diff --git a/WIP/6-gated_pixelcnn_cropped/3d_gated_pixelcnn_conv.ipynb b/WIP/6-gated_pixelcnn_cropped/3d_gated_pixelcnn_conv.ipynb
index 02e1f98..8df9dee 100644
--- a/WIP/6-gated_pixelcnn_cropped/3d_gated_pixelcnn_conv.ipynb
+++ b/WIP/6-gated_pixelcnn_cropped/3d_gated_pixelcnn_conv.ipynb
@@ -35,7 +35,8 @@
       "source": [
         "import tensorflow as tf\n",
         "import tensorflow.keras as keras\n",
-        "import numpy as np"
+        "import numpy as np\n",
+        "import math"
       ],
       "execution_count": 0,
       "outputs": []
@@ -62,10 +63,10 @@
         "colab": {}
       },
       "source": [
-        "def print_test_3d(test_matrix):\n",
-        "    for i in range(test_matrix.shape[1]):\n",
+        "def print_3d(matrix_3d):\n",
+        "    for i in range(matrix_3d.shape[0]):\n",
         "        print(f'Depth {i}')\n",
-        "        print(test_matrix[0,i,:,:,0])"
+        "        print(matrix_3d[i,...])"
       ],
       "execution_count": 0,
       "outputs": []
@@ -79,12 +80,12 @@
           "base_uri": "https://localhost:8080/",
           "height": 538
         },
-        "outputId": "e7fd642b-23ac-4d79-a1f0-564853e64a79"
+        "outputId": "456c0c0f-9701-40c2-d545-c8091d50a59a"
       },
       "source": [
-        "print_test_3d(test_ones_3d)"
+        "print_3d(test_ones_3d.squeeze())"
       ],
-      "execution_count": 4,
+      "execution_count": 146,
       "outputs": [
         {
           "output_type": "stream",
@@ -229,6 +230,7 @@
         "        if self.mask_type == 'V':\n",
         "            mask[center_h + 1:, :, :, :] = 0.\n",
         "        else:\n",
+        "            mask[:center_h, :, :] = 0.\n",
         "            mask[center_h, center_w + (self.mask_type == 'B'):, :, :] = 0.\n",
         "            mask[center_h + 1:, :, :] = 0.\n",
         "\n",
@@ -243,31 +245,6 @@
       "execution_count": 0,
       "outputs": []
     },
-    {
-      "cell_type": "code",
-      "metadata": {
-        "id": "PmmR19cRgNyV",
-        "colab_type": "code",
-        "colab": {}
-      },
-      "source": [
-        "def build_test_stack_2d(mask_type, input_shape=(5, 5, 1), kernel_size=(3, 3)):\n",
-        "  inputs = tf.keras.layers.Input(shape=input_shape)\n",
-        "  \n",
-        "  x = MaskedConv2D(mask_type=mask_type,\n",
-        "                   filters=1,\n",
-        "                   kernel_size=kernel_size, \n",
-        "                   padding='same',\n",
-        "                   kernel_initializer='ones', \n",
-        "                   bias_initializer='zeros')(inputs)\n",
-        "\n",
-        "  stack = tf.keras.Model(inputs=inputs, outputs=x)\n",
-        "  stack.compile(optimizer='adam', loss='mse')\n",
-        "  return stack"
-      ],
-      "execution_count": 0,
-      "outputs": []
-    },
     {
       "cell_type": "markdown",
       "metadata": {
@@ -295,9 +272,9 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 104
+          "height": 208
         },
-        "outputId": "78b74814-4929-4e48-e45c-a167b80ae83d"
+        "outputId": "3f769c93-2f50-452d-ea81-503488f3c7c1"
       },
       "source": [
         "mask_type = 'V'\n",
@@ -319,13 +296,23 @@
         "x = conv(x)\n",
         "result = cropping(x)\n",
         "\n",
-        "print(result.numpy().squeeze())\n"
+        "print('MASK')\n",
+        "print(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print(result.numpy().squeeze())"
       ],
-      "execution_count": 93,
+      "execution_count": 111,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "[[1. 1. 1.]\n",
+            " [1. 1. 1.]\n",
+            " [0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [2. 3. 3. 3. 2.]\n",
             " [4. 6. 6. 6. 4.]\n",
@@ -336,6 +323,16 @@
         }
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "iHY6UE2_p5oc",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Horizontal stack A"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -343,20 +340,40 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 104
+          "height": 208
         },
-        "outputId": "6bc566e2-2bbd-49f0-a4c6-0574125c6cd8"
+        "outputId": "f9a62e80-29f4-4c9d-b511-620a3eb76037"
       },
       "source": [
-        "horizontal_A_stack = build_test_stack_2d('A', kernel_size=(1, 3))\n",
-        "val = horizontal_A_stack.predict(test_ones_2d)\n",
