mlc-ai-notebooks/9_Computational_Graph_Optimization.ipynb

{
  "cells": [
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "Mpn1ti5Urdsv"
      },
      "source": [
        "# Ep9: Computational Graph Optimization\n",
        "\n",
        "主要是做算子融合"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "BVp0fHyRkYj6"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "tvm versin: 0.13.dev0\n"
          ]
        }
      ],
      "source": [
        "import tvm\n",
        "from tvm.ir.module import IRModule\n",
        "\n",
        "from tvm.script import tir as T, relax as R\n",
        "from tvm import te\n",
        "\n",
        "from tvm import relax\n",
        "# This is needed for deferring annotation parsing in TVMScript\n",
        "from __future__ import annotations \n",
        "\n",
        "import numpy as np\n",
        "import pickle\n",
        "\n",
        "print('tvm versin: %s' % tvm.__version__)"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "aBhwDhrlrSiK"
      },
      "source": [
        "## Pattern Match and Rewriting\n",
        "\n"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "F8Zqd1_qrmXN"
      },
      "source": [
        "1. 原始 code 如下，包含 2 个算子: relax.multipy 和 relax.add\n",
        "2. 背景：2 个算子可以融合成 `relax.ewise_fma`，功能相同，但融合后的计算效率更高。\n",
        "3. 目标，通过 compiler 中经典的 pattern match and rewrite 计算进行算子融合。"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "sR45I3mvm695"
      },
      "outputs": [],
      "source": [
        "@tvm.script.ir_module\n",
        "class MyModule:\n",
        "    @R.function\n",
        "    def main(x: R.Tensor((3, 4), \"float32\"), y: R.Tensor((3, 4), \"float32\")):\n",
        "        with R.dataflow():\n",
        "            lv0 = R.multiply(x, y)\n",
        "            gv0 = R.add(lv0, y)\n",
        "            R.output(gv0)\n",
        "        return gv0\n",
        "\n",
        "\n",
        "relax_func = MyModule[\"main\"]"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "SWGxwuDB0PMV"
      },
      "source": [
        "Before we dive into how to do that exactly, let us first examine the data structure that makes up the MyModule. Each IRModule contains a collection of functions, and the function body is composed of a set of data structures called abstract syntax trees (AST)."
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "TBUIvV6c4Gzw"
      },
      "source": [
        "Each function is represented by a relax.Function node"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 74,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SYM-u_IJ0vNy",
        "outputId": "26709d64-71b9-4d45-86d2-2d110524e6e7"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "tvm.relax.expr.Function"
            ]
          },
          "execution_count": 74,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "type(relax_func)"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "V4NAP23h4QM4"
      },
      "source": [
        "The function contains a list of parameters"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 75,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BeNcHrEd4Mw8",
        "outputId": "74afca54-7afd-4632-ceb6-2b592766896f"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[relax.expr.Var(0xa8f2d80), relax.expr.Var(0xa750480)]"
            ]
          },
          "execution_count": 75,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "relax_func.params"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "kQYT-1Ke4VQf"
      },
      "source": [
        "The function contains a body fields that represents its return value and set of binding blocks in the function."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 76,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0hhXxL_20x-N",
        "outputId": "ae106676-8101-4efc-e31f-4dc4513cc524"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "tvm.relax.expr.SeqExpr"
            ]
          },
          "execution_count": 76,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "func_body = relax_func.body\n",
        "type(func_body)"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "J4dQoc525CxL"
      },
      "source": [
        "The function body SeqExpr contains a sequence of (binding) blocks"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 77,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3nM-F8Gh2PzX",
        "outputId": "9df0aeb1-cc23-49b7-8b9c-fa8a81c6f5a3"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "[relax.expr.DataflowBlock(0x3444520)]"
            ]
          },
          "execution_count": 77,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "func_body.blocks"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 78,
      "metadata": {
        "id": "CV_byz4_2jmD"
      },
      "outputs": [],
      "source": [
        "dataflow_block = func_body.blocks[0]"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "OsDRAIB_5Lig"
      },
      "source": [
        "In our particular case, we have a single data flow block that contains\n",
        "two bindings. Each binding corresponds to one of the following two lines\n",
        "\n",
        "```python\n",
        "lv0 = relax.multiply(x, y)\n",
        "gv0 = relax.add(lv0, y)\n",
        "```"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Zw8Yw9SV2o7H"
      },
      "outputs": [],
      "source": [
        "dataflow_block.bindings"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "A8RZMKD22slK"
      },
      "outputs": [],
      "source": [
        "binding = dataflow_block.bindings[0]"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "yHKybEKE5hq3"
      },
      "source": [
        "Each binding have a var field that corresponds to the left hand side of the binding (lv0, gv0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 79,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nzzMhaSS29UR",
        "outputId": "6e03bfef-c441-4f40-dfda-0b88bb69bfcb"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "relax.expr.DataflowVar(0x6c36300)"
            ]
          },
          "execution_count": 79,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "binding.var"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "-cAOJAmF5o5Z"
      },
      "source": [
        "And its value field corresponds to the right-hand side of the binding. Each value field corresponds to a `relax.Call` node representing a call into a primitive function"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 80,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qfP5rEdD2_1c",
        "outputId": "0151ffb1-c5a6-4a06-db9f-b89d263e0895"
      },
      "outputs": [
        {
          "data": {
            "text/plain": [
              "CallNode(Op(relax.multiply), [relax.expr.Var(0x3463080), relax.expr.Var(0x3463300)], (nullptr), [])"
            ]
          },
          "execution_count": 80,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "binding.value"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "1oV8O46kWsyG"
      },
      "source": [
        "![image.png](data:image/png;base64,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)"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "iogK4y4kWxm9"
      },
      "source": [
        "The above figure summarizes the data structure involved in this particular function."
