Calculating Cosine Similarity with Different Implementations

TensorPrimitivesCosineSimSample.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Consine Similarity Samples\n",
    "\n",
    "Different examples showing how to compute cosine similarity using TorchSharp, a custom implementation, and tensor primitives."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "var movies = new [] {\n",
    "    new {Title=\"The Lion King\", Embedding= new [] {0.10022575f, -0.23998135f}},\n",
    "    new {Title=\"Inception\", Embedding= new [] {0.10327095f, 0.2563685f}},\n",
    "    new {Title=\"Toy Story\", Embedding= new [] {0.095857024f, -0.201278f}},\n",
    "    new {Title=\"Pulp Function\", Embedding= new [] {0.106827796f, 0.21676421f}},\n",
    "    new {Title=\"Shrek\", Embedding= new [] {0.09568083f, -0.21177962f}}\n",
    "};"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "// \"A movie that's fun for the whole family\" Embedding\n",
    "var queryEmbedding = new[] {0.12217915f, -0.034832448f };"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## TorchSharp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div></div><div></div><div><strong>Installed Packages</strong><ul><li><span>TorchSharp-cpu, 0.100.5</span></li></ul></div></div>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "Loading extensions from `C:\\Users\\luquinta\\.nuget\\packages\\skiasharp\\2.88.3\\interactive-extensions\\dotnet\\SkiaSharp.DotNet.Interactive.dll`"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#r \"nuget: TorchSharp-cpu, 0.100.5\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "using TorchSharp;\n",
    "using static TorchSharp.torch.nn.functional;"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "var top3MoviesTorchSharp = \n",
    "    movies\n",
    "        .Select(movie => \n",
    "            new {\n",
    "                Title=movie.Title, \n",
    "                Embedding=movie.Embedding, \n",
    "                Similarity=cosine_similarity(torch.tensor(queryEmbedding),torch.tensor(movie.Embedding),0L).item<float>()})\n",
    "        .OrderByDescending(movies => movies.Similarity)\n",
    "        .Take(3);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<details open=\"open\" class=\"dni-treeview\"><summary><span class=\"dni-code-hint\"><code>[ { Title = Toy Story, Embedding = System.Single[], Similarity = 0.66102695 }, { Title = Shrek, Embedding = System.Single[], Similarity = 0.6457999 }, { Title = The Lion King, Embedding = System.Single[], Similarity = 0.62360466 } ]</code></span></summary><div><table><thead><tr></tr></thead><tbody><tr><td><i>(values)</i></td><td><table><thead><tr><th><i>index</i></th><th>value</th></tr></thead><tbody><tr><td>0</td><td><details class=\"dni-treeview\"><summary><span class=\"dni-code-hint\"><code>{ Title = Toy Story, Embedding = System.Single[], Similarity = 0.66102695 }</code></span></summary><div><table><thead><tr></tr></thead><tbody><tr><td>Title</td><td><div class=\"dni-plaintext\"><pre>Toy Story</pre></div></td></tr><tr><td>Embedding</td><td><div class=\"dni-plaintext\"><pre>[ 0.095857024, -0.201278 ]</pre></div></td></tr><tr><td>Similarity</td><td><div class=\"dni-plaintext\"><pre>0.66102695</pre></div></td></tr></tbody></table></div></details></td></tr><tr><td>1</td><td><details class=\"dni-treeview\"><summary><span class=\"dni-code-hint\"><code>{ Title = Shrek, Embedding = System.Single[], Similarity = 0.6457999 }</code></span></summary><div><table><thead><tr></tr></thead><tbody><tr><td>Title</td><td><div class=\"dni-plaintext\"><pre>Shrek</pre></div></td></tr><tr><td>Embedding</td><td><div class=\"dni-plaintext\"><pre>[ 0.09568083, -0.21177962 ]</pre></div></td></tr><tr><td>Similarity</td><td><div