-        "print(val[0,:,:,0].squeeze())"
+        "mask_type = 'A'\n",
+        "kernel_size=(3, 3)\n",
+        "\n",
+        "conv = MaskedConv2D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "result = conv(test_ones_2d)\n",
+        "\n",
+        "print('MASK')\n",
+        "print(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print(result.numpy().squeeze())"
       ],
-      "execution_count": 9,
+      "execution_count": 113,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "[[0. 0. 0.]\n",
+            " [1. 0. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "[[0. 1. 1. 1. 1.]\n",
             " [0. 1. 1. 1. 1.]\n",
             " [0. 1. 1. 1. 1.]\n",
@@ -367,6 +384,16 @@
         }
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "jMuS-vgWqAWK",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Horizontal stack B"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -374,20 +401,40 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 104
+          "height": 208
         },
-        "outputId": "8e8b83ab-2c0b-4ae9-e5c0-eceb6ab55f66"
+        "outputId": "64e2dfcf-ba0f-4b8b-b92d-8b5504340a4a"
       },
       "source": [
-        "horizontal_B_stack = build_test_stack_2d('B', kernel_size=(1, 3))\n",
-        "val = horizontal_B_stack.predict(test_ones_2d)\n",
-        "print(val[0,:,:,0].squeeze())"
+        "mask_type = 'B'\n",
+        "kernel_size=(3, 3)\n",
+        "\n",
+        "conv = MaskedConv2D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "result = conv(test_ones_2d)\n",
+        "\n",
+        "print('MASK')\n",
+        "print(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print(result.numpy().squeeze())"
       ],
-      "execution_count": 10,
+      "execution_count": 115,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "[[0. 0. 0.]\n",
+            " [1. 1. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "[[1. 2. 2. 2. 2.]\n",
             " [1. 2. 2. 2. 2.]\n",
             " [1. 2. 2. 2. 2.]\n",
@@ -408,6 +455,16 @@
         "### Tests with kernel_size 4"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "WNykK-WpqMlu",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Vertical stack"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -415,30 +472,68 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 104
+          "height": 225
         },
-        "outputId": "fe952753-ad58-4d40-9b3e-70abd2a96a4c"
+        "outputId": "f2ec68a3-95e6-4113-f0ba-9c37c7c63589"
       },
       "source": [
-        "vertical_stack = build_test_stack_2d('V', kernel_size=(4, 4))\n",
-        "val = vertical_stack.predict(test_ones_2d)\n",
-        "print(val[0,:,:,0].squeeze())"
+        "mask_type = 'V'\n",
+        "kernel_size=(4, 4)\n",
+        "\n",
+        "padding = keras.layers.ZeroPadding2D(padding=((1,0),0))\n",
+        "\n",
+        "conv = MaskedConv2D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "cropping = keras.layers.Cropping2D(cropping=((0, 1), 0))\n",
+        "\n",
+        "\n",
+        "x = padding(test_ones_2d)\n",
+        "x = conv(x)\n",
+        "result = cropping(x)\n",
+        "\n",
+        "print('MASK')\n",
+        "print(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print(result.numpy().squeeze())"
       ],
-      "execution_count": 11,
+      "execution_count": 119,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "[[1. 1. 1. 1.]\n",
+            " [1. 1. 1. 1.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]]\n"
+            " [6. 8. 8. 6. 4.]\n",
+            " [6. 8. 8. 6. 4.]\n",
+            " [6. 8. 8. 6. 4.]]\n"
           ],
           "name": "stdout"
         }
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "E5jUGK3_qbT8",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Horizontal stack A"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -446,20 +541,41 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 104
+          "height": 225
         },
-        "outputId": "76861f8f-c601-4e97-e6f8-1b3818104cb7"
+        "outputId": "bd56a4a9-985e-4746-c110-02f6a37bf964"
       },
       "source": [
-        "horizontal_A_stack = build_test_stack_2d('A', kernel_size=(1, 4))\n",
-        "val = horizontal_A_stack.predict(test_ones_2d)\n",
-        "print(val[0,:,:,0].squeeze())"
+        "mask_type = 'A'\n",
+        "kernel_size=(4, 4)\n",
+        "\n",
