Copy_of_What_Is_Similarity_Between_Sentences.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Eddiebee/AI-Craft/blob/main/Copy_of_What_Is_Similarity_Between_Sentences.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "epTxUkgF0lr9"
      },
      "source": [
        "<a target=\"_blank\" href=\"https://colab.research.google.com/github/cohere-ai/notebooks/blob/main/notebooks/What_Is_Similarity_Between_Sentences.ipynb\">\n",
        "  <img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/>\n",
        "</a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "psRggLwvhi1E"
      },
      "source": [
        "# Similarity Between Words and Sentences\n",
        "\n",
        "Sentence embeddings are the bread and butter of language models, as they associate each sentence with a particular list of numbers (a vector), in a way that similar sentences give similar vectors. We can think of embeddings as a way to locate each sentence in space (a high dimensional space, but a space nonetheless), in a way that similar sentences are located close by. Once we have each sentence somewhere in space, it’s natural to think of distances between them. And an even more intuitive way to think of distances is to think of similarities, i.e., a score assigned to each pair of sentences, which is high when these sentences are similar, and low when they are different. The similarity is a very useful concept in large language models, as it can be used for search, for translation, for summarization, and in many other applications.\n",
        "\n",
        "In this notebook, we understand the intuition behind similarities between sentences, including dot product and cosine similarity.\n",
        "\n",
        "_Read the accompanying [blog post here](https://docs.cohere.com/docs/similarity-between-words-and-sentences)._"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_P5_7ksz0lsK"
      },
      "source": [
        "## Overview\n",
        "\n",
        "We'll do the following steps:\n",
        "- **Step 1: Turn Text into Embeddings** - Use Cohere's Embed endpoint to get sentence embeddings.\n",
        "- **Step 2: Calculate Dot Products** - Calculate the dot products between each pair of sentence embeddings to understand similarity between them.\n",
        "- **Step 3: Calculate Cosine Similarities** - Use scikit-learn to get the cosine similarity for each pair of sentence embeddings."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "suCXBjG_0lsL"
      },
      "source": [
        "## Setup\n",
        "\n",
        "We'll start by installing the tools we'll need and then importing them."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "0XVQfp790lsL",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "599ccce0-2db9-4498-f889-7c93bcbeeaf0"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m150.3/150.3 kB\u001b[0m \u001b[31m1.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m3.1/3.1 MB\u001b[0m \u001b[31m32.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m75.6/75.6 kB\u001b[0m \u001b[31m908.6 kB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m77.9/77.9 kB\u001b[0m \u001b[31m7.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m58.3/58.3 kB\u001b[0m \u001b[31m4.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h"
          ]
        }
      ],
      "source": [
        "! pip install cohere -q"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "metadata": {
        "id": "Zg5H7gd7V0Mu"
      },
      "outputs": [],
      "source": [
        "import cohere\n",
        "import numpy as np\n",
        "import seaborn as sns\n",
        "import altair as alt\n",
        "from sklearn.metrics.pairwise import cosine_similarity"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Rut3HaRP0lsO"
      },
      "source": [
        "Fill in your Cohere API key in the next cell. To do this, begin by [signing up to Cohere](https://os.cohere.ai/) (for free!) if you haven't yet. Then get your API key [here](https://dashboard.cohere.com/api-keys)."