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "_8Q1V8b16K3N"
      },
      "source": [
        "One approach to rewrite the program would be to traverse MyModule's AST recursively and generate a transformed AST. We can certainly do that using the python API available. However, we can use extra tooling support to simplify the process. The following code block follows a design pattern called **visitor pattern** that allows us to visit each AST node and rewrite them to transformed versions."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 81,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 157
        },
        "id": "Rgz_VxzPsEx8",
        "outputId": "434671e8-3b3e-4d45-dd44-d5c67badd0a8"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div class=\"highlight\" style=\"background: \"><pre style=\"line-height: 125%\"><span></span><span style=\"color: #AA22FF\">@R</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>function\n",
              "<span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">main</span>(x: Tensor((<span style=\"color: #008000\">3</span>, <span style=\"color: #008000\">4</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>), y: Tensor((<span style=\"color: #008000\">3</span>, <span style=\"color: #008000\">4</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>)) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> Tensor(<span style=\"color: #008000; font-weight: bold\">None</span>, <span style=\"color: #BA2121\">&quot;float32&quot;</span>, ndim <span style=\"color: #AA22FF; font-weight: bold\">=</span> <span style=\"color: #008000\">2</span>):\n",
              "    <span style=\"color: #007979; font-style: italic\"># block 0</span>\n",
              "    <span style=\"color: #008000; font-weight: bold\">with</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>dataflow():\n",
              "        lv0: Tensor((<span style=\"color: #008000\">3</span>, <span style=\"color: #008000\">4</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>multiply(x, y)\n",
              "        gv0: Tensor((<span style=\"color: #008000\">3</span>, <span style=\"color: #008000\">4</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>ewise_fma(x, y, y)\n",
              "        R<span style=\"color: #AA22FF; font-weight: bold\">.</span>output(gv0)\n",
              "    <span style=\"color: #008000; font-weight: bold\">return</span> gv0\n",
              "</pre></div>\n"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "@relax.expr_functor.mutator\n",
        "class EwiseFMARewriter(relax.PyExprMutator):\n",
        "    def visit_call_(self, call):\n",
        "        call = self.visit_expr_post_order(call)\n",
        "        add_op = tvm.ir.Op.get(\"relax.add\")\n",
        "        multiply_op = tvm.ir.Op.get(\"relax.multiply\")\n",
        "        ewise_fma_op = tvm.ir.Op.get(\"relax.ewise_fma\")\n",
        "\n",
        "        if call.op != add_op:\n",
        "            return call\n",
        "\n",
        "        value = self.lookup_binding(call.args[0])\n",
        "        if not isinstance(value, relax.Call) or value.op != multiply_op:\n",
        "            return call\n",
        "        \n",
        "        fma_call = relax.Call(\n",
        "            ewise_fma_op, [value.args[0], value.args[1], call.args[1]], None, None\n",
        "        )\n",
        "        return fma_call\n",
        "\n",
        "\n",
        "updated_fn = EwiseFMARewriter().visit_expr(MyModule[\"main\"])\n",
        "updated_fn.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "x24qphBw7Kli"
      },
      "source": [
        "We can go ahead and run the code. Note that the result rewrites gv0 to the fused operator but leaves lv0 in the code. We can use `remove_all_unused` to further simplify the code block."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 82,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 140
        },
        "id": "Nne2dRY6OzxI",
        "outputId": "abe1c9c4-6ad5-48f2-f28a-d3e8bfdef450"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div class=\"highlight\" style=\"background: \"><pre style=\"line-height: 125%\"><span></span><span style=\"color: #AA22FF\">@R</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>function\n",
              "<span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">main</span>(x: Tensor((<span style=\"color: #008000\">3</span>, <span style=\"color: #008000\">4</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>), y: Tensor((<span style=\"color: #008000\">3</span>, <span style=\"color: #008000\">4</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>)) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> Tensor(<span style=\"color: #008000; font-weight: bold\">None</span>, <span style=\"color: #BA2121\">&quot;float32&quot;</span>, ndim <span style=\"color: #AA22FF; font-weight: bold\">=</span> <span style=\"color: #008000\">2</span>):\n",
              "    <span style=\"color: #007979; font-style: italic\"># block 0</span>\n",
              "    <span style=\"color: #008000; font-weight: bold\">with</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>dataflow():\n",
              "        gv0: Tensor((<span style=\"color: #008000\">3</span>, <span style=\"color: #008000\">4</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>ewise_fma(x, y, y)\n",
              "        R<span style=\"color: #AA22FF; font-weight: bold\">.</span>output(gv0)\n",
              "    <span style=\"color: #008000; font-weight: bold\">return</span> gv0\n",
              "</pre></div>\n"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "relax.analysis.remove_all_unused(updated_fn).show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "xQxmQX7xc8I9"
      },
      "source": [
        "## Fuse Linear and ReLU"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "xhtWogRl7q0V"
      },
      "source": [
        "Now we have get a basic taste of graph rewriting. Let us try it on an end to end model. "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 83,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cEANm229O4p2",
        "outputId": "6d4f91f2-d81c-49c9-e0bc-ac58d670d8c7"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "File ‘fasionmnist_mlp_params.pkl’ already there; not retrieving.\n",
            "\n"
          ]
        }
      ],
      "source": [
        "!wget -nc https://github.com/mlc-ai/web-data/raw/main/models/fasionmnist_mlp_params.pkl"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 84,
      "metadata": {
        "id": "jhkS8P3kVl58"
      },
      "outputs": [],
      "source": [
        "import pickle as pkl\n",
        "mlp_params = pkl.load(open(\"fasionmnist_mlp_params.pkl\", \"rb\"))"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "wOEMkZtQ8DSx"
      },
      "source": [
        "The following code reconstructs the FashionMNIST MLP model we used in our past chapters. To simplify our explaination, we directly construct the model using high-level operators such as `relax.op.add` and `relax.op.dense`."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 85,
      "metadata": {
        "id": "mpsG2ZTEUp68"
      },
      "outputs": [],
      "source": [
        "def create_model():\n",
        "    bb = relax.BlockBuilder()\n",
        "    x = relax.Var(\"x\", (1, 784), relax.DynTensorType(2, \"float32\"))\n",
        "    w0 = relax.const(mlp_params[\"w0\"], \"float32\")\n",
        "    b0 = relax.const(mlp_params[\"b0\"], \"float32\")\n",
        "    w1 = relax.const(mlp_params[\"w1\"], \"float32\")\n",
        "    b1 = relax.const(mlp_params[\"b1\"], \"float32\")\n",
        "\n",
        "    with bb.function(\"main\", [x]):\n",
        "        with bb.dataflow():\n",
        "            lv0 = bb.emit(relax.op.dense(x, w0))\n",
        "            lv1 = bb.emit(relax.op.add(lv0, b0))\n",
        "            lv2 = bb.emit(relax.op.relu(lv1))\n",
        "            lv3 = bb.emit(relax.op.dense(lv2, w1))\n",
        "            lv4 = bb.emit(relax.op.add(lv3, b1))\n",
        "            gv = bb.emit_output(lv4)\n",
        "        bb.emit_func_output(gv)\n",
        "\n",
        "    return bb.get()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 86,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 280
        },
        "id": "2sbHd9ZkVD1R",
        "outputId": "9e90aa17-6e1f-435d-9e75-88662bcb0abf"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div class=\"highlight\" style=\"background: \"><pre style=\"line-height: 125%\"><span></span><span style=\"color: #AA22FF\">@tvm</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>script<span style=\"color: #AA22FF; font-weight: bold\">.</span>ir_module\n",
              "<span style=\"color: #008000; font-weight: bold\">class</span> <span style=\"color: #1E90FF; font-weight: bold\">Module</span>:\n",