+        "conv = MaskedConv2D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "result = conv(test_ones_2d)\n",
+        "\n",
+        "print('MASK')\n",
+        "print(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print(result.numpy().squeeze())"
       ],
-      "execution_count": 12,
+      "execution_count": 120,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "[[0. 0. 0. 0.]\n",
+            " [1. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "[[0. 1. 1. 1. 1.]\n",
             " [0. 1. 1. 1. 1.]\n",
             " [0. 1. 1. 1. 1.]\n",
@@ -470,6 +586,16 @@
         }
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "KqORK7mLqvPP",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Horizontal stack B"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -477,20 +603,41 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 104
+          "height": 225
         },
-        "outputId": "1c35dccd-8136-4ccb-bb33-26fa728acbd9"
+        "outputId": "5a6e6421-a0f9-4665-f453-b9fb0b80c325"
       },
       "source": [
-        "horizontal_B_stack = build_test_stack_2d('B', kernel_size=(1, 4))\n",
-        "val = horizontal_B_stack.predict(test_ones_2d)\n",
-        "print(val[0,:,:,0].squeeze())"
+        "mask_type = 'B'\n",
+        "kernel_size=(4, 4)\n",
+        "\n",
+        "conv = MaskedConv2D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "result = conv(test_ones_2d)\n",
+        "\n",
+        "print('MASK')\n",
+        "print(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print(result.numpy().squeeze())"
       ],
-      "execution_count": 13,
+      "execution_count": 121,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "[[0. 0. 0. 0.]\n",
+            " [1. 1. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "[[1. 2. 2. 2. 2.]\n",
             " [1. 2. 2. 2. 2.]\n",
             " [1. 2. 2. 2. 2.]\n",
@@ -581,14 +728,17 @@
         "\n",
         "\n",
         "        if self.mask_type == 'D':\n",
-        "            mask[center_d:, :, :, :, :] = 0.\n",
+        "            mask[center_d+1:, :, :, :, :] = 0.\n",
         "        elif self.mask_type == 'V':\n",
-        "            mask[center_d, center_h:, :, :, :] = 0.\n",
+        "            mask[:center_d, :, :, :, :] = 0.\n",
+        "            mask[center_d, center_h+1:, :, :, :] = 0.\n",
         "            mask[center_d + 1:, :, :, :, :] = 0.\n",
         "        else:\n",
+        "            mask[:center_d, :, :, :, :] = 0.\n",
+        "            mask[:, :center_h, :, :, :] = 0.\n",
         "            mask[center_d, center_h, center_w + (self.mask_type == 'B'):, :, :] = 0.\n",
-        "            mask[center_d + 1:, :, :, :, :] = 0.\n",
         "            mask[:, center_h + 1:, :, :, :] = 0.\n",
+        "            mask[center_d + 1:, :, :, :, :] = 0.\n",
         "\n",
         "        self.mask = tf.constant(mask, dtype=tf.float32, name='mask')\n",
         "\n",
@@ -602,38 +752,23 @@
       "outputs": []
     },
     {
-      "cell_type": "code",
+      "cell_type": "markdown",
       "metadata": {
-        "id": "A-D6ge8jr04d",
-        "colab_type": "code",
-        "colab": {}
+        "id": "TJSNB3m4Apfi",
+        "colab_type": "text"
       },
       "source": [
-        "def build_test_stack_3d(mask_type, input_shape=(5, 5, 5, 1), kernel_size=(3, 3, 3)):\n",
-        "  inputs = tf.keras.layers.Input(shape=input_shape)\n",
-        "  \n",
-        "  x = MaskedConv3D(mask_type=mask_type,\n",
-        "                   filters=1,\n",
-        "                   kernel_size=kernel_size, \n",
-        "                   padding='same',\n",
-        "                   kernel_initializer='ones', \n",
-        "                   bias_initializer='zeros')(inputs)\n",
-        "\n",
-        "  stack = tf.keras.Model(inputs=inputs, outputs=x)\n",
-        "  stack.compile(optimizer='adam', loss='mse')\n",
-        "  return stack"
-      ],
-      "execution_count": 0,
-      "outputs": []
+        "### Tests with kernel_size 3"
+      ]
     },
     {
       "cell_type": "markdown",
       "metadata": {
-        "id": "TJSNB3m4Apfi",
+        "id": "-NIs7IZdrVnu",
         "colab_type": "text"
       },
       "source": [
-        "### Tests with kernel_size 3"
+        "#### Depth stack"
       ]
     },
     {
@@ -643,20 +778,56 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 538
+          "height": 799
