diff --git a/.gitignore b/.gitignore
index e6458330..051e5a71 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,2 +1,3 @@
 *~
-*.pyc
\ No newline at end of file
+*.pyc
+.DS_Store
\ No newline at end of file
diff --git a/docs/examples/dcn.ipynb b/docs/examples/dcn.ipynb
index 12da7d01..c73d92a4 100644
--- a/docs/examples/dcn.ipynb
+++ b/docs/examples/dcn.ipynb
@@ -14,7 +14,10 @@
       "execution_count": null,
       "metadata": {
         "cellView": "form",
-        "id": "uWqCArLO_kez"
+        "id": "uWqCArLO_kez",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -37,22 +40,22 @@
         "id": "ikhIvrku-i-L"
       },
       "source": [
-        "# Deep \u0026 Cross Network (DCN)\n",
-        "\n",
-        "\u003ctable class=\"tfo-notebook-buttons\" align=\"left\"\u003e\n",
-        "  \u003ctd\u003e\n",
-        "    \u003ca target=\"_blank\" href=\"https://www.tensorflow.org/recommenders/examples/dcn\"\u003e\u003cimg src=\"https://www.tensorflow.org/images/tf_logo_32px.png\" /\u003eView on TensorFlow.org\u003c/a\u003e\n",
-        "  \u003c/td\u003e\n",
-        "  \u003ctd\u003e\n",
-        "    \u003ca target=\"_blank\" href=\"https://colab.research.google.com/github/tensorflow/recommenders/blob/main/docs/examples/dcn.ipynb\"\u003e\u003cimg src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" /\u003eRun in Google Colab\u003c/a\u003e\n",
-        "  \u003c/td\u003e\n",
-        "  \u003ctd\u003e\n",
-        "    \u003ca target=\"_blank\" href=\"https://github.com/tensorflow/recommenders/blob/main/docs/examples/dcn.ipynb\"\u003e\u003cimg src=\"https://www.tensorflow.org/images/GitHub-Mark-32px.png\" /\u003eView source on GitHub\u003c/a\u003e\n",
-        "  \u003c/td\u003e\n",
-        "  \u003ctd\u003e\n",
-        "    \u003ca href=\"https://storage.googleapis.com/tensorflow_docs/recommenders/docs/examples/dcn.ipynb\"\u003e\u003cimg src=\"https://www.tensorflow.org/images/download_logo_32px.png\" /\u003eDownload notebook\u003c/a\u003e\n",
-        "  \u003c/td\u003e\n",
-        "\u003c/table\u003e"
+        "# Deep & Cross Network (DCN)\n",
+        "\n",
+        "<table class=\"tfo-notebook-buttons\" align=\"left\">\n",
+        "  <td>\n",
+        "    <a target=\"_blank\" href=\"https://www.tensorflow.org/recommenders/examples/dcn\"><img src=\"https://www.tensorflow.org/images/tf_logo_32px.png\" />View on TensorFlow.org</a>\n",
+        "  </td>\n",
+        "  <td>\n",
+        "    <a target=\"_blank\" href=\"https://colab.research.google.com/github/tensorflow/recommenders/blob/main/docs/examples/dcn.ipynb\"><img src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" />Run in Google Colab</a>\n",
+        "  </td>\n",
+        "  <td>\n",
+        "    <a target=\"_blank\" href=\"https://github.com/tensorflow/recommenders/blob/main/docs/examples/dcn.ipynb\"><img src=\"https://www.tensorflow.org/images/GitHub-Mark-32px.png\" />View source on GitHub</a>\n",
+        "  </td>\n",
+        "  <td>\n",
+        "    <a href=\"https://storage.googleapis.com/tensorflow_docs/recommenders/docs/examples/dcn.ipynb\"><img src=\"https://www.tensorflow.org/images/download_logo_32px.png\" />Download notebook</a>\n",
+        "  </td>\n",
+        "</table>"
       ]
     },
     {
@@ -61,16 +64,16 @@
         "id": "Q-rOX95bAye4"
       },
       "source": [
-        "This tutorial demonstrates how to use Deep \u0026 Cross Network (DCN) to effectively learn feature crosses.\n",
+        "This tutorial demonstrates how to use Deep & Cross Network (DCN) to effectively learn feature crosses.\n",
         "\n",
-        "##Background\n",
+        "## Background\n",
         "\n",
         "**What are feature crosses and why are they important?** Imagine that we are building a recommender system to sell a blender to customers. Then, a customer's past purchase history such as `purchased_bananas` and `purchased_cooking_books`, or geographic features, are single features. If one has purchased both bananas **and** cooking books, then this customer will more likely click on the recommended blender. The combination of `purchased_bananas` and `purchased_cooking_books` is referred to as a **feature cross**, which provides additional interaction information beyond the individual features.\n",
-        "\u003cdiv\u003e\n",
-        "\u003ccenter\u003e\n",