              "    <span style=\"color: #AA22FF\">@R</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>function\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">main</span>(x: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">784</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>)) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> Tensor(<span style=\"color: #008000; font-weight: bold\">None</span>, <span style=\"color: #BA2121\">&quot;float32&quot;</span>, ndim <span style=\"color: #AA22FF; font-weight: bold\">=</span> <span style=\"color: #008000\">2</span>):\n",
              "        <span style=\"color: #007979; font-style: italic\"># block 0</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">with</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>dataflow():\n",
              "            lv: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>nn<span style=\"color: #AA22FF; font-weight: bold\">.</span>dense(x, meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">0</span>])\n",
              "            lv1: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>add(lv, meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">1</span>])\n",
              "            lv2: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>nn<span style=\"color: #AA22FF; font-weight: bold\">.</span>relu(lv1)\n",
              "            lv3: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>nn<span style=\"color: #AA22FF; font-weight: bold\">.</span>dense(lv2, meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">2</span>])\n",
              "            lv4: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>add(lv3, meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">3</span>])\n",
              "            gv: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> lv4\n",
              "            R<span style=\"color: #AA22FF; font-weight: bold\">.</span>output(gv)\n",
              "        <span style=\"color: #008000; font-weight: bold\">return</span> gv\n",
              "    \n",
              "</pre></div>\n"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "MLPModel = create_model()\n",
        "\n",
        "MLPModel.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "19dVn4w-8pqV"
      },
      "source": [
        "We aim to \"fuse\" the dense and add operations into a single group. The following code achieves that through the following steps:\n",
        "\n",
        "- Identify `dense` and `add` patterns.\n",
        "- Generate another fused sub-function that calls into the dense and add operators.\n",
        "- Replace `dense` and `add` with the fused sub-functions."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 87,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 560
        },
        "id": "DOEj4iu5htFE",
        "outputId": "3dd81363-5f00-4fb9-a499-0b0061b0084d"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div class=\"highlight\" style=\"background: \"><pre style=\"line-height: 125%\"><span></span><span style=\"color: #AA22FF\">@tvm</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>script<span style=\"color: #AA22FF; font-weight: bold\">.</span>ir_module\n",
              "<span style=\"color: #008000; font-weight: bold\">class</span> <span style=\"color: #1E90FF; font-weight: bold\">Module</span>:\n",
              "    <span style=\"color: #AA22FF\">@R</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>function\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">fused_dense_add0</span>(x: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">784</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>), w: Tensor((<span style=\"color: #008000\">128</span>, <span style=\"color: #008000\">784</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>), b: Tensor((<span style=\"color: #008000\">128</span>,), <span style=\"color: #BA2121\">&quot;float32&quot;</span>)) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> Tensor(<span style=\"color: #008000; font-weight: bold\">None</span>, <span style=\"color: #BA2121\">&quot;float32&quot;</span>, ndim <span style=\"color: #AA22FF; font-weight: bold\">=</span> <span style=\"color: #008000\">2</span>):\n",
              "        <span style=\"color: #007979; font-style: italic\"># block 0</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">with</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>dataflow():\n",
              "            lv: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>nn<span style=\"color: #AA22FF; font-weight: bold\">.</span>dense(x, w)\n",
              "            gv: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>add(lv, b)\n",
              "            R<span style=\"color: #AA22FF; font-weight: bold\">.</span>output(gv)\n",
              "        <span style=\"color: #008000; font-weight: bold\">return</span> gv\n",
              "    \n",
              "    <span style=\"color: #AA22FF\">@R</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>function\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">fused_dense_add1</span>(x1: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>), w1: Tensor((<span style=\"color: #008000\">10</span>, <span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>), b1: Tensor((<span style=\"color: #008000\">10</span>,), <span style=\"color: #BA2121\">&quot;float32&quot;</span>)) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> Tensor(<span style=\"color: #008000; font-weight: bold\">None</span>, <span style=\"color: #BA2121\">&quot;float32&quot;</span>, ndim <span style=\"color: #AA22FF; font-weight: bold\">=</span> <span style=\"color: #008000\">2</span>):\n",
              "        <span style=\"color: #007979; font-style: italic\"># block 0</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">with</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>dataflow():\n",
              "            lv1: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>nn<span style=\"color: #AA22FF; font-weight: bold\">.</span>dense(x1, w1)\n",
              "            gv1: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>add(lv1, b1)\n",
              "            R<span style=\"color: #AA22FF; font-weight: bold\">.</span>output(gv1)\n",
              "        <span style=\"color: #008000; font-weight: bold\">return</span> gv1\n",
              "    \n",
              "    <span style=\"color: #AA22FF\">@R</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>function\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">main</span>(x2: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">784</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>)) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> Tensor(<span style=\"color: #008000; font-weight: bold\">None</span>, <span style=\"color: #BA2121\">&quot;float32&quot;</span>, ndim <span style=\"color: #AA22FF; font-weight: bold\">=</span> <span style=\"color: #008000\">2</span>):\n",
              "        <span style=\"color: #007979; font-style: italic\"># block 0</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">with</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>dataflow():\n",
              "            lv11: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> fused_dense_add0(x2, meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">0</span>], meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">1</span>])\n",
              "            lv2: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> relax<span style=\"color: #AA22FF; font-weight: bold\">.</span>nn<span style=\"color: #AA22FF; font-weight: bold\">.</span>relu(lv11)\n",
              "            lv4: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> fused_dense_add1(lv2, meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">2</span>], meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">3</span>])\n",
              "            gv2: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> lv4\n",
              "            R<span style=\"color: #AA22FF; font-weight: bold\">.</span>output(gv2)\n",
              "        <span style=\"color: #008000; font-weight: bold\">return</span> gv2\n",
              "    \n",
              "</pre></div>\n"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "@relax.expr_functor.mutator\n",
        "class DenseAddFusor(relax.PyExprMutator):\n",
        "    def __init__(self, mod: IRModule) -> None:\n",
        "        super().__init__()\n",
        "        self.mod_ = mod\n",
        "        # cache pre-defined ops\n",
        "        self.add_op = tvm.ir.Op.get(\"relax.add\")\n",
        "        self.dense_op = tvm.ir.Op.get(\"relax.nn.dense\")\n",
        "        self.counter = 0\n",
        "\n",
        "    def transform(self) -> IRModule:\n",
        "        for global_var, func in self.mod_.functions.items():\n",
        "            if not isinstance(func, relax.Function):\n",
        "                continue\n",
        "            # avoid already fused primitive functions\n",
        "            if \"Primitive\" in func.attrs.keys() and func.attrs[\"Primitive\"] != 0:\n",
        "                continue\n",
        "            updated_func = self.visit_expr(func)\n",