         },
-        "outputId": "15429ddd-ce04-4e04-b79b-6fe7cfedac96"
+        "outputId": "f98454bb-09ed-416f-8d20-af28b813881a"
       },
       "source": [
-        "depth_stack = build_test_stack_3d('D')\n",
-        "val = depth_stack.predict(test_ones_3d)\n",
-        "print_test_3d(val)"
+        "mask_type = 'D'\n",
+        "kernel_size=(3, 3, 3)\n",
+        "\n",
+        "padding = keras.layers.ZeroPadding3D(padding=((1,0),0,0))\n",
+        "\n",
+        "conv = MaskedConv3D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "cropping = keras.layers.Cropping3D(cropping=((0, 1), 0, 0))\n",
+        "\n",
+        "\n",
+        "x = padding(test_ones_3d)\n",
+        "x = conv(x)\n",
+        "result = cropping(x)\n",
+        "\n",
+        "print('MASK')\n",
+        "print_3d(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print_3d(result.numpy().squeeze())\n"
       ],
-      "execution_count": 16,
+      "execution_count": 154,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "Depth 0\n",
+            "[[1. 1. 1.]\n",
+            " [1. 1. 1.]\n",
+            " [1. 1. 1.]]\n",
+            "Depth 1\n",
+            "[[1. 1. 1.]\n",
+            " [1. 1. 1.]\n",
+            " [1. 1. 1.]]\n",
+            "Depth 2\n",
+            "[[0. 0. 0.]\n",
+            " [0. 0. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "Depth 0\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [0. 0. 0. 0. 0.]\n",
@@ -670,28 +841,38 @@
             " [6. 9. 9. 9. 6.]\n",
             " [4. 6. 6. 6. 4.]]\n",
             "Depth 2\n",
-            "[[4. 6. 6. 6. 4.]\n",
-            " [6. 9. 9. 9. 6.]\n",
-            " [6. 9. 9. 9. 6.]\n",
-            " [6. 9. 9. 9. 6.]\n",
-            " [4. 6. 6. 6. 4.]]\n",
+            "[[ 8. 12. 12. 12.  8.]\n",
+            " [12. 18. 18. 18. 12.]\n",
+            " [12. 18. 18. 18. 12.]\n",
+            " [12. 18. 18. 18. 12.]\n",
+            " [ 8. 12. 12. 12.  8.]]\n",
             "Depth 3\n",
-            "[[4. 6. 6. 6. 4.]\n",
-            " [6. 9. 9. 9. 6.]\n",
-            " [6. 9. 9. 9. 6.]\n",
-            " [6. 9. 9. 9. 6.]\n",
-            " [4. 6. 6. 6. 4.]]\n",
+            "[[ 8. 12. 12. 12.  8.]\n",
+            " [12. 18. 18. 18. 12.]\n",
+            " [12. 18. 18. 18. 12.]\n",
+            " [12. 18. 18. 18. 12.]\n",
+            " [ 8. 12. 12. 12.  8.]]\n",
             "Depth 4\n",
-            "[[4. 6. 6. 6. 4.]\n",
-            " [6. 9. 9. 9. 6.]\n",
-            " [6. 9. 9. 9. 6.]\n",
-            " [6. 9. 9. 9. 6.]\n",
-            " [4. 6. 6. 6. 4.]]\n"
+            "[[ 8. 12. 12. 12.  8.]\n",
+            " [12. 18. 18. 18. 12.]\n",
+            " [12. 18. 18. 18. 12.]\n",
+            " [12. 18. 18. 18. 12.]\n",
+            " [ 8. 12. 12. 12.  8.]]\n"
           ],
           "name": "stdout"
         }
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "u4K8-zY0timI",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Vertical stack"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -699,55 +880,101 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 538
+          "height": 799
         },
-        "outputId": "4d50016e-5f3f-4124-d02e-beda5b8fc4e2"
+        "outputId": "b836ef8d-4cbf-40f2-f086-4a83dd7a0ddf"
       },
       "source": [
-        "vertical_stack = build_test_stack_3d('V', kernel_size=(1, 3, 3))\n",
-        "val = vertical_stack.predict(test_ones_3d)\n",
-        "print_test_3d(val)"
+        "mask_type = 'V'\n",
+        "kernel_size=(3, 3, 3)\n",
+        "\n",
+        "padding = keras.layers.ZeroPadding3D(padding=(0,(1,0),0))\n",
+        "\n",
+        "conv = MaskedConv3D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "cropping = keras.layers.Cropping3D(cropping=(0,(0, 1), 0))\n",
+        "\n",
+        "\n",
+        "x = padding(test_ones_3d)\n",
+        "x = conv(x)\n",
+        "result = cropping(x)\n",
+        "\n",
+        "print('MASK')\n",
+        "print_3d(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print_3d(result.numpy().squeeze())\n"
       ],
-      "execution_count": 17,
+      "execution_count": 157,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "Depth 0\n",
+            "[[0. 0. 0.]\n",
+            " [0. 0. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "Depth 1\n",
+            "[[1. 1. 1.]\n",
+            " [1. 1. 1.]\n",
+            " [0. 0. 0.]]\n",