-        "\u003cimg src=\"http://drive.google.com/uc?export=view\u0026id=1e8pYZHM1ZSwqBLYVkKDoGg0_2t2UPc2y\" width=\"600\"/\u003e\n",
-        "\u003c/center\u003e\n",
-        "\u003c/div\u003e\n",
+        "<div>\n",
+        "<center>\n",
+        "<img src=\"./images/blender-draw.gif\" width=\"600\"/>\n",
+        "</center>\n",
+        "</div>\n",
         "\n",
         "\n",
         "\n",
@@ -78,28 +81,29 @@
         "**What are the challenges in learning feature crosses?** In Web-scale applications, data are mostly categorical, leading to large and sparse feature space. Identifying effective feature crosses in this setting often requires\n",
         "manual feature engineering or exhaustive search. Traditional feed-forward multilayer perceptron (MLP) models are universal function approximators; however, they cannot efficiently approximate even 2nd or 3rd-order feature crosses [[1](https://arxiv.org/pdf/2008.13535.pdf), [2](https://static.googleusercontent.com/media/research.google.com/en//pubs/archive/18fa88ad519f25dc4860567e19ab00beff3f01cb.pdf)].\n",
         "\n",
-        "**What is Deep \u0026 Cross Network (DCN)?** DCN was designed to learn explicit and bounded-degree cross features more effectively. It starts with an input layer (typically an embedding layer), followed by a *cross network* containing multiple cross layers that models explicit feature interactions, and then combines\n",
+        "**What is Deep & Cross Network (DCN)?** DCN was designed to learn explicit and bounded-degree cross features more effectively. It starts with an input layer (typically an embedding layer), followed by a *cross network* containing multiple cross layers that models explicit feature interactions, and then combines\n",
         "with a *deep network* that models implicit feature interactions.\n",
         "\n",
         "\n",
         "*   Cross Network. This is the core of DCN. It explicitly applies feature crossing at each layer, and the highest\n",
         "polynomial degree increases with layer depth. The following figure shows the $(i+1)$-th cross layer.\n",
-        "\u003cdiv class=\"fig figcenter fighighlight\"\u003e\n",
-        "\u003ccenter\u003e\n",
-        "  \u003cimg src=\"http://drive.google.com/uc?export=view\u0026id=1QvIDptMxixFNp6P4bBqMN4AYAhAIAYQZ\" width=\"50%\" style=\"display:block\"\u003e\n",
-        "  \u003c/center\u003e\n",
-        "\u003c/div\u003e\n",
+        "<div class=\"fig figcenter fighighlight\">\n",
+        "<center>\n",
+        "  <img src=\"./images/dcn-formula.png\" width=\"50%\" style=\"display:block\">\n",
+        "  </center>\n",
+        "</div>\n",
+        "\n",
         "*   Deep Network. It is a traditional feedforward multilayer perceptron (MLP).\n",
         "\n",
         "The deep network and cross network are then combined to form DCN [[1](https://arxiv.org/pdf/2008.13535.pdf)]. Commonly, we could stack a deep network on top of the cross network (stacked structure); we could also place them in parallel (parallel structure). \n",
         "\n",
         "\n",
-        "\u003cdiv class=\"fig figcenter fighighlight\"\u003e\n",
-        "\u003ccenter\u003e\n",
-        "  \u003cimg src=\"http://drive.google.com/uc?export=view\u0026id=1WtDUCV6b-eetUnWVCAmcPh8mJFut5EUd\" hspace=\"40\" width=\"30%\" style=\"margin: 0px 100px 0px 0px;\"\u003e\n",
-        "  \u003cimg src=\"http://drive.google.com/uc?export=view\u0026id=1xo_twKb847hasfss7JxF0UtFX_rEb4nt\" width=\"20%\"\u003e\n",
-        "  \u003c/center\u003e\n",
-        "\u003c/div\u003e"
+        "<div class=\"fig figcenter fighighlight\">\n",
+        "<center>\n",
+        "  <img src=\"./images/dcn-parallel.png\" hspace=\"40\" width=\"30%\" style=\"margin: 0px 100px 0px 0px;\">\n",
+        "  <img src=\"./images/dcn-stack.png\" width=\"20%\">\n",
+        "  </center>\n",
+        "</div>"
       ]
     },
     {
@@ -124,7 +128,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "PjfZWVEWAmxS"
+        "id": "PjfZWVEWAmxS",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -136,7 +143,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "DqsyLA0UHeCl"
+        "id": "DqsyLA0UHeCl",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -199,7 +209,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "9rT3f6C3GX0u"