        "            updated_func = relax.analysis.remove_all_unused(updated_func)\n",
        "            self.builder_.update_func(global_var, updated_func)\n",
        "        \n",
        "        return self.builder_.get()\n",
        "\n",
        "    def visit_call_(self, call):\n",
        "        call = self.visit_expr_post_order(call)\n",
        "\n",
        "        def match_call(node, op):\n",
        "            if not isinstance(node, relax.Call):\n",
        "                return False\n",
        "            return node.op == op\n",
        "          \n",
        "        # pattern match dense => add\n",
        "        if not match_call(call, self.add_op):\n",
        "            return call\n",
        "\n",
        "        value = self.lookup_binding(call.args[0])\n",
        "        if value is None:\n",
        "            return call\n",
        "\n",
        "        if not match_call(value, self.dense_op):\n",
        "            return call\n",
        "\n",
        "        x = value.args[0]\n",
        "        w = value.args[1]\n",
        "        b = call.args[1]\n",
        "\n",
        "        # construct a new fused primitive function\n",
        "        param_x = relax.Var(\"x\", x.shape_, x._checked_type_)\n",
        "        param_w = relax.Var(\"w\", w.shape_, w._checked_type_)\n",
        "        param_b = relax.Var(\"b\", b.shape_, b._checked_type_)\n",
        "\n",
        "        bb = relax.BlockBuilder()\n",
        "\n",
        "        fn_name = \"fused_dense_add%d\" % (self.counter)\n",
        "        self.counter += 1\n",
        "        with bb.function(fn_name, [param_x, param_w, param_b]):\n",
        "            with bb.dataflow():\n",
        "                lv0 = bb.emit(relax.op.nn.dense(param_x, param_w))\n",
        "                gv = bb.emit_output(relax.op.add(lv0, param_b))\n",
        "            bb.emit_func_output(gv)\n",
        "\n",
        "        # Add Primitive attribute to the fused funtions\n",
        "        fused_fn = bb.get()[fn_name].with_attr(\"Primitive\", 1)\n",
        "        global_var = self.builder_.add_func(fused_fn, fn_name)\n",
        "\n",
        "        # construct call into the fused function\n",
        "        return relax.Call(global_var, [x, w, b], None, None)\n",
        "\n",
        "@tvm.ir.transform.module_pass(opt_level=2, name=\"DeseAddFuse\")\n",
        "class FuseDenseAddPass:\n",
        "    \"\"\"The wrapper for the LowerTensorIR pass.\"\"\"\n",
        "    def transform_module(self, mod, ctx):\n",
        "        return DenseAddFusor(mod).transform()\n",
        "\n",
        "\n",
        "MLPFused = FuseDenseAddPass()(MLPModel)\n",
        "MLPFused.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "MNt4wN409XfN"
      },
      "source": [
        "### Why Creating a Sub-function\n",
        "\n",
        "In the above example, we created two sub-functions with the prefix `fuse_dense_add`. These sub-function bodies contain information about the operations performed by the fused operator. An alternative to this rewriting is simply creating a separate primitive operation for the fused operator (like `ewise_fma`). However, as we are looking into fusing more operators, there can be an exponential amount of possible combinations. A sub-function that groups the fused operation together provides the same amount of information for follow-up code lowering without introducing a dedicated high-level operator for each fusion pattern."
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "n-4rxNjX-nhY"
      },
      "source": [
        "## Map to TensorIR Calls"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "4UB0YVIE-z6c"
      },
      "source": [
        "The fused IRModule only contains calls into high-level operations. To further low-level optimization and code generation, we need to translate those high-level primitive operators into corresponding TensorIR functions (or environment library functions).\n",
        "\n",
        "The following code  remaps high-level operations to the corresponding TensorIR functions. Here we leverage the internal block builder in each Mutator and return the transformed value using `call_te`."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 89,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "xqQnVkrHVNNU",
        "outputId": "fc8282ae-05ee-4a62-9d42-87bdf2370ff3"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div class=\"highlight\" style=\"background: \"><pre style=\"line-height: 125%\"><span></span><span style=\"color: #AA22FF\">@tvm</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>script<span style=\"color: #AA22FF; font-weight: bold\">.</span>ir_module\n",
              "<span style=\"color: #008000; font-weight: bold\">class</span> <span style=\"color: #1E90FF; font-weight: bold\">Module</span>:\n",
              "    <span style=\"color: #AA22FF\">@R</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>function\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">main</span>(x: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">784</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>)) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> Tensor(<span style=\"color: #008000; font-weight: bold\">None</span>, <span style=\"color: #BA2121\">&quot;float32&quot;</span>, ndim <span style=\"color: #AA22FF; font-weight: bold\">=</span> <span style=\"color: #008000\">2</span>):\n",
              "        <span style=\"color: #007979; font-style: italic\"># block 0</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">with</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>dataflow():\n",
              "            lv1: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> fused_dense_add0(x, meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">0</span>], meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">1</span>])\n",
              "            lv2 <span style=\"color: #AA22FF; font-weight: bold\">=</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>call_tir(relu, (lv1,), (<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), dtype<span style=\"color: #AA22FF; font-weight: bold\">=</span><span style=\"color: #BA2121\">&quot;float32&quot;</span>)\n",
              "            lv4: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> fused_dense_add1(lv2, meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">2</span>], meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">3</span>])\n",
              "            gv: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> lv4\n",
              "            R<span style=\"color: #AA22FF; font-weight: bold\">.</span>output(gv)\n",
              "        <span style=\"color: #008000; font-weight: bold\">return</span> gv\n",
              "    \n",
              "    <span style=\"color: #AA22FF\">@T</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>prim_func\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">dense</span>(rxplaceholder: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">784</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], rxplaceholder_1: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">784</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], T_matmul_NT: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>]) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> <span style=\"color: #008000; font-weight: bold\">None</span>:\n",
              "        <span style=\"color: #007979; font-style: italic\"># function attr dict</span>\n",
              "        T<span style=\"color: #AA22FF; font-weight: bold\">.</span>func_attr({<span style=\"color: #BA2121\">&quot;global_symbol&quot;</span>: <span style=\"color: #BA2121\">&quot;dense&quot;</span>, <span style=\"color: #BA2121\">&quot;tir.noalias&quot;</span>: <span style=\"color: #008000; font-weight: bold\">True</span>, <span style=\"color: #BA2121\">&quot;layout_free_buffers&quot;</span>: [<span style=\"color: #008000\">1</span>]})\n",
              "        <span style=\"color: #007979; font-style: italic\"># body</span>\n",
              "        <span style=\"color: #007979; font-style: italic\"># with T.block(&quot;root&quot;)</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">for</span> i0, i1, i2 <span style=\"color: #008000; font-weight: bold\">in</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>grid(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), <span style=\"color: #008000\">784</span>):\n",
              "            <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>block(<span style=\"color: #BA2121\">&quot;T_matmul_NT&quot;</span>):\n",
              "                i <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(<span style=\"color: #008000\">1</span>, i0)\n",
              "                j <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), i1)\n",