+            "Depth 2\n",
+            "[[0. 0. 0.]\n",
+            " [0. 0. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "Depth 0\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]]\n",
+            " [4. 6. 6. 6. 4.]\n",
+            " [4. 6. 6. 6. 4.]\n",
+            " [4. 6. 6. 6. 4.]]\n",
             "Depth 1\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]]\n",
+            " [4. 6. 6. 6. 4.]\n",
+            " [4. 6. 6. 6. 4.]\n",
+            " [4. 6. 6. 6. 4.]]\n",
             "Depth 2\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]]\n",
+            " [4. 6. 6. 6. 4.]\n",
+            " [4. 6. 6. 6. 4.]\n",
+            " [4. 6. 6. 6. 4.]]\n",
             "Depth 3\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]]\n",
+            " [4. 6. 6. 6. 4.]\n",
+            " [4. 6. 6. 6. 4.]\n",
+            " [4. 6. 6. 6. 4.]]\n",
             "Depth 4\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]\n",
-            " [2. 3. 3. 3. 2.]]\n"
+            " [4. 6. 6. 6. 4.]\n",
+            " [4. 6. 6. 6. 4.]\n",
+            " [4. 6. 6. 6. 4.]]\n"
           ],
           "name": "stdout"
         }
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "oSlof4GvuaLu",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Horizontal stack A"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -755,20 +982,49 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 538
+          "height": 799
         },
-        "outputId": "a1b5bdfc-5415-459f-94d9-e9dffe5eaff3"
+        "outputId": "cfea8fc3-f6af-41d6-98aa-5453eb3612ca"
       },
       "source": [
-        "horizontal_A_stack = build_test_stack_3d('A', kernel_size=(1, 1, 3))\n",
-        "val = horizontal_A_stack.predict(test_ones_3d)\n",
-        "print_test_3d(val)"
+        "mask_type = 'A'\n",
+        "kernel_size=(3, 3, 3)\n",
+        "\n",
+        "conv = MaskedConv3D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "result = conv(test_ones_3d)\n",
+        "\n",
+        "print('MASK')\n",
+        "print_3d(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print_3d(result.numpy().squeeze())"
       ],
-      "execution_count": 18,
+      "execution_count": 168,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "Depth 0\n",
+            "[[0. 0. 0.]\n",
+            " [0. 0. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "Depth 1\n",
+            "[[0. 0. 0.]\n",
+            " [1. 0. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "Depth 2\n",
+            "[[0. 0. 0.]\n",
+            " [0. 0. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "Depth 0\n",
             "[[0. 1. 1. 1. 1.]\n",
             " [0. 1. 1. 1. 1.]\n",
@@ -804,6 +1060,16 @@
         }
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "ixWP3Ey7vs-H",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Horizontal stack B"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -811,20 +1077,49 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 538
+          "height": 799
         },
-        "outputId": "ce9a0d0d-bda7-408e-b42b-7d50858b26ed"
+        "outputId": "50244671-766c-41e2-8999-1e4f34844622"
       },
       "source": [
-        "horizontal_B_stack = build_test_stack_3d('B', kernel_size=(1, 1, 3))\n",
-        "val = horizontal_B_stack.predict(test_ones_3d)\n",
-        "print_test_3d(val)"
+        "mask_type = 'B'\n",
+        "kernel_size=(3, 3, 3)\n",
+        "\n",
+        "conv = MaskedConv3D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "result = conv(test_ones_3d)\n",
+        "\n",
+        "print('MASK')\n",
+        "print_3d(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print_3d(result.numpy().squeeze())"
       ],
-      "execution_count": 19,
+      "execution_count": 169,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "Depth 0\n",
+            "[[0. 0. 0.]\n",
+            " [0. 0. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "Depth 1\n",
+            "[[0. 0. 0.]\n",
+            " [1. 1. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "Depth 2\n",
+            "[[0. 0. 0.]\n",
+            " [0. 0. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "Depth 0\n",
             "[[1. 2. 2. 2. 2.]\n",
             " [1. 2. 2. 2. 2.]\n",
@@ -870,6 +1165,16 @@