+        "id": "9rT3f6C3GX0u",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -236,7 +249,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "vrQWVYajgmNV"
+        "id": "vrQWVYajgmNV",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -267,7 +283,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "bwgAH2FTR4Fe"
+        "id": "bwgAH2FTR4Fe",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -312,7 +331,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "EwBwSHz_N3pW"
+        "id": "EwBwSHz_N3pW",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -340,7 +362,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "X6gD-NTF4eoj"
+        "id": "X6gD-NTF4eoj",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -361,7 +386,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "nFhrC7fV6szW"
+        "id": "nFhrC7fV6szW",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -382,7 +410,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "F8ZXXbmKuB8p"
+        "id": "F8ZXXbmKuB8p",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -394,7 +425,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "Tzg3KLKW2sdA"
+        "id": "Tzg3KLKW2sdA",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -416,7 +450,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "l4PM-goX6FoD"
+        "id": "l4PM-goX6FoD",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -452,7 +489,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "N8dga2Qck5IV"
+        "id": "N8dga2Qck5IV",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -506,7 +546,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "7Y_n3EPosR4A"
+        "id": "7Y_n3EPosR4A",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -535,7 +578,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "a5-l91jR_zEo"
+        "id": "a5-l91jR_zEo",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -559,7 +605,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "l9qhEcHq_VfI"
+        "id": "l9qhEcHq_VfI",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -588,7 +637,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "6lrDcBjiwnHU"
+        "id": "6lrDcBjiwnHU",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -685,7 +737,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "qeFjmfUbgzcS"
+        "id": "qeFjmfUbgzcS",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -706,7 +761,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "gTDk3GloquHO"
+        "id": "gTDk3GloquHO",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -743,7 +801,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "Zy3kWb5Dh0E7"
+        "id": "Zy3kWb5Dh0E7",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -758,18 +819,21 @@
       },
       "source": [
         "**DCN (stacked).** We first train a DCN model with a stacked structure, that is, the inputs are fed to a cross network followed by a deep network.\n",
-        "\u003cdiv\u003e\n",
-        "\u003ccenter\u003e\n",
-        "\u003cimg src=\"http://drive.google.com/uc?export=view\u0026id=1X8qoMtIYKJz4yBYifvfw4QpAwrjr70e_\" width=\"140\"/\u003e\n",
-        "\u003c/center\u003e\n",
-        "\u003c/div\u003e\n"
+        "<div>\n",
+        "<center>\n",
+        "<img src=\"./images/dcn-stacked-simple.png\" width=\"140\"/>\n",
+        "</center>\n",
+        "</div>\n"
       ]
     },
     {
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "hiuYPJWhgw3J"
+        "id": "hiuYPJWhgw3J",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -785,18 +849,21 @@
       "source": [
         "**Low-rank DCN.** To reduce the training and serving cost, we leverage low-rank techniques to approximate the DCN weight matrices. The rank is passed in through argument `projection_dim`; a smaller `projection_dim` results in a lower cost. Note that `projection_dim` needs to be smaller than (input size)/2 to reduce the cost. In practice, we've observed using low-rank DCN with rank (input size)/4 consistently preserved the accuracy of a full-rank DCN.\n",
         "\n",
-        "\u003cdiv\u003e\n",
-        "\u003ccenter\u003e\n",
-        "\u003cimg src=\"http://drive.google.com/uc?export=view\u0026id=1ZZfUTNdxjGAaAuwNrweKkLJ1PGxMmiCm\" width=\"400\"/\u003e\n",
-        "\u003c/center\u003e\n",
-        "\u003c/div\u003e\n"
+        "<div>\n",
+        "<center>\n",
+        "<img src=\"./images/dcn_lowrank_simple.png\" width=\"400\"/>\n",
+        "</center>\n",