class=\"dni-plaintext\"><pre>0.6457999</pre></div></td></tr></tbody></table></div></details></td></tr><tr><td>2</td><td><details class=\"dni-treeview\"><summary><span class=\"dni-code-hint\"><code>{ Title = The Lion King, Embedding = System.Single[], Similarity = 0.62360466 }</code></span></summary><div><table><thead><tr></tr></thead><tbody><tr><td>Title</td><td><div class=\"dni-plaintext\"><pre>The Lion King</pre></div></td></tr><tr><td>Embedding</td><td><div class=\"dni-plaintext\"><pre>[ 0.10022575, -0.23998135 ]</pre></div></td></tr><tr><td>Similarity</td><td><div class=\"dni-plaintext\"><pre>0.62360466</pre></div></td></tr></tbody></table></div></details></td></tr></tbody></table></td></tr></tbody></table></div></details><style>\r\n",
       ".dni-code-hint {\r\n",
       "    font-style: italic;\r\n",
       "    overflow: hidden;\r\n",
       "    white-space: nowrap;\r\n",
       "}\r\n",
       ".dni-treeview {\r\n",
       "    white-space: nowrap;\r\n",
       "}\r\n",
       ".dni-treeview td {\r\n",
       "    vertical-align: top;\r\n",
       "    text-align: start;\r\n",
       "}\r\n",
       "details.dni-treeview {\r\n",
       "    padding-left: 1em;\r\n",
       "}\r\n",
       "table td {\r\n",
       "    text-align: start;\r\n",
       "}\r\n",
       "table tr { \r\n",
       "    vertical-align: top; \r\n",
       "    margin: 0em 0px;\r\n",
       "}\r\n",
       "table tr td pre \r\n",
       "{ \r\n",
       "    vertical-align: top !important; \r\n",
       "    margin: 0em 0px !important;\r\n",
       "} \r\n",
       "table th {\r\n",
       "    text-align: start;\r\n",
       "}\r\n",
       "</style>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "top3MoviesTorchSharp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Custom Implementation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "public float CosineSimilarityCustom(ReadOnlySpan<float> vector1, ReadOnlySpan<float> vector2)\n",
    "{\n",
    "    if (vector1.Length != vector2.Length)\n",
    "        throw new ArgumentException(\"Vectors must have the same length\");\n",
    "\n",
    "    float dotProduct = 0f;\n",
    "    float magnitude1 = 0f;\n",
    "    float magnitude2 = 0f;\n",
    "\n",
    "    for (int i = 0; i < vector1.Length; i++)\n",
    "    {\n",
    "        dotProduct += vector1[i] * vector2[i];\n",
    "        magnitude1 += vector1[i] * vector1[i];\n",
    "        magnitude2 += vector2[i] * vector2[i];\n",
    "    }\n",
    "\n",
    "    magnitude1 = MathF.Sqrt(magnitude1);\n",
    "    magnitude2 = MathF.Sqrt(magnitude2);\n",
    "\n",
    "    if (magnitude1 == 0 || magnitude2 == 0)\n",
    "        return 0;  // handle the case where one or both vectors have zero magnitude\n",
    "\n",
    "    return dotProduct / (magnitude1 * magnitude2);\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "var top3MoviesNaive = \n",
    "    movies\n",
    "        .Select(movie => \n",
    "            new {\n",
    "                Title=movie.Title, \n",
    "                Embedding=movie.Embedding, \n",
    "                Similarity=CosineSimilarityCustom(queryEmbedding.AsSpan(),movie.Embedding.AsSpan())})\n",
    "        .OrderByDescending(movies => movies.Similarity)\n",