         "### Tests with kernel_size 4"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "_ZFtCkYhvwGW",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Depth stack"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -877,20 +1182,64 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 538
+          "height": 937
         },
-        "outputId": "a55145ec-594e-4a81-cc3b-45df89d9e086"
+        "outputId": "e78c3adf-e32c-44fd-e0de-3a7db406bbd9"
       },
       "source": [
-        "depth_stack = build_test_stack_3d('D', kernel_size=(4, 4, 4))\n",
-        "val = depth_stack.predict(test_ones_3d)\n",
-        "print_test_3d(val)"
+        "mask_type = 'D'\n",
+        "kernel_size=(4, 4, 4)\n",
+        "\n",
+        "padding = keras.layers.ZeroPadding3D(padding=((1,0),0,0))\n",
+        "\n",
+        "conv = MaskedConv3D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "cropping = keras.layers.Cropping3D(cropping=((0, 1), 0, 0))\n",
+        "\n",
+        "\n",
+        "x = padding(test_ones_3d)\n",
+        "x = conv(x)\n",
+        "result = cropping(x)\n",
+        "\n",
+        "print('MASK')\n",
+        "print_3d(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print_3d(result.numpy().squeeze())\n"
       ],
-      "execution_count": 24,
+      "execution_count": 171,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "Depth 0\n",
+            "[[1. 1. 1. 1.]\n",
+            " [1. 1. 1. 1.]\n",
+            " [1. 1. 1. 1.]\n",
+            " [1. 1. 1. 1.]]\n",
+            "Depth 1\n",
+            "[[1. 1. 1. 1.]\n",
+            " [1. 1. 1. 1.]\n",
+            " [1. 1. 1. 1.]\n",
+            " [1. 1. 1. 1.]]\n",
+            "Depth 2\n",
+            "[[0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "Depth 3\n",
+            "[[0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "Depth 0\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [0. 0. 0. 0. 0.]\n",
@@ -904,28 +1253,38 @@
             " [ 9. 12. 12.  9.  6.]\n",
             " [ 6.  8.  8.  6.  4.]]\n",
             "Depth 2\n",
-            "[[ 9. 12. 12.  9.  6.]\n",
-            " [12. 16. 16. 12.  8.]\n",
-            " [12. 16. 16. 12.  8.]\n",
-            " [ 9. 12. 12.  9.  6.]\n",
-            " [ 6.  8.  8.  6.  4.]]\n",
+            "[[18. 24. 24. 18. 12.]\n",
+            " [24. 32. 32. 24. 16.]\n",
+            " [24. 32. 32. 24. 16.]\n",
+            " [18. 24. 24. 18. 12.]\n",
+            " [12. 16. 16. 12.  8.]]\n",
             "Depth 3\n",
-            "[[ 9. 12. 12.  9.  6.]\n",
-            " [12. 16. 16. 12.  8.]\n",
-            " [12. 16. 16. 12.  8.]\n",
-            " [ 9. 12. 12.  9.  6.]\n",
-            " [ 6.  8.  8.  6.  4.]]\n",
+            "[[18. 24. 24. 18. 12.]\n",
+            " [24. 32. 32. 24. 16.]\n",
+            " [24. 32. 32. 24. 16.]\n",
+            " [18. 24. 24. 18. 12.]\n",
+            " [12. 16. 16. 12.  8.]]\n",
             "Depth 4\n",
-            "[[ 9. 12. 12.  9.  6.]\n",
-            " [12. 16. 16. 12.  8.]\n",
-            " [12. 16. 16. 12.  8.]\n",
-            " [ 9. 12. 12.  9.  6.]\n",
-            " [ 6.  8.  8.  6.  4.]]\n"
+            "[[18. 24. 24. 18. 12.]\n",
+            " [24. 32. 32. 24. 16.]\n",
+            " [24. 32. 32. 24. 16.]\n",
+            " [18. 24. 24. 18. 12.]\n",
+            " [12. 16. 16. 12.  8.]]\n"
           ],
           "name": "stdout"
         }
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "3iXH2zkbv-qY",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Vertical stack"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -933,55 +1292,109 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 538
+          "height": 937
         },
-        "outputId": "6eef6945-5102-4847-ede8-a52b1ed54443"
+        "outputId": "3f0d6424-8783-42b5-f705-1e47592a40eb"
       },
       "source": [
-        "vertical_stack = build_test_stack_3d('V', kernel_size=(1, 4, 4))\n",
-        "val = vertical_stack.predict(test_ones_3d)\n",
-        "print_test_3d(val)"
+        "mask_type = 'V'\n",
+        "kernel_size=(4, 4, 4)\n",
+        "\n",
+        "padding = keras.layers.ZeroPadding3D(padding=(0,(1,0),0))\n",
+        "\n",
+        "conv = MaskedConv3D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "cropping = keras.layers.Cropping3D(cropping=(0,(0, 1), 0))\n",
+        "\n",
+        "\n",
+        "x = padding(test_ones_3d)\n",
+        "x = conv(x)\n",
+        "result = cropping(x)\n",
+        "\n",
+        "print('MASK')\n",