              "                k <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>reduce(<span style=\"color: #008000\">784</span>, i2)\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>reads(rxplaceholder[i, k], rxplaceholder_1[j, k])\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>writes(T_matmul_NT[i, j])\n",
              "                <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>init():\n",
              "                    T_matmul_NT[i, j] <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>float32(<span style=\"color: #008000\">0</span>)\n",
              "                T_matmul_NT[i, j] <span style=\"color: #AA22FF; font-weight: bold\">=</span> T_matmul_NT[i, j] <span style=\"color: #AA22FF; font-weight: bold\">+</span> rxplaceholder[i, k] <span style=\"color: #AA22FF; font-weight: bold\">*</span> rxplaceholder_1[j, k]\n",
              "    \n",
              "    <span style=\"color: #AA22FF\">@T</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>prim_func\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">add1</span>(rxplaceholder: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], rxplaceholder_1: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], T_add: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>]) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> <span style=\"color: #008000; font-weight: bold\">None</span>:\n",
              "        <span style=\"color: #007979; font-style: italic\"># function attr dict</span>\n",
              "        T<span style=\"color: #AA22FF; font-weight: bold\">.</span>func_attr({<span style=\"color: #BA2121\">&quot;global_symbol&quot;</span>: <span style=\"color: #BA2121\">&quot;add1&quot;</span>, <span style=\"color: #BA2121\">&quot;tir.noalias&quot;</span>: <span style=\"color: #008000; font-weight: bold\">True</span>})\n",
              "        <span style=\"color: #007979; font-style: italic\"># body</span>\n",
              "        <span style=\"color: #007979; font-style: italic\"># with T.block(&quot;root&quot;)</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">for</span> i0, i1 <span style=\"color: #008000; font-weight: bold\">in</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>grid(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>)):\n",
              "            <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>block(<span style=\"color: #BA2121\">&quot;T_add&quot;</span>):\n",
              "                ax0 <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(<span style=\"color: #008000\">1</span>, i0)\n",
              "                ax1 <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>), i1)\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>reads(rxplaceholder[ax0, ax1], rxplaceholder_1[ax1])\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>writes(T_add[ax0, ax1])\n",
              "                T_add[ax0, ax1] <span style=\"color: #AA22FF; font-weight: bold\">=</span> rxplaceholder[ax0, ax1] <span style=\"color: #AA22FF; font-weight: bold\">+</span> rxplaceholder_1[ax1]\n",
              "    \n",
              "    <span style=\"color: #AA22FF\">@T</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>prim_func\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">add</span>(rxplaceholder: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], rxplaceholder_1: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], T_add: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>]) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> <span style=\"color: #008000; font-weight: bold\">None</span>:\n",
              "        <span style=\"color: #007979; font-style: italic\"># function attr dict</span>\n",
              "        T<span style=\"color: #AA22FF; font-weight: bold\">.</span>func_attr({<span style=\"color: #BA2121\">&quot;global_symbol&quot;</span>: <span style=\"color: #BA2121\">&quot;add&quot;</span>, <span style=\"color: #BA2121\">&quot;tir.noalias&quot;</span>: <span style=\"color: #008000; font-weight: bold\">True</span>})\n",
              "        <span style=\"color: #007979; font-style: italic\"># body</span>\n",
              "        <span style=\"color: #007979; font-style: italic\"># with T.block(&quot;root&quot;)</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">for</span> i0, i1 <span style=\"color: #008000; font-weight: bold\">in</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>grid(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)):\n",
              "            <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>block(<span style=\"color: #BA2121\">&quot;T_add&quot;</span>):\n",
              "                ax0 <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(<span style=\"color: #008000\">1</span>, i0)\n",
              "                ax1 <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), i1)\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>reads(rxplaceholder[ax0, ax1], rxplaceholder_1[ax1])\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>writes(T_add[ax0, ax1])\n",
              "                T_add[ax0, ax1] <span style=\"color: #AA22FF; font-weight: bold\">=</span> rxplaceholder[ax0, ax1] <span style=\"color: #AA22FF; font-weight: bold\">+</span> rxplaceholder_1[ax1]\n",
              "    \n",
              "    <span style=\"color: #AA22FF\">@R</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>function\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">fused_dense_add1</span>(x1: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>), w: Tensor((<span style=\"color: #008000\">10</span>, <span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>), b: Tensor((<span style=\"color: #008000\">10</span>,), <span style=\"color: #BA2121\">&quot;float32&quot;</span>)) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> Tensor(<span style=\"color: #008000; font-weight: bold\">None</span>, <span style=\"color: #BA2121\">&quot;float32&quot;</span>, ndim <span style=\"color: #AA22FF; font-weight: bold\">=</span> <span style=\"color: #008000\">2</span>):\n",
              "        <span style=\"color: #007979; font-style: italic\"># block 0</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">with</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>dataflow():\n",
              "            lv <span style=\"color: #AA22FF; font-weight: bold\">=</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>call_tir(dense1, (x1, w), (<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), dtype<span style=\"color: #AA22FF; font-weight: bold\">=</span><span style=\"color: #BA2121\">&quot;float32&quot;</span>)\n",
              "            gv1 <span style=\"color: #AA22FF; font-weight: bold\">=</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>call_tir(add1, (lv, b), (<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), dtype<span style=\"color: #AA22FF; font-weight: bold\">=</span><span style=\"color: #BA2121\">&quot;float32&quot;</span>)\n",
              "            R<span style=\"color: #AA22FF; font-weight: bold\">.</span>output(gv1)\n",
              "        <span style=\"color: #008000; font-weight: bold\">return</span> gv1\n",
              "    \n",
              "    <span style=\"color: #AA22FF\">@T</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>prim_func\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">dense1</span>(rxplaceholder: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], rxplaceholder_1: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>), T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], T_matmul_NT: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>]) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> <span style=\"color: #008000; font-weight: bold\">None</span>:\n",
              "        <span style=\"color: #007979; font-style: italic\"># function attr dict</span>\n",
              "        T<span style=\"color: #AA22FF; font-weight: bold\">.</span>func_attr({<span style=\"color: #BA2121\">&quot;global_symbol&quot;</span>: <span style=\"color: #BA2121\">&quot;dense1&quot;</span>, <span style=\"color: #BA2121\">&quot;tir.noalias&quot;</span>: <span style=\"color: #008000; font-weight: bold\">True</span>, <span style=\"color: #BA2121\">&quot;layout_free_buffers&quot;</span>: [<span style=\"color: #008000\">1</span>]})\n",
              "        <span style=\"color: #007979; font-style: italic\"># body</span>\n",
              "        <span style=\"color: #007979; font-style: italic\"># with T.block(&quot;root&quot;)</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">for</span> i0, i1, i2 <span style=\"color: #008000; font-weight: bold\">in</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>grid(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>), T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)):\n",
              "            <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>block(<span style=\"color: #BA2121\">&quot;T_matmul_NT&quot;</span>):\n",
              "                i <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(<span style=\"color: #008000\">1</span>, i0)\n",