    "        .Take(3);\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<details open=\"open\" class=\"dni-treeview\"><summary><span class=\"dni-code-hint\"><code>[ { Title = Toy Story, Embedding = System.Single[], Similarity = 0.66102695 }, { Title = Shrek, Embedding = System.Single[], Similarity = 0.64579993 }, { Title = The Lion King, Embedding = System.Single[], Similarity = 0.6236047 } ]</code></span></summary><div><table><thead><tr></tr></thead><tbody><tr><td><i>(values)</i></td><td><table><thead><tr><th><i>index</i></th><th>value</th></tr></thead><tbody><tr><td>0</td><td><details class=\"dni-treeview\"><summary><span class=\"dni-code-hint\"><code>{ Title = Toy Story, Embedding = System.Single[], Similarity = 0.66102695 }</code></span></summary><div><table><thead><tr></tr></thead><tbody><tr><td>Title</td><td><div class=\"dni-plaintext\"><pre>Toy Story</pre></div></td></tr><tr><td>Embedding</td><td><div class=\"dni-plaintext\"><pre>[ 0.095857024, -0.201278 ]</pre></div></td></tr><tr><td>Similarity</td><td><div class=\"dni-plaintext\"><pre>0.66102695</pre></div></td></tr></tbody></table></div></details></td></tr><tr><td>1</td><td><details class=\"dni-treeview\"><summary><span class=\"dni-code-hint\"><code>{ Title = Shrek, Embedding = System.Single[], Similarity = 0.64579993 }</code></span></summary><div><table><thead><tr></tr></thead><tbody><tr><td>Title</td><td><div class=\"dni-plaintext\"><pre>Shrek</pre></div></td></tr><tr><td>Embedding</td><td><div class=\"dni-plaintext\"><pre>[ 0.09568083, -0.21177962 ]</pre></div></td></tr><tr><td>Similarity</td><td><div class=\"dni-plaintext\"><pre>0.64579993</pre></div></td></tr></tbody></table></div></details></td></tr><tr><td>2</td><td><details class=\"dni-treeview\"><summary><span class=\"dni-code-hint\"><code>{ Title = The Lion King, Embedding = System.Single[], Similarity = 0.6236047 }</code></span></summary><div><table><thead><tr></tr></thead><tbody><tr><td>Title</td><td><div class=\"dni-plaintext\"><pre>The Lion King</pre></div></td></tr><tr><td>Embedding</td><td><div class=\"dni-plaintext\"><pre>[ 0.10022575, -0.23998135 ]</pre></div></td></tr><tr><td>Similarity</td><td><div class=\"dni-plaintext\"><pre>0.6236047</pre></div></td></tr></tbody></table></div></details></td></tr></tbody></table></td></tr></tbody></table></div></details><style>\r\n",
       ".dni-code-hint {\r\n",
       "    font-style: italic;\r\n",
       "    overflow: hidden;\r\n",
       "    white-space: nowrap;\r\n",
       "}\r\n",
       ".dni-treeview {\r\n",
       "    white-space: nowrap;\r\n",
       "}\r\n",
       ".dni-treeview td {\r\n",
       "    vertical-align: top;\r\n",
       "    text-align: start;\r\n",
       "}\r\n",
       "details.dni-treeview {\r\n",
       "    padding-left: 1em;\r\n",
       "}\r\n",
       "table td {\r\n",
       "    text-align: start;\r\n",
       "}\r\n",
       "table tr { \r\n",
       "    vertical-align: top; \r\n",
       "    margin: 0em 0px;\r\n",
       "}\r\n",
       "table tr td pre \r\n",
       "{ \r\n",
       "    vertical-align: top !important; \r\n",
       "    margin: 0em 0px !important;\r\n",
       "} \r\n",
       "table th {\r\n",
       "    text-align: start;\r\n",
       "}\r\n",
       "</style>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "top3MoviesNaive"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Tensor Primitives"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "#r \"C:\\Users\\luquinta\\Downloads\\System.Numerics.Tensors\\netstandard\\System.Numerics.Tensors.dll\"\n",
    "#r \"C:\\Users\\luquinta\\Downloads\\Microsoft.Bcl.Numerics\\Microsoft.Bcl.Numerics.dll\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "using System.Numerics.Tensors;"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [],
   "source": [
    "var top3MoviesNaive = \n",
    "    movies\n",
    "        .Select(movie => \n",
    "            new {\n",
    "                Title=movie.Title, \n",
    "                Embedding=movie.Embedding, \n",
    "                Similarity=TensorPrimitives.CosineSimilarity(queryEmbedding.AsSpan(),movie.Embedding.AsSpan())})\n",
    "        .OrderByDescending(movies => movies.Similarity)\n",
    "        .Take(3);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "dotnet_interactive": {
     "language": "csharp"