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import userdata\n",
        "COHERE_TRIAL_KEY = userdata.get('COHERE_TRIAL_KEY')"
      ],
      "metadata": {
        "id": "c5v09knt10gm"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "metadata": {
        "id": "EzvjPBYy0lsP"
      },
      "outputs": [],
      "source": [
        "# Paste your API key here. Remember to not share publicly\n",
        "co = cohere.Client(COHERE_TRIAL_KEY)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4_YQVuAbf3Xh"
      },
      "source": [
        "## Step 1: Turn Text into Embeddings\n",
        "\n",
        "In this notebook, we'll work with three sentences and store them in a Python list `texts`."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "metadata": {
        "id": "LbRdqqgC0lsQ"
      },
      "outputs": [],
      "source": [
        "texts = [\"I like to be in my house\",\n",
        "         \"I enjoy staying home\",\n",
        "         \"the isotope 238u decays to 206pb\"]"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KxvVoL6A0lsR"
      },
      "source": [
        "To get the corresponding sentence embeddings, we call the [Embed endpoint](https://docs.cohere.com/reference/embed) with `co.embed()`. We supply three parameters:\n",
        "- `texts` - our list of sentences\n",
        "- `model` - we use `embed-english-v3.0`, Cohere's latest (at the time of writing) English-only embeddings model to generate the embeddings\n",
        "- `input_type` - we use `search_document` to indicate that we intend to use the embeddings for search use-cases\n",
        "\n",
        "You'll learn about these parameters in more detail in the [LLMU Module on Text Representation](https://docs.cohere.com/docs/intro-text-representation)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "metadata": {
        "id": "6HujlbFsIZjW"
      },
      "outputs": [],
      "source": [
        "response = co.embed(\n",
        "    texts=texts,\n",
        "    model='embed-english-v3.0',\n",
        "    input_type='search_document'\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2cB6L5C40lsS"
      },
      "source": [
        "The embeddings are stored in the `embeddings` value of the response. After getting the embeddings, we separate them by sentence."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QIZoRxXVNPQY",
        "outputId": "139a2475-0644-4696-d093-f42fead761fc"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Embedding for sentence 1 [ 0.05075073  0.03753662 -0.02958679 ... -0.07366943 -0.00842285\n",
            " -0.01834106]\n",
            "Embedding for sentence 2 [ 0.04333496  0.05407715 -0.02453613 ... -0.06182861 -0.01983643\n",
            " -0.0096283 ]\n",
            "Embedding for sentence 3 [ 0.02383423  0.00695801 -0.04669189 ... -0.04006958 -0.02304077\n",
            "  0.01963806]\n"
          ]
        }
      ],
      "source": [
        "embeddings = response.embeddings\n",
        "\n",
        "[sentence1, sentence2, sentence3] = embeddings\n",
        "\n",
        "print(\"Embedding for sentence 1\", np.array(sentence1))\n",
        "print(\"Embedding for sentence 2\", np.array(sentence2))\n",
        "print(\"Embedding for sentence 3\", np.array(sentence3))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9pR9yDy1ot17"
      },
      "source": [
        "Note that the embeddings are vectors (lists) of 1024 numbers, so they are truncated here (thus the dots in between).  One would expect that the vectors corresponding to sentences 1 and 2 are similar to each other and that both are different from the vector corresponding to sentence 3. However, from inspection, this is not very clear. We need to calculate some similarities to see if this is the case.  We will do that in the following two sections.\n",
        "\n",
        "## Step 2: Calculate Dot Products\n",
        "\n",
        "Let’s calculate the dot products between the three sentences to understand how similar they are to each other."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8b-FViGlJQLe",
        "outputId": "b957f7df-d7d1-4049-e635-f518261bb05f"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Similarity between sentences 1 and 2: 0.8187172323456495\n",
            "Similarity between sentences 1 and 3: 0.1952170633623921\n",
            "Similarity between sentences 2 and 3: 0.19826654576138197\n"
          ]
        }
      ],
      "source": [
        "print(\"Similarity between sentences 1 and 2:\", np.dot(sentence1, sentence2))\n",
        "print(\"Similarity between sentences 1 and 3:\", np.dot(sentence1, sentence3))\n",
        "print(\"Similarity between sentences 2 and 3:\", np.dot(sentence2, sentence3))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "xKFbCIv80lsX"
      },
      "source": [
        "The similarity between sentences 1 and 2 (0.8188) is much larger than the similarities between the other pairs. This confirms our predictions.\n",
        "\n",
        "Just for consistency, let’s calculate the similarities between each sentence and itself, to confirm that a sentence and itself has the highest similarity score."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sb-MX-MdhlPj",