              "                j <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>), i1)\n",
              "                k <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>reduce(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), i2)\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>reads(rxplaceholder[i, k], rxplaceholder_1[j, k])\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>writes(T_matmul_NT[i, j])\n",
              "                <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>init():\n",
              "                    T_matmul_NT[i, j] <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>float32(<span style=\"color: #008000\">0</span>)\n",
              "                T_matmul_NT[i, j] <span style=\"color: #AA22FF; font-weight: bold\">=</span> T_matmul_NT[i, j] <span style=\"color: #AA22FF; font-weight: bold\">+</span> rxplaceholder[i, k] <span style=\"color: #AA22FF; font-weight: bold\">*</span> rxplaceholder_1[j, k]\n",
              "    \n",
              "    <span style=\"color: #AA22FF\">@T</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>prim_func\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">relu</span>(rxplaceholder: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], compute: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>]) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> <span style=\"color: #008000; font-weight: bold\">None</span>:\n",
              "        <span style=\"color: #007979; font-style: italic\"># function attr dict</span>\n",
              "        T<span style=\"color: #AA22FF; font-weight: bold\">.</span>func_attr({<span style=\"color: #BA2121\">&quot;global_symbol&quot;</span>: <span style=\"color: #BA2121\">&quot;relu&quot;</span>, <span style=\"color: #BA2121\">&quot;tir.noalias&quot;</span>: <span style=\"color: #008000; font-weight: bold\">True</span>})\n",
              "        <span style=\"color: #007979; font-style: italic\"># body</span>\n",
              "        <span style=\"color: #007979; font-style: italic\"># with T.block(&quot;root&quot;)</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">for</span> i0, i1 <span style=\"color: #008000; font-weight: bold\">in</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>grid(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)):\n",
              "            <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>block(<span style=\"color: #BA2121\">&quot;compute&quot;</span>):\n",
              "                i0_1 <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(<span style=\"color: #008000\">1</span>, i0)\n",
              "                i1_1 <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), i1)\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>reads(rxplaceholder[i0_1, i1_1])\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>writes(compute[i0_1, i1_1])\n",
              "                compute[i0_1, i1_1] <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>max(rxplaceholder[i0_1, i1_1], T<span style=\"color: #AA22FF; font-weight: bold\">.</span>float32(<span style=\"color: #008000\">0</span>))\n",
              "    \n",
              "    <span style=\"color: #AA22FF\">@R</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>function\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">fused_dense_add0</span>(x2: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">784</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>), w1: Tensor((<span style=\"color: #008000\">128</span>, <span style=\"color: #008000\">784</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>), b1: Tensor((<span style=\"color: #008000\">128</span>,), <span style=\"color: #BA2121\">&quot;float32&quot;</span>)) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> Tensor(<span style=\"color: #008000; font-weight: bold\">None</span>, <span style=\"color: #BA2121\">&quot;float32&quot;</span>, ndim <span style=\"color: #AA22FF; font-weight: bold\">=</span> <span style=\"color: #008000\">2</span>):\n",
              "        <span style=\"color: #007979; font-style: italic\"># block 0</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">with</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>dataflow():\n",
              "            lv3 <span style=\"color: #AA22FF; font-weight: bold\">=</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>call_tir(dense, (x2, w1), (<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), dtype<span style=\"color: #AA22FF; font-weight: bold\">=</span><span style=\"color: #BA2121\">&quot;float32&quot;</span>)\n",
              "            gv2 <span style=\"color: #AA22FF; font-weight: bold\">=</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>call_tir(add, (lv3, b1), (<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), dtype<span style=\"color: #AA22FF; font-weight: bold\">=</span><span style=\"color: #BA2121\">&quot;float32&quot;</span>)\n",
              "            R<span style=\"color: #AA22FF; font-weight: bold\">.</span>output(gv2)\n",
              "        <span style=\"color: #008000; font-weight: bold\">return</span> gv2\n",
              "    \n",
              "</pre></div>\n"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "@relax.expr_functor.mutator\n",
        "class LowerToTensorIR(relax.PyExprMutator):\n",
        "    def __init__(self, mod: IRModule, op_map) -> None:\n",
        "        super().__init__()\n",
        "        self.mod_ = mod\n",
        "        self.op_map = {\n",
        "            tvm.ir.Op.get(k): v for k, v in op_map.items()\n",
        "        }\n",
        "\n",
        "\n",
        "    def visit_call_(self, call):\n",
        "        call = self.visit_expr_post_order(call)\n",
        "\n",
        "        if call.op in self.op_map:\n",
        "            return self.op_map[call.op](self.builder_, call)\n",
        "        return call\n",
        "\n",
        "    def transform(self) -> IRModule:\n",
        "        for global_var, func in self.mod_.functions.items():\n",
        "            if not isinstance(func, relax.Function):\n",
        "                continue\n",
        "            updated_func = self.visit_expr(func)\n",
        "            self.builder_.update_func(global_var, updated_func)\n",
        "        \n",
        "        return self.builder_.get()\n",
        "\n",
        "\n",
        "def map_dense(bb, call):\n",
        "    x, w = call.args\n",
        "    return bb.call_te(topi.nn.dense, x, w)\n",
        "\n",
        "def map_add(bb, call):\n",
        "    a, b = call.args\n",
        "    return bb.call_te(topi.add, a, b)\n",
        "\n",
        "def map_relu(bb, call):\n",
        "    return bb.call_te(topi.nn.relu, call.args[0])\n",
        "\n",
        "\n",
        "op_map = {\n",
        "  \"relax.nn.dense\": map_dense,\n",
        "  \"relax.add\": map_add,\n",
        "  \"relax.nn.relu\": map_relu\n",
        "}\n",
        "\n",
        "@tvm.ir.transform.module_pass(opt_level=0, name=\"LowerToTensorIR\")\n",
        "class LowerToTensorIRPass:\n",
        "    \"\"\"The wrapper for the LowerTensorIR pass.\"\"\"\n",
        "    def transform_module(self, mod, ctx):\n",
        "        return LowerToTensorIR(mod, op_map).transform()\n",
        "\n",
        "\n",
        "MLPModelTIR = LowerToTensorIRPass()(MLPFused)\n",
        "MLPModelTIR.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "St-WgM3_BBgS"
      },
      "source": [
        "Note that in the above code. `fused_dense_add0` and `fused_dense_add1` still are high-level relax functions that calls into the corresponding TensorIR dense and add functions. We can turn them into a single TensorIR function, which then can be used for follow-up optimization and code generation phases."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 90,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "HDehRUH4Ay7H",
        "outputId": "6e35574c-98e6-4e7b-f909-c88ada13a575"
      },
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div class=\"highlight\" style=\"background: \"><pre style=\"line-height: 125%\"><span></span><span style=\"color: #AA22FF\">@tvm</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>script<span style=\"color: #AA22FF; font-weight: bold\">.</span>ir_module\n",
              "<span style=\"color: #008000; font-weight: bold\">class</span> <span style=\"color: #1E90FF; font-weight: bold\">Module</span>:\n",
              "    <span style=\"color: #AA22FF\">@T</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>prim_func\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">fused_dense_add0</span>(x: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">784</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], w: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">784</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], b: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], T_add: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>]) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> <span style=\"color: #008000; font-weight: bold\">None</span>:\n",
