LLMs-from-scratch/ch03/02_bonus_efficient-multihead-attention/mha-implementations.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "6f678e62-7bcb-4405-86ae-dce94f494303",
      "metadata": {
        "id": "6f678e62-7bcb-4405-86ae-dce94f494303"
      },
      "source": [
        "# Efficient Multi-Head Attention Implementations"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "2f9bb1b6-a1e5-4e0a-884d-0f31b374a8d6",
      "metadata": {
        "id": "2f9bb1b6-a1e5-4e0a-884d-0f31b374a8d6"
      },
      "source": [
        "## Multi-head attention implementations from chapter 3"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "id": "7898551e-f582-48ac-9f66-3632abe2a93f",
      "metadata": {
        "id": "7898551e-f582-48ac-9f66-3632abe2a93f",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "840126fe-fffa-46d4-9717-41aef89d5052"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Running on cuda\n"
          ]
        }
      ],
      "source": [
        "import torch\n",
        "\n",
        "torch.manual_seed(123)\n",
        "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
        "print(f\"Running on {device}\")\n",
        "\n",
        "batch_size = 8\n",
        "context_len = 1024\n",
        "embed_dim = 768\n",
        "embeddings = torch.randn((batch_size, context_len, embed_dim), device=device)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "id": "297c93ed-aec0-4896-bb89-42c4b294d3d1",
      "metadata": {
        "id": "297c93ed-aec0-4896-bb89-42c4b294d3d1",
        "outputId": "5af9d36b-37c9-4f6e-c370-58a46db02632",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "torch.Size([8, 1024, 9216])\n"
          ]
        }
      ],
      "source": [
        "from ch03 import MultiHeadAttentionWrapper as Ch03_MHA_Wrapper\n",
        "\n",
        "mha_ch03_wrapper = Ch03_MHA_Wrapper(\n",
        "    d_in=embed_dim,\n",
        "    d_out=embed_dim,\n",
        "    block_size=context_len,\n",
        "    dropout=0.0,\n",
        "    num_heads=12,\n",
        "    qkv_bias=False\n",
        ").to(device)\n",
        "\n",
        "out = mha_ch03_wrapper(embeddings)\n",
        "print(out.shape)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "id": "4ee6a61b-d25c-4a0c-8a59-f285544e3710",
      "metadata": {
        "id": "4ee6a61b-d25c-4a0c-8a59-f285544e3710",
        "outputId": "1c7ffc71-3b51-4ee8-beab-261625b1473e",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "torch.Size([8, 1024, 768])\n"
          ]
        }
      ],
      "source": [
        "from ch03 import MultiHeadAttention as Ch03_MHA\n",
        "\n",
        "mha_ch03 = Ch03_MHA(\n",
        "    d_in=embed_dim,\n",
        "    d_out=embed_dim,\n",
        "    block_size=context_len,\n",
        "    dropout=0.0,\n",
        "    num_heads=12,\n",
        "    qkv_bias=False\n",
        ").to(device)\n",
        "\n",
        "out = mha_ch03(embeddings)\n",
        "print(out.shape)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "73cd11da-ea3b-4081-b483-c4965dfefbc4",
      "metadata": {
        "id": "73cd11da-ea3b-4081-b483-c4965dfefbc4"
      },
      "source": [
        "## An alternative multi-head attention with combined weights"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "1fa1a5ea-eaff-4d2d-aaf0-b34cdb6fd4dd",
      "metadata": {
        "id": "1fa1a5ea-eaff-4d2d-aaf0-b34cdb6fd4dd"
      },
      "source": [
        "- The code for the `MultiHeadAttentionAlt` class below is based on code that was kindly shared by [Rayed Bin Wahed](https://github.com/rasbt/LLMs-from-scratch/discussions/51)\n",
        "- The main difference between the `MultiHeadAttentionAlt` class and the `MultiHeadAttention` class used in chapter 3 is that `MultiHeadAttentionAlt` uses a single weight matrix, `self.qkv = nn.Linear(d_in, 3 * d_out, bias=qkv_bias)` instead of separate weight matrices:\n",
        "\n",
        "  - `self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)`\n",
        "  - `self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)`\n",
        "  - `self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)`\n",
        "\n",
        "- Here, `self.qkv` combines all three weight matrices `self.W_query`, `self.W_key`, and `self.W_value` to carry out the query, key, and value computation in a single step\n",
        "- Using `q, k, v = qkv.unbind(0)`, we obtain the individual query, key, and value tensors, which are then used similarly to the query, key, and value tensors in the `MultiHeadAttention` class in chapter 3"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "id": "9a6bd0a2-f27c-4602-afa0-c96cd295c1a6",
      "metadata": {
        "id": "9a6bd0a2-f27c-4602-afa0-c96cd295c1a6",
        "outputId": "3c225fe5-73a9-4df0-c513-6296f4bb5261",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "torch.Size([8, 1024, 768])\n"