        "outputId": "339826aa-9122-4e57-9017-66ee544b0839"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Similarity between sentences 1 and 1: 1.0001818411403134\n",
            "Similarity between sentences 2 and 2: 0.9997857473557472\n",
            "Similarity between sentences 3 and 3: 0.9998230785105998\n"
          ]
        }
      ],
      "source": [
        "print(\"Similarity between sentences 1 and 1:\", np.dot(sentence1, sentence1))\n",
        "print(\"Similarity between sentences 2 and 2:\", np.dot(sentence2, sentence2))\n",
        "print(\"Similarity between sentences 3 and 3:\", np.dot(sentence3, sentence3))"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qX98N4470lsZ"
      },
      "source": [
        "This checks out—the similarity between a sentence and itself is around 1, which is higher than all the other similarities."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DWEpWsI0o8gn"
      },
      "source": [
        "## Step 3: Calculate Cosine Similarities\n",
        "\n",
        "We use the `cosine_similarity()` function from scikit-learn to measure cosine similarity between the three sentences."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "D4ZkpwMSL_oH",
        "outputId": "22281596-10c2-4df0-fcca-5f4d3242dbb6"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cosine similarity between sentences 1 and 2: 0.8187305165459494\n",
            "Cosine similarity between sentences 1 and 3: 0.195216586304462\n",
            "Cosine similarity between sentences 2 and 3: 0.1983053317541077\n"
          ]
        }
      ],
      "source": [
        "print(\"Cosine similarity between sentences 1 and 2:\", cosine_similarity([sentence1], [sentence2])[0][0])\n",
        "print(\"Cosine similarity between sentences 1 and 3:\", cosine_similarity([sentence1], [sentence3])[0][0])\n",
        "print(\"Cosine similarity between sentences 2 and 3:\", cosine_similarity([sentence2], [sentence3])[0][0])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hOIXLBU_0lsa"
      },
      "source": [
        "Now let’s check the similarity between each sentence and itself."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tZ7ls1JlkngY",
        "outputId": "5832c532-ce33-4f7a-ef69-882890d36a38"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cosine similarity between sentences 1 and 1: 1.0000000000000007\n",
            "Cosine similarity between sentences 2 and 2: 0.9999999999999999\n",
            "Cosine similarity between sentences 3 and 3: 0.9999999999999999\n"
          ]
        }
      ],
      "source": [
        "print(\"Cosine similarity between sentences 1 and 1:\", cosine_similarity([sentence1], [sentence1])[0][0])\n",
        "print(\"Cosine similarity between sentences 2 and 2:\", cosine_similarity([sentence2], [sentence2])[0][0])\n",
        "print(\"Cosine similarity between sentences 3 and 3:\", cosine_similarity([sentence3], [sentence3])[0][0])"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TYoCPTsh0lsb"
      },
      "source": [
        "We also plot the results in a grid."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 653
        },
        "id": "sO3AxCUJq7AH",
        "outputId": "8cc6af01-2240-4293-8805-405649ed4d55"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[Text(0.5, 0, 'I like to be in my house'),\n",
              " Text(1.5, 0, 'I enjoy staying home'),\n",
              " Text(2.5, 0, 'the isotope 238u decays to 206pb')]"
            ]
          },
          "metadata": {},
          "execution_count": 12
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ],
      "source": [
        "# Get pairwise dot product similarities\n",
        "dot_product_similarities = [[cosine_similarity([embeddings[i]], [embeddings[j]])[0][0] for i in range(len(embeddings))] for j in range(len(embeddings))]\n",
        "\n",
        "# Plot in 3x3 grid\n",
        "ax = sns.heatmap(dot_product_similarities, vmin=0, vmax=1,\n",
        "                 linewidths=1, linecolor='grey',\n",
        "                 xticklabels=texts,\n",
        "                 yticklabels=texts,\n",
        ")\n",
        "ax.set_xticklabels(labels=texts, rotation=45)"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "EAvhbYd70lsb"
      },
      "source": [
        "Notice that the dot product and cosine distance give nearly identical values. The reason for this is that the embedding is normalized (meaning each vector has norm equal to 1). When the embedding is not normalized, the dot product and cosine distance would give different values."
      ]
    }
  ],
  "metadata": {
    "colab": {
      "provenance": [],
      "include_colab_link": true
    },
    "gpuClass": "standard",
    "kernelspec": {
      "display_name": "Python 3 (ipykernel)",
      "language": "python",
      "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": "3.11.5"
    },
    "vscode": {
      "interpreter": {
        "hash": "1fb8019e3560b882083e525615cf48e713d3a7345a15eb723d805e91aa410aac"
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}