv2d(dx, num_outputs=32, kernel_size=3)\n",
-    "        dx = tf.contrib.layers.conv2d(dx, num_outputs=1, kernel_size=3,\n",
-    "                                      activation_fn=None)\n",
-    "        primal = primal + dx\n",
-    "    \n",
-    "    x_result_lpd = primal\n",
+    "with tf.variable_scope('primal', reuse=tf.AUTO_REUSE):\n",
+    "    with tf.name_scope('learned_primal_dual'):\n",
+    "        primal = tf_fbp_op(y)\n",
+    "        dual = tf.identity(y)\n",
+    "\n",
+    "        for i in range(1):\n",
+    "            dy = tf.concat([dual, tf_op(primal)], axis=-1)\n",
+    "            dy = tf.contrib.layers.conv2d(dy, num_outputs=32, kernel_size=3)\n",
+    "            dy = tf.contrib.layers.conv2d(dy, num_outputs=1, kernel_size=3,\n",
+    "                                          activation_fn=None)\n",
+    "            dual = dual + dy\n",
+    "\n",
+    "            dx = tf.concat([primal, tf_op_adj(dual)], axis=-1)\n",
+    "            dx = tf.contrib.layers.conv2d(dx, num_outputs=32, kernel_size=3)\n",
+    "            dx = tf.contrib.layers.conv2d(dx, num_outputs=1, kernel_size=3,\n",
+    "                                          activation_fn=None)\n",
+    "            primal = primal + dx\n",
     "\n",
-    "with tf.name_scope('optimizer'):\n",
-    "    loss = tf.reduce_mean((x_result_lpd - x_true) ** 2)\n",
-    "    optimizer = tf.train.AdamOptimizer().minimize(loss)\n",
+    "        x_result_lpd = primal\n",
     "\n",
-    "# Initialize all TF variables\n",
-    "session.run(tf.global_variables_initializer())\n",
+    "    with tf.name_scope('optimizer'):\n",
+    "        loss = tf.reduce_mean((x_result_lpd - x_true) ** 2)\n",
+    "        optimizer = tf.train.AdamOptimizer().minimize(loss)\n",
     "\n",
+    "# Initialize all current variables\n",
+    "session.run([v.initializer for v in tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES, scope='primal')]);"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [],
+   "source": [
     "max_iter = 10000\n",
     "for i in range(max_iter):\n",
     "    batch = mnist.train.next_batch(5)\n",
@@ -548,24 +513,13 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 26,
+   "execution_count": null,
    "metadata": {
     "scrolled": false
    },
-   "outputs": [
-    {
-     "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAQQAAAOVCAYAAACPrZ7FAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAEDAIACAgEEAAAGDAADC/wChiFBCfsFv+AAAAABJRU5ErkJggg==\n",
-      "text/plain": [
-       "<matplotlib.figure.Figure at 0x7f8398212278>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
-    "visualize(x_result_lpd, y, rand_indices)"
+    "visualize([x_result_denoise, x_result_lpd], y, rand_indices)"
    ]
   }
  ],