+        "</div>\n"
       ]
     },
     {
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "NYxbHI7ZNJX7"
+        "id": "NYxbHI7ZNJX7",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -818,7 +885,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "iBPpwD4cGtXF"
+        "id": "iBPpwD4cGtXF",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -839,7 +909,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "a1yj3pp0glEL"
+        "id": "a1yj3pp0glEL",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -872,14 +945,14 @@
         "\n",
         "*   *Concatenating cross layers.* The inputs are fed in parallel to multiple cross layers to capture complementary feature crosses.\n",
         "\n",
-        "\u003cdiv class=\"fig figcenter fighighlight\"\u003e\n",
-        "\u003ccenter\u003e\n",
-        "  \u003cimg src=\"http://drive.google.com/uc?export=view\u0026id=11RpNuj9s0OgSav9TUuGA7v7PuFLL6nVR\" hspace=40 width=\"600\" style=\"display:block;\"\u003e\n",
-        "  \u003cdiv class=\"figcaption\"\u003e\n",
-        "  \u003cb\u003eLeft\u003c/b\u003e: DCN with a parallel structure; \u003cb\u003eRight\u003c/b\u003e: Concatenating cross layers. \n",
-        "  \u003c/div\u003e\n",
-        "  \u003c/center\u003e\n",
-        "\u003c/div\u003e"
+        "<div class=\"fig figcenter fighighlight\">\n",
+        "<center>\n",
+        "  <img src=\"./images/dcn-more-simple.png\" hspace=40 width=\"600\" style=\"display:block;\">\n",
+        "  <div class=\"figcaption\">\n",
+        "  <b>Left</b>: DCN with a parallel structure; <b>Right</b>: Concatenating cross layers. \n",
+        "  </div>\n",
+        "  </center>\n",
+        "</div>"
       ]
     },
     {
@@ -903,7 +976,10 @@
       "cell_type": "code",
       "execution_count": null,
       "metadata": {
-        "id": "47ibaEBJxOoe"
+        "id": "47ibaEBJxOoe",
+        "vscode": {
+          "languageId": "python"
+        }
       },
       "outputs": [],
       "source": [
@@ -951,12 +1027,12 @@
         "---\n",
         "\n",
         "\n",
-        "##References\n",
-        "[DCN V2: Improved Deep \u0026 Cross Network and Practical Lessons for Web-scale Learning to Rank Systems](https://arxiv.org/pdf/2008.13535.pdf). \\\n",
+        "## References\n",
+        "[DCN V2: Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systems](https://arxiv.org/pdf/2008.13535.pdf). \\\n",
         "*Ruoxi Wang, Rakesh Shivanna, Derek Zhiyuan Cheng, Sagar Jain, Dong Lin, Lichan Hong, Ed Chi. (2020)*\n",
         "\n",
         "\n",
-        "[Deep \u0026 Cross Network for Ad Click Predictions](https://arxiv.org/pdf/1708.05123.pdf). \\\n",
+        "[Deep & Cross Network for Ad Click Predictions](https://arxiv.org/pdf/1708.05123.pdf). \\\n",
         "*Ruoxi Wang, Bin Fu, Gang Fu, Mingliang Wang. (AdKDD 2017)*"
       ]
     }
diff --git a/docs/examples/images/blender-draw.gif b/docs/examples/images/blender-draw.gif
new file mode 100644
index 00000000..a2f8da97
Binary files /dev/null and b/docs/examples/images/blender-draw.gif differ
diff --git a/docs/examples/images/dcn-formula.png b/docs/examples/images/dcn-formula.png
new file mode 100644
index 00000000..75d072a0
Binary files /dev/null and b/docs/examples/images/dcn-formula.png differ
diff --git a/docs/examples/images/dcn-more-simple.png b/docs/examples/images/dcn-more-simple.png
new file mode 100644
index 00000000..f95d9c85
Binary files /dev/null and b/docs/examples/images/dcn-more-simple.png differ
diff --git a/docs/examples/images/dcn-parallel.png b/docs/examples/images/dcn-parallel.png
new file mode 100644
index 00000000..6a370d55
Binary files /dev/null and b/docs/examples/images/dcn-parallel.png differ
diff --git a/docs/examples/images/dcn-stack.png b/docs/examples/images/dcn-stack.png
new file mode 100644
index 00000000..df56c649
Binary files /dev/null and b/docs/examples/images/dcn-stack.png differ
diff --git a/docs/examples/images/dcn-stacked-simple.png b/docs/examples/images/dcn-stacked-simple.png
new file mode 100644
index 00000000..4181a42d
Binary files /dev/null and b/docs/examples/images/dcn-stacked-simple.png differ
diff --git a/docs/examples/images/dcn_lowrank_simple.png b/docs/examples/images/dcn_lowrank_simple.png
new file mode 100644
index 00000000..b445bade
Binary files /dev/null and b/docs/examples/images/dcn_lowrank_simple.png differ