+        "print_3d(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print_3d(result.numpy().squeeze())"
       ],
-      "execution_count": 25,
+      "execution_count": 172,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "Depth 0\n",
+            "[[0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "Depth 1\n",
+            "[[1. 1. 1. 1.]\n",
+            " [1. 1. 1. 1.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "Depth 2\n",
+            "[[0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "Depth 3\n",
+            "[[0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "Depth 0\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]]\n",
+            " [6. 8. 8. 6. 4.]\n",
+            " [6. 8. 8. 6. 4.]\n",
+            " [6. 8. 8. 6. 4.]]\n",
             "Depth 1\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]]\n",
+            " [6. 8. 8. 6. 4.]\n",
+            " [6. 8. 8. 6. 4.]\n",
+            " [6. 8. 8. 6. 4.]]\n",
             "Depth 2\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]]\n",
+            " [6. 8. 8. 6. 4.]\n",
+            " [6. 8. 8. 6. 4.]\n",
+            " [6. 8. 8. 6. 4.]]\n",
             "Depth 3\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]]\n",
+            " [6. 8. 8. 6. 4.]\n",
+            " [6. 8. 8. 6. 4.]\n",
+            " [6. 8. 8. 6. 4.]]\n",
             "Depth 4\n",
             "[[0. 0. 0. 0. 0.]\n",
             " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]\n",
-            " [3. 4. 4. 3. 2.]]\n"
+            " [6. 8. 8. 6. 4.]\n",
+            " [6. 8. 8. 6. 4.]\n",
+            " [6. 8. 8. 6. 4.]]\n"
           ],
           "name": "stdout"
         }
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "ANQnjmvPwI3K",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Horizontal stack A"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -989,20 +1402,57 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 538
+          "height": 937
         },
-        "outputId": "cd8ec46e-f49e-4761-d7d3-f1d328148263"
+        "outputId": "298206a2-bd57-469f-f497-84ab21a970b5"
       },
       "source": [
-        "horizontal_A_stack = build_test_stack_3d('A', kernel_size=(1, 1, 4))\n",
-        "val = horizontal_A_stack.predict(test_ones_3d)\n",
-        "print_test_3d(val)"
+        "mask_type = 'A'\n",
+        "kernel_size=(4, 4, 4)\n",
+        "\n",
+        "conv = MaskedConv3D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "result = conv(test_ones_3d)\n",
+        "\n",
+        "print('MASK')\n",
+        "print_3d(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print_3d(result.numpy().squeeze())"
       ],
-      "execution_count": 26,
+      "execution_count": 173,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "Depth 0\n",
+            "[[0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "Depth 1\n",
+            "[[0. 0. 0. 0.]\n",
+            " [1. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "Depth 2\n",
+            "[[0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "Depth 3\n",
+            "[[0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]\n",
+            " [0. 0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "Depth 0\n",
             "[[0. 1. 1. 1. 1.]\n",
             " [0. 1. 1. 1. 1.]\n",
@@ -1038,6 +1488,16 @@
         }
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "KJzAJ40EwTL3",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Horizontal stack B"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -1045,20 +1505,49 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 538
+          "height": 799
         },
-        "outputId": "95efb670-2e51-462d-8b0d-97e4cc89852b"
+        "outputId": "fa304a57-4168-4bfd-d722-1503a2bc54e1"
       },
       "source": [
-        "horizontal_B_stack = build_test_stack_3d('B', kernel_size=(1, 1, 4))\n",
-        "val = horizontal_B_stack.predict(test_ones_3d)\n",
-        "print_test_3d(val)"
+        "mask_type = 'B'\n",
+        "kernel_size=(3, 3, 3)\n",
+        "\n",
+        "conv = MaskedConv3D(mask_type=mask_type,\n",
+        "                filters=1,\n",
+        "                kernel_size=kernel_size, \n",
+        "                padding='same',\n",
+        "                kernel_initializer='ones', \n",
+        "                bias_initializer='zeros')\n",
+        "\n",
+        "result = conv(test_ones_3d)\n",