          ]
        }
      ],
      "source": [
        "import torch.nn as nn\n",
        "\n",
        "\n",
        "class MultiHeadAttentionCombinedQKV(nn.Module):\n",
        "    def __init__(self, d_in, d_out, num_heads, block_size, dropout=0.0, qkv_bias=False):\n",
        "        super().__init__()\n",
        "\n",
        "        assert d_out % num_heads == 0, \"embed_dim is indivisible by num_heads\"\n",
        "\n",
        "        self.num_heads = num_heads\n",
        "        self.block_size = block_size\n",
        "        self.head_dim = d_out // num_heads\n",
        "\n",
        "        self.qkv = nn.Linear(d_in, 3 * d_out, bias=qkv_bias)\n",
        "        self.proj = nn.Linear(d_in, d_out)\n",
        "        self.dropout = nn.Dropout(dropout)\n",
        "\n",
        "        self.register_buffer(\n",
        "            \"mask\", torch.triu(torch.ones(block_size, block_size), diagonal=1)\n",
        "        )\n",
        "\n",
        "    def forward(self, x):\n",
        "        batch_size, num_tokens, embed_dim = x.shape\n",
        "\n",
        "        # (b, num_tokens, embed_dim) --> (b, num_tokens, 3 * embed_dim)\n",
        "        qkv = self.qkv(x)\n",
        "\n",
        "        # (b, num_tokens, 3 * embed_dim) --> (b, num_tokens, 3, num_heads, head_dim)\n",
        "        qkv = qkv.reshape(batch_size, num_tokens, 3, self.num_heads, self.head_dim)\n",
        "\n",
        "        # (b, num_tokens, 3, num_heads, head_dim) --> (3, b, num_heads, num_tokens, head_dim)\n",
        "        qkv = qkv.permute(2, 0, 3, 1, 4)\n",
        "\n",
        "        # (3, b, num_heads, num_tokens, head_dim) -> 3 times (b, num_head, num_tokens, head_dim)\n",
        "        queries, keys, values = qkv.unbind(0)\n",
        "\n",
        "        # (b, num_heads, num_tokens, head_dim) --> (b, num_heads, num_tokens, num_tokens)\n",
        "        attn_scores = queries @ keys.transpose(-2, -1)\n",
        "        attn_scores = attn_scores.masked_fill(\n",
        "            self.mask.bool()[:num_tokens, :num_tokens], -torch.inf\n",
        "        )\n",
        "\n",
        "        attn_weights = torch.softmax(attn_scores / keys.shape[-1]**-0.5, dim=-1)\n",
        "        attn_weights = self.dropout(attn_weights)\n",
        "\n",
        "        # (b, num_heads, num_tokens, num_tokens) --> (b, num_heads, num_tokens, head_dim)\n",
        "        context_vec = attn_weights @ values\n",
        "\n",
        "        # (b, num_heads, num_tokens, head_dim) --> (b, num_tokens, num_heads, head_dim)\n",
        "        context_vec = context_vec.transpose(1, 2)\n",
        "\n",
        "        # (b, num_tokens, num_heads, head_dim) --> (b, num_tokens, embed_dim)\n",
        "        context_vec = context_vec.reshape(batch_size, num_tokens, embed_dim)\n",
        "\n",
        "        context_vec = self.proj(context_vec)\n",
        "\n",
        "        return context_vec\n",
        "\n",
        "\n",
        "mha_combined_qkv = MultiHeadAttentionCombinedQKV(\n",
        "    d_in=embed_dim,\n",
        "    d_out=embed_dim,\n",
        "    block_size=context_len,\n",
        "    dropout=0.0,\n",
        "    num_heads=12,\n",
        "    qkv_bias=False\n",
        ").to(device)\n",
        "\n",
        "out = mha_combined_qkv(embeddings)\n",
        "print(out.shape)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "48a042d3-ee78-4c29-bf63-d92fe6706632",
      "metadata": {
        "id": "48a042d3-ee78-4c29-bf63-d92fe6706632"
      },
      "source": [
        "## Multihead attention with PyTorch's scaled dot product attention"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "f78e346f-3b85-44e6-9feb-f01131381148",
      "metadata": {
        "id": "f78e346f-3b85-44e6-9feb-f01131381148"
      },
      "source": [
        "- The implementation below uses PyTorch's [`scaled_dot_product_attention`](https://pytorch.org/docs/stable/generated/torch.nn.functional.scaled_dot_product_attention.html) function, which implements a memory-optimized version of self-attention calld [flash attention](https://arxiv.org/abs/2205.14135)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "id": "1b8e5a0d-1f65-4a03-bf6e-723f0cc428f5",
      "metadata": {
        "id": "1b8e5a0d-1f65-4a03-bf6e-723f0cc428f5"
      },
      "outputs": [],
      "source": [
        "class MultiHeadAttentionPyTorch(nn.Module):\n",
        "    def __init__(self, d_in, d_out, num_heads, block_size, dropout=0.0, qkv_bias=False):\n",
        "        super().__init__()\n",
        "\n",
        "        assert d_out % num_heads == 0, \"embed_dim is indivisible by num_heads\"\n",
        "\n",
        "        self.num_heads = num_heads\n",
        "        self.block_size = block_size\n",
        "        self.head_dim = d_out // num_heads\n",