              "        <span style=\"color: #007979; font-style: italic\"># function attr dict</span>\n",
              "        T<span style=\"color: #AA22FF; font-weight: bold\">.</span>func_attr({<span style=\"color: #BA2121\">&quot;tir.noalias&quot;</span>: <span style=\"color: #008000; font-weight: bold\">True</span>, <span style=\"color: #BA2121\">&quot;global_symbol&quot;</span>: <span style=\"color: #BA2121\">&quot;fused_dense_add0&quot;</span>})\n",
              "        <span style=\"color: #007979; font-style: italic\"># body</span>\n",
              "        <span style=\"color: #007979; font-style: italic\"># with T.block(&quot;root&quot;)</span>\n",
              "        T_matmul_NT <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>alloc_buffer([<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)], dtype<span style=\"color: #AA22FF; font-weight: bold\">=</span><span style=\"color: #BA2121\">&quot;float32&quot;</span>)\n",
              "        <span style=\"color: #008000; font-weight: bold\">for</span> i0, i1, i2 <span style=\"color: #008000; font-weight: bold\">in</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>grid(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), <span style=\"color: #008000\">784</span>):\n",
              "            <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>block(<span style=\"color: #BA2121\">&quot;T_matmul_NT&quot;</span>):\n",
              "                i <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(<span style=\"color: #008000\">1</span>, i0)\n",
              "                j <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), i1)\n",
              "                k <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>reduce(<span style=\"color: #008000\">784</span>, i2)\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>reads(x[i, k], w[j, k])\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>writes(T_matmul_NT[i, j])\n",
              "                <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>init():\n",
              "                    T_matmul_NT[i, j] <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>float32(<span style=\"color: #008000\">0</span>)\n",
              "                T_matmul_NT[i, j] <span style=\"color: #AA22FF; font-weight: bold\">=</span> T_matmul_NT[i, j] <span style=\"color: #AA22FF; font-weight: bold\">+</span> x[i, k] <span style=\"color: #AA22FF; font-weight: bold\">*</span> w[j, k]\n",
              "        <span style=\"color: #008000; font-weight: bold\">for</span> i0, i1 <span style=\"color: #008000; font-weight: bold\">in</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>grid(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)):\n",
              "            <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>block(<span style=\"color: #BA2121\">&quot;T_add&quot;</span>):\n",
              "                ax0 <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(<span style=\"color: #008000\">1</span>, i0)\n",
              "                ax1 <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), i1)\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>reads(T_matmul_NT[ax0, ax1], b[ax1])\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>writes(T_add[ax0, ax1])\n",
              "                T_add[ax0, ax1] <span style=\"color: #AA22FF; font-weight: bold\">=</span> T_matmul_NT[ax0, ax1] <span style=\"color: #AA22FF; font-weight: bold\">+</span> b[ax1]\n",
              "    \n",
              "    <span style=\"color: #AA22FF\">@T</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>prim_func\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">relu</span>(rxplaceholder: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], compute: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>]) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> <span style=\"color: #008000; font-weight: bold\">None</span>:\n",
              "        <span style=\"color: #007979; font-style: italic\"># function attr dict</span>\n",
              "        T<span style=\"color: #AA22FF; font-weight: bold\">.</span>func_attr({<span style=\"color: #BA2121\">&quot;global_symbol&quot;</span>: <span style=\"color: #BA2121\">&quot;relu&quot;</span>, <span style=\"color: #BA2121\">&quot;tir.noalias&quot;</span>: <span style=\"color: #008000; font-weight: bold\">True</span>})\n",
              "        <span style=\"color: #007979; font-style: italic\"># body</span>\n",
              "        <span style=\"color: #007979; font-style: italic\"># with T.block(&quot;root&quot;)</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">for</span> i0, i1 <span style=\"color: #008000; font-weight: bold\">in</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>grid(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)):\n",
              "            <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>block(<span style=\"color: #BA2121\">&quot;compute&quot;</span>):\n",
              "                i0_1 <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(<span style=\"color: #008000\">1</span>, i0)\n",
              "                i1_1 <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), i1)\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>reads(rxplaceholder[i0_1, i1_1])\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>writes(compute[i0_1, i1_1])\n",
              "                compute[i0_1, i1_1] <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>max(rxplaceholder[i0_1, i1_1], T<span style=\"color: #AA22FF; font-weight: bold\">.</span>float32(<span style=\"color: #008000\">0</span>))\n",
              "    \n",
              "    <span style=\"color: #AA22FF\">@T</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>prim_func\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">fused_dense_add1</span>(x: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], w: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>), T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], b: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>], T_add: T<span style=\"color: #AA22FF; font-weight: bold\">.</span>Buffer[(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>)), <span style=\"color: #BA2121\">&quot;float32&quot;</span>]) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> <span style=\"color: #008000; font-weight: bold\">None</span>:\n",
              "        <span style=\"color: #007979; font-style: italic\"># function attr dict</span>\n",
              "        T<span style=\"color: #AA22FF; font-weight: bold\">.</span>func_attr({<span style=\"color: #BA2121\">&quot;tir.noalias&quot;</span>: <span style=\"color: #008000; font-weight: bold\">True</span>, <span style=\"color: #BA2121\">&quot;global_symbol&quot;</span>: <span style=\"color: #BA2121\">&quot;fused_dense_add1&quot;</span>})\n",
              "        <span style=\"color: #007979; font-style: italic\"># body</span>\n",
              "        <span style=\"color: #007979; font-style: italic\"># with T.block(&quot;root&quot;)</span>\n",
              "        T_matmul_NT <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>alloc_buffer([<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>)], dtype<span style=\"color: #AA22FF; font-weight: bold\">=</span><span style=\"color: #BA2121\">&quot;float32&quot;</span>)\n",
              "        <span style=\"color: #008000; font-weight: bold\">for</span> i0, i1, i2 <span style=\"color: #008000; font-weight: bold\">in</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>grid(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>), T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>)):\n",
              "            <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>block(<span style=\"color: #BA2121\">&quot;T_matmul_NT&quot;</span>):\n",
              "                i <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(<span style=\"color: #008000\">1</span>, i0)\n",
              "                j <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>), i1)\n",
              "                k <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>reduce(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">128</span>), i2)\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>reads(x[i, k], w[j, k])\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>writes(T_matmul_NT[i, j])\n",
              "                <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>init():\n",
              "                    T_matmul_NT[i, j] <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>float32(<span style=\"color: #008000\">0</span>)\n",
              "                T_matmul_NT[i, j] <span style=\"color: #AA22FF; font-weight: bold\">=</span> T_matmul_NT[i, j] <span style=\"color: #AA22FF; font-weight: bold\">+</span> x[i, k] <span style=\"color: #AA22FF; font-weight: bold\">*</span> w[j, k]\n",