+        "\n",
+        "print('MASK')\n",
+        "print_3d(conv.mask.numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print_3d(result.numpy().squeeze())"
       ],
-      "execution_count": 27,
+      "execution_count": 174,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "MASK\n",
+            "Depth 0\n",
+            "[[0. 0. 0.]\n",
+            " [0. 0. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "Depth 1\n",
+            "[[0. 0. 0.]\n",
+            " [1. 1. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "Depth 2\n",
+            "[[0. 0. 0.]\n",
+            " [0. 0. 0.]\n",
+            " [0. 0. 0.]]\n",
+            "\n",
+            "OUTPUT\n",
             "Depth 0\n",
             "[[1. 2. 2. 2. 2.]\n",
             " [1. 2. 2. 2. 2.]\n",
@@ -1101,7 +1590,7 @@
         "colab_type": "text"
       },
       "source": [
-        "## Creating 2D cropping solution"
+        "## Creating 2D cropped solution"
       ]
     },
     {
@@ -1112,8 +1601,6 @@
         "colab": {}
       },
       "source": [
-        "import math\n",
-        "\n",
         "class VerticalCroppedConv2d(tf.keras.Model):\n",
         "    def __init__(self,\n",
         "                 filters,\n",
@@ -1286,6 +1773,16 @@
         "###Tests with kernel_size 3"
       ]
     },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "DdcDFcMbxwpZ",
+        "colab_type": "text"
+      },
+      "source": [
+        "#### Vertical stack"
+      ]
+    },
     {
       "cell_type": "code",
       "metadata": {
@@ -1293,24 +1790,57 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 86
+          "height": 191
         },
-        "outputId": "194d0b97-c5ed-4cbc-c82c-82a4885ce7a9"
+        "outputId": "49313976-a256-48c4-b198-0e4bf5a66bc9"
       },
       "source": [
-        "vertical_stack = build_test_croppedv_stack_2d()\n",
-        "val = vertical_stack.predict(test_ones_2d)\n",
-        "print(val[0,:,:,0].squeeze())"
+        "kernel_size=(2, 3)\n",
+        "kernel_h, kernel_w = kernel_size\n",
+        "\n",
+        "\n",
+        "\n",
+        "padding1 = keras.layers.ZeroPadding2D(padding=((1,0),0))\n",
+        "\n",
+        "padding2 = keras.layers.ZeroPadding2D(padding=((kernel_h-1, 0),(int((kernel_w-1)/2),int((kernel_w-1)/2))))\n",
+        "conv = keras.layers.Conv2D(filters=1,\n",
+        "                           kernel_size=kernel_size,\n",
+        "                           strides=1,\n",
+        "                           padding='valid',\n",
+        "                           kernel_initializer='ones', \n",
+        "                           bias_initializer='zeros')\n",
+        "\n",
+        "cropping = keras.layers.Cropping2D(cropping=((0, 1), 0))\n",
+        "\n",
+        "\n",
+        "x = padding1(test_ones_2d)\n",
+        "x = padding2(x)\n",
+        "x = conv(x)\n",
+        "result = cropping(x)\n",
+        "\n",
+        "\n",
+        "print('KERNEL')\n",
+        "print(conv.weights[0].numpy().squeeze())\n",
+        "print('')\n",
+        "print('OUTPUT')\n",
+        "print(result.numpy().squeeze())\n",
+        "\n"
       ],
-      "execution_count": 59,
+      "execution_count": 182,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
-            "[[2. 3. 3. 3. 2.]\n",
+            "KERNEL\n",
+            "[[1. 1. 1.]\n",
+            " [1. 1. 1.]]\n",
+            "\n",
+            "OUTPUT\n",
+            "[[0. 0. 0. 0. 0.]\n",
+            " [2. 3. 3. 3. 2.]\n",
             " [4. 6. 6. 6. 4.]\n",
-            " [6. 9. 9. 9. 6.]\n",
-            " [6. 9. 9. 9. 6.]]\n"
+            " [4. 6. 6. 6. 4.]\n",
+            " [4. 6. 6. 6. 4.]]\n"
           ],
           "name": "stdout"
         }
@@ -1338,8 +1868,25 @@
       "source": [
         "REFERENCES\n",
         "\n",
+        "https://wiki.math.uwaterloo.ca/statwiki/index.php?title=STAT946F17/Conditional_Image_Generation_with_PixelCNN_Decoders#Gated_PixelCNN\n",
+        "\n",
+        "https://www.slideshare.net/suga93/conditional-image-generation-with-pixelcnn-decoders\n",
+        "\n",
         "https://www.youtube.com/watch?v=1BURwCCYNEI"
       ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "jTI9ts7i7Wch",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        ""
+      ],
+      "execution_count": 0,
+      "outputs": []
     }
   ]
 }
\ No newline at end of file