    },
    "polyglot_notebook": {
     "kernelName": "csharp"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<details open=\"open\" class=\"dni-treeview\"><summary><span class=\"dni-code-hint\"><code>[ { Title = Toy Story, Embedding = System.Single[], Similarity = 0.66102695 }, { Title = Shrek, Embedding = System.Single[], Similarity = 0.64579993 }, { Title = The Lion King, Embedding = System.Single[], Similarity = 0.6236047 } ]</code></span></summary><div><table><thead><tr></tr></thead><tbody><tr><td><i>(values)</i></td><td><table><thead><tr><th><i>index</i></th><th>value</th></tr></thead><tbody><tr><td>0</td><td><details class=\"dni-treeview\"><summary><span class=\"dni-code-hint\"><code>{ Title = Toy Story, Embedding = System.Single[], Similarity = 0.66102695 }</code></span></summary><div><table><thead><tr></tr></thead><tbody><tr><td>Title</td><td><div class=\"dni-plaintext\"><pre>Toy Story</pre></div></td></tr><tr><td>Embedding</td><td><div class=\"dni-plaintext\"><pre>[ 0.095857024, -0.201278 ]</pre></div></td></tr><tr><td>Similarity</td><td><div class=\"dni-plaintext\"><pre>0.66102695</pre></div></td></tr></tbody></table></div></details></td></tr><tr><td>1</td><td><details class=\"dni-treeview\"><summary><span class=\"dni-code-hint\"><code>{ Title = Shrek, Embedding = System.Single[], Similarity = 0.64579993 }</code></span></summary><div><table><thead><tr></tr></thead><tbody><tr><td>Title</td><td><div class=\"dni-plaintext\"><pre>Shrek</pre></div></td></tr><tr><td>Embedding</td><td><div class=\"dni-plaintext\"><pre>[ 0.09568083, -0.21177962 ]</pre></div></td></tr><tr><td>Similarity</td><td><div class=\"dni-plaintext\"><pre>0.64579993</pre></div></td></tr></tbody></table></div></details></td></tr><tr><td>2</td><td><details class=\"dni-treeview\"><summary><span class=\"dni-code-hint\"><code>{ Title = The Lion King, Embedding = System.Single[], Similarity = 0.6236047 }</code></span></summary><div><table><thead><tr></tr></thead><tbody><tr><td>Title</td><td><div class=\"dni-plaintext\"><pre>The Lion King</pre></div></td></tr><tr><td>Embedding</td><td><div class=\"dni-plaintext\"><pre>[ 0.10022575, -0.23998135 ]</pre></div></td></tr><tr><td>Similarity</td><td><div class=\"dni-plaintext\"><pre>0.6236047</pre></div></td></tr></tbody></table></div></details></td></tr></tbody></table></td></tr></tbody></table></div></details><style>\r\n",
       ".dni-code-hint {\r\n",
       "    font-style: italic;\r\n",
       "    overflow: hidden;\r\n",
       "    white-space: nowrap;\r\n",
       "}\r\n",
       ".dni-treeview {\r\n",
       "    white-space: nowrap;\r\n",
       "}\r\n",
       ".dni-treeview td {\r\n",
       "    vertical-align: top;\r\n",
       "    text-align: start;\r\n",
       "}\r\n",
       "details.dni-treeview {\r\n",
       "    padding-left: 1em;\r\n",
       "}\r\n",
       "table td {\r\n",
       "    text-align: start;\r\n",
       "}\r\n",
       "table tr { \r\n",
       "    vertical-align: top; \r\n",
       "    margin: 0em 0px;\r\n",
       "}\r\n",
       "table tr td pre \r\n",
       "{ \r\n",
       "    vertical-align: top !important; \r\n",
       "    margin: 0em 0px !important;\r\n",
       "} \r\n",
       "table th {\r\n",
       "    text-align: start;\r\n",
       "}\r\n",
       "</style>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "top3MoviesNaive"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".NET (C#)",
   "language": "C#",
   "name": ".net-csharp"
  },
  "language_info": {
   "name": "polyglot-notebook"
  },
  "polyglot_notebook": {
   "kernelInfo": {
    "defaultKernelName": "csharp",
    "items": [
     {
      "aliases": [],
      "languageName": "csharp",
      "name": "csharp"
     }
    ]
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}