        "        self.d_out = d_out\n",
        "\n",
        "        self.qkv = nn.Linear(d_in, 3 * d_out, bias=qkv_bias)\n",
        "        self.proj = nn.Linear(d_in, d_out)\n",
        "        self.dropout = dropout\n",
        "\n",
        "        self.register_buffer(\n",
        "            \"mask\", torch.triu(torch.ones(block_size, block_size), diagonal=1)\n",
        "        )\n",
        "\n",
        "    def forward(self, x):\n",
        "        batch_size, num_tokens, embed_dim = x.shape\n",
        "\n",
        "        # (b, num_tokens, embed_dim) --> (b, num_tokens, 3 * embed_dim)\n",
        "        qkv = self.qkv(x)\n",
        "\n",
        "        # (b, num_tokens, 3 * embed_dim) --> (b, num_tokens, 3, num_heads, head_dim)\n",
        "        qkv = qkv.reshape(batch_size, num_tokens, 3, self.num_heads, self.head_dim)\n",
        "\n",
        "        # (b, num_tokens, 3, num_heads, head_dim) --> (3, b, num_heads, num_tokens, head_dim)\n",
        "        qkv = qkv.permute(2, 0, 3, 1, 4)\n",
        "\n",
        "        # (3, b, num_heads, num_tokens, head_dim) -> 3 times (b, num_heads, num_tokens, head_dim)\n",
        "        queries, keys, values = qkv.unbind(0)\n",
        "\n",
        "        use_dropout = 0. if not self.training else self.dropout\n",
        "        context_vec = nn.functional.scaled_dot_product_attention(\n",
        "            queries, keys, values, attn_mask=None, dropout_p=use_dropout, is_causal=True)\n",
        "\n",
        "        # Combine heads, where self.d_out = self.num_heads * self.head_dim\n",
        "        context_vec = context_vec.transpose(1, 2).contiguous().view(batch_size, num_tokens, self.d_out)\n",
        "\n",
        "        return context_vec"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "id": "fbc8ba92-3471-41cb-b1b2-4c0ef5be392b",
      "metadata": {
        "id": "fbc8ba92-3471-41cb-b1b2-4c0ef5be392b",
        "outputId": "f3e7933d-16d3-45e5-f03d-610319004579",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "torch.Size([8, 1024, 768])\n"
          ]
        }
      ],
      "source": [
        "mha_pytorch = MultiHeadAttentionPyTorch(\n",
        "    d_in=embed_dim,\n",
        "    d_out=embed_dim,\n",
        "    block_size=context_len,\n",
        "    dropout=0.0,\n",
        "    num_heads=12,\n",
        "    qkv_bias=False\n",
        ").to(device)\n",
        "\n",
        "out = mha_pytorch(embeddings)\n",
        "print(out.shape)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "8877de71-f84f-4f6d-bc87-7552013b6301",
      "metadata": {
        "id": "8877de71-f84f-4f6d-bc87-7552013b6301"
      },
      "source": [
        "## Speed comparison"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "id": "a97c0b2e-6593-49d8-98bc-2267b3aa610f",
      "metadata": {
        "id": "a97c0b2e-6593-49d8-98bc-2267b3aa610f",
        "outputId": "bb928da8-6ac0-4d15-cf12-4903d73708fc",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "41.1 ms ± 9.08 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
          ]
        }
      ],
      "source": [
        "%timeit mha_ch03_wrapper(embeddings)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "id": "19db9c2c-8e75-431a-8eef-0b4d8284e6e6",
      "metadata": {
        "id": "19db9c2c-8e75-431a-8eef-0b4d8284e6e6",
        "outputId": "54f8e05e-0cb2-4e4a-cacd-27a309a3be8b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "6.58 ms ± 582 ns per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
          ]
        }
      ],
      "source": [
        "%timeit mha_ch03(embeddings)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "id": "aa526ee0-7a88-4f34-a49a-f8f97da83779",
      "metadata": {
        "id": "aa526ee0-7a88-4f34-a49a-f8f97da83779",
        "outputId": "415e959e-b648-4f1e-f05e-8b8444e74bee",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "7.2 ms ± 327 ns per loop (mean ± std. dev. of 7 runs, 100 loops each)\n"
          ]
        }
      ],
      "source": [
        "%timeit mha_combined_qkv(embeddings)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "id": "cc2b4256-16d8-4c34-9fd0-d4b4af0e60fa",
      "metadata": {
        "id": "cc2b4256-16d8-4c34-9fd0-d4b4af0e60fa",
        "outputId": "05b7c696-1b97-4f18-8430-481bb8940b6b",
        "colab": {
          "base_uri": "https://localhost:8080/"
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2.38 ms ± 386 ns per loop (mean ± std. dev. of 7 runs, 1000 loops each)\n"
          ]
        }
      ],
      "source": [
        "%timeit mha_pytorch(embeddings)"
      ]
    }
  ],
  "metadata": {
    "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": "3.10.12"
    },
    "colab": {
      "provenance": [],
      "machine_shape": "hm",
      "gpuType": "A100"
    },
    "accelerator": "GPU"
  },
  "nbformat": 4,
  "nbformat_minor": 5
}