              "        <span style=\"color: #008000; font-weight: bold\">for</span> i0, i1 <span style=\"color: #008000; font-weight: bold\">in</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>grid(<span style=\"color: #008000\">1</span>, T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>)):\n",
              "            <span style=\"color: #008000; font-weight: bold\">with</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>block(<span style=\"color: #BA2121\">&quot;T_add&quot;</span>):\n",
              "                ax0 <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(<span style=\"color: #008000\">1</span>, i0)\n",
              "                ax1 <span style=\"color: #AA22FF; font-weight: bold\">=</span> T<span style=\"color: #AA22FF; font-weight: bold\">.</span>axis<span style=\"color: #AA22FF; font-weight: bold\">.</span>spatial(T<span style=\"color: #AA22FF; font-weight: bold\">.</span>int64(<span style=\"color: #008000\">10</span>), i1)\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>reads(T_matmul_NT[ax0, ax1], b[ax1])\n",
              "                T<span style=\"color: #AA22FF; font-weight: bold\">.</span>writes(T_add[ax0, ax1])\n",
              "                T_add[ax0, ax1] <span style=\"color: #AA22FF; font-weight: bold\">=</span> T_matmul_NT[ax0, ax1] <span style=\"color: #AA22FF; font-weight: bold\">+</span> b[ax1]\n",
              "    \n",
              "    <span style=\"color: #AA22FF\">@R</span><span style=\"color: #AA22FF; font-weight: bold\">.</span>function\n",
              "    <span style=\"color: #008000; font-weight: bold\">def</span> <span style=\"color: #1E90FF\">main</span>(x: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">784</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>)) <span style=\"color: #AA22FF; font-weight: bold\">-&gt;</span> Tensor(<span style=\"color: #008000; font-weight: bold\">None</span>, <span style=\"color: #BA2121\">&quot;float32&quot;</span>, ndim <span style=\"color: #AA22FF; font-weight: bold\">=</span> <span style=\"color: #008000\">2</span>):\n",
              "        <span style=\"color: #007979; font-style: italic\"># block 0</span>\n",
              "        <span style=\"color: #008000; font-weight: bold\">with</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>dataflow():\n",
              "            lv1 <span style=\"color: #AA22FF; font-weight: bold\">=</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>call_tir(fused_dense_add0, (x, meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">0</span>], meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">1</span>]), (<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), dtype<span style=\"color: #AA22FF; font-weight: bold\">=</span><span style=\"color: #BA2121\">&quot;float32&quot;</span>)\n",
              "            lv2 <span style=\"color: #AA22FF; font-weight: bold\">=</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>call_tir(relu, (lv1,), (<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">128</span>), dtype<span style=\"color: #AA22FF; font-weight: bold\">=</span><span style=\"color: #BA2121\">&quot;float32&quot;</span>)\n",
              "            lv4 <span style=\"color: #AA22FF; font-weight: bold\">=</span> R<span style=\"color: #AA22FF; font-weight: bold\">.</span>call_tir(fused_dense_add1, (lv2, meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">2</span>], meta[relay<span style=\"color: #AA22FF; font-weight: bold\">.</span>Constant][<span style=\"color: #008000\">3</span>]), (<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), dtype<span style=\"color: #AA22FF; font-weight: bold\">=</span><span style=\"color: #BA2121\">&quot;float32&quot;</span>)\n",
              "            gv: Tensor((<span style=\"color: #008000\">1</span>, <span style=\"color: #008000\">10</span>), <span style=\"color: #BA2121\">&quot;float32&quot;</span>) <span style=\"color: #AA22FF; font-weight: bold\">=</span> lv4\n",
              "            R<span style=\"color: #AA22FF; font-weight: bold\">.</span>output(gv)\n",
              "        <span style=\"color: #008000; font-weight: bold\">return</span> gv\n",
              "    \n",
              "</pre></div>\n"
            ],
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "MLPModelFinal = relax.transform.FuseTIR()(MLPModelTIR)\n",
        "MLPModelFinal.show()"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "c9NR2_yzrseJ"
      },
      "source": [
        "## Build and Run"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "wfC4YioXBeaC"
      },
      "source": [
        "We can go ahead and build the final module and try it out on an example picture."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 91,
      "metadata": {
        "id": "3VYApwmKruio"
      },
      "outputs": [],
      "source": [
        "import torch\n",
        "import torchvision\n",
        "\n",
        "test_data = torchvision.datasets.FashionMNIST(\n",
        "    root=\"data\",\n",
        "    train=False,\n",
        "    download=True,\n",
        "    transform=torchvision.transforms.ToTensor()\n",
        ")\n",
        "test_loader = torch.utils.data.DataLoader(test_data, batch_size=1, shuffle=True)\n",
        "class_names = ['T-shirt/top', 'Trouser', 'Pullover', 'Dress', 'Coat',\n",
        "               'Sandal', 'Shirt', 'Sneaker', 'Bag', 'Ankle boot']\n",
        "\n",
        "img, label = next(iter(test_loader))\n",
        "img = img.reshape(1, 28, 28).numpy()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 92,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 287
        },
        "id": "AzdKeWBUrzFo",
        "outputId": "a25a074c-3092-4b21-bc30-c45440e51fd5"
      },
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 432x288 with 2 Axes>"
            ]
          },
          "metadata": {
            "needs_background": "light"
          },
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Class: Sneaker\n"
          ]
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "plt.figure()\n",
        "plt.imshow(img[0])\n",
        "plt.colorbar()\n",
        "plt.grid(False)\n",
        "plt.show()\n",
        "\n",
        "print(\"Class:\", class_names[label[0]])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 93,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Z9LxKCoPfLbI",
        "outputId": "40099c3b-0954-479b-d128-808bb5ae6a03"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "MLPModule Prediction: Sneaker\n"
          ]
        }
      ],
      "source": [
        "ex = relax.vm.build(MLPModelFinal, target=\"llvm\")\n",
        "vm = relax.VirtualMachine(ex, tvm.cpu())\n",
        "data_nd = tvm.nd.array(img.reshape(1, 784))\n",
        "\n",
        "nd_res = vm[\"main\"](data_nd)\n",
        "\n",
        "pred_kind = np.argmax(nd_res.numpy(), axis=1)\n",
        "print(\"MLPModule Prediction:\", class_names[pred_kind[0]])"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "pWiGeyAaBmAL"
      },
      "source": [
        "## Discussions"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "hOuaPK7zB3Ny"
      },
      "source": [
        "This section comes back to our common theme of **transformation** among computational graphs. Despite being minimum, this sequence of transformations covers two important optimizations we commonly do in MLC process -- fusion and loop level code lowering.\n",
        "\n",
        "Real-world MLC process can contain more powerful and robust transformations. For example, our fusion pass can create duplicated dense computations in which a dense operator is referenced in two follow-ups add operations. A robust fusion pass will detect that and choose to skip such cases. Additionally, we do not want to have to write down rules for each combination. Instead, TVM's internal fusor will analyze the TensorIR function loop patterns and use them in fusion decisions.\n",
        "\n",
        "Notably, each of these transformations is composable with each other. For example, we can choose to use our version of customized fusor to support additional new fusion patterns that we want to explore and then feed into an existing fusor to handle the rest of the steps.\n"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "XeE0HR7tBoOi"
      },
      "source": [
        "![image.png](data:image/png;base64,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)"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "metadata": {
        "id": "mIQMuYaJ7sbJ"
      },
      "source": [
        "## Summary\n",
        "\n",
        "- We can optimize tensor programs by rewriting computational graph data structures.\n",
        "- Visitor pattern to rewrite call nodes.\n",
        "- We can perform computational graph transformations, such as fusion and loop-level program lowering.\n"
      ]
    }
  ],
  "metadata": {
    "colab": {
      "collapsed_sections": [],
      "include_colab_link": true,
      "name": "9 Computational Graph Optimization.ipynb",
      "provenance": []
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.16"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}