aleksejalex · September 6, 2025 13:44
diff --git a/mermin_machine4_by_ag_published.ipynb b/mermin_machine4_by_ag_published.ipynb
 {
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyP13M19fTjeE34p19UU1tqV",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/gist/aleksejalex/65bf81fc978d2ca72e984276aea64cf1/mermin_machine4_by_ag_published.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "implementace masiny:"
      ],
      "metadata": {
        "id": "fyJGm8dFRBzN"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "ciH5joezQZch"
      },
      "outputs": [],
      "source": [
        "import random\n",
        "import copy\n",
        "\n",
        "class Detector:\n",
        "    def __init__(self, detector_id: str = \"test\", setting: str = '1'):\n",
        "        self.name = detector_id\n",
        "        self.timestep = None\n",
        "        self.setting = setting # only '1', '2' or '3'\n",
        "        #self.possible_outputs = {'R', 'G'}\n",
        "\n",
        "    def receive(self, particle):\n",
        "        if not type(particle.flavour) == dict:\n",
        "            raise ValueError(f\"particle.flavour not a 'dict'! \")\n",
        "        if self.setting not in particle.flavour:\n",
        "            raise ValueError(f\"invalid setting {self.setting}; valid: {list(particle.flavour)}\")\n",
        "        return particle.flavour[self.setting]\n",
        "\n",
        "    def set_setting(self, new_setting):\n",
        "        self.setting = new_setting\n",
        "\n",
        "class Generator:\n",
        "    def __init__(self, generator_id: str = \"test\", settings=None):\n",
        "        self.name = generator_id\n",
        "        self.timestep = None\n",
        "        self.seed = None\n",
        "        if settings is None:\n",
        "            settings = ['1', '2', '3']\n",
        "        self.settings = settings  # better to keep ordered\n",
        "        self.possible_outputs = ['R', 'G']\n",
        "\n",
        "    class Particle:\n",
        "        def __init__(self, settings: list, outcomes: list):\n",
        "            self.particle_id = \"test\"\n",
        "            self.flavour = {}\n",
        "            for setting in settings:\n",
        "                self.flavour[setting] = random.choice(outcomes)\n",
        "\n",
        "        def __str__(self):\n",
        "            return f\"Particle '{self.particle_id}' with flavour {self.flavour}\"\n",
        "\n",
        "    def emit_particle(self) -> Particle:\n",
        "        particle = self.Particle(settings=self.settings, outcomes=self.possible_outputs)\n",
        "        return particle\n",
        "\n",
        "\n",
        "def experiment_maker(num_of_steps : int) -> list:\n",
        "    settings = ['1', '2', '3']\n",
        "    g1 = Generator()\n",
        "    d1 = Detector(setting='1')\n",
        "    d2 = Detector(setting='1')\n",
        "    results = []\n",
        "    for _ in range(num_of_steps):\n",
        "        p = g1.emit_particle()\n",
        "        s1 = random.choice(settings)\n",
        "        d1.set_setting(s1)\n",
        "        s2 = random.choice(settings)\n",
        "        d2.set_setting(s2)\n",
        "        result_on_d1 = d1.receive(copy.deepcopy(p))\n",
        "        result_on_d2 = d2.receive(copy.deepcopy(p))\n",
        "        #result_on_d1 = d1.receive(p)\n",
        "        #result_on_d2 = d2.receive(p)\n",
        "        result_of_step = f\"{s1}{s2}{result_on_d1}{result_on_d2}\"\n",
        "        results.append(result_of_step)\n",
        "    return results\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "PIcKKyyAQf3O"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "Le0hUB-NQfyC"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "spusteni simulace a samotna simulace vc. vypisu vysledku"
      ],
      "metadata": {
        "id": "gLXt4IlgRGsj"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "g1 = Generator()\n",
        "p = g1.emit_particle()\n",
        "print(p)\n",
        "\n",
        "d1 = Detector(setting='1')\n",
        "print(d1.receive(p))\n",
        "\n",
        "d2 = Detector(setting='2')\n",
        "print(d2.receive(p))\n",
        "\n",
        "N = 100\n",
        "\n",
        "results = experiment_maker(num_of_steps=N)\n",
        "if N <= 100:\n",
        "    print(results)\n",
        "case_a_same_colour = 0\n",
        "case_a_total = 0\n",
        "case_b_same_colour = 0\n",
        "case_b_total = 0\n",
        "case_b_diff_colour = 0\n",
        "\n",
        "for result in results:\n",
        "    if result[0] == result[1]:\n",
        "        case_a_total = case_a_total + 1\n",
        "        #detectors have the same settings\n",
        "        if result[2] == result[3]:\n",
        "            #print(f\"same settings, same output!\")\n",
        "            case_a_same_colour = case_a_same_colour + 1\n",
        "        else:\n",
        "            pass\n",
        "\n",
        "    if result[0] != result[1]:\n",
        "        #detectors have different settings\n",
        "        case_b_total = case_b_total + 1\n",
        "        if result[2] == result[3]:\n",
        "            #print(f\"different settings, same outputs\")\n",
        "            case_b_same_colour = case_b_same_colour + 1\n",
        "        else:\n",
        "            pass\n",
        "            case_b_diff_colour = case_b_diff_colour + 1\n",
        "\n",
        "print(f\"case a: total={case_a_total}, same col={case_a_same_colour}, ratio={case_a_same_colour/case_a_total}\")\n",
        "print(f\"case b: total={case_b_total}, same col={case_b_same_colour}, ratio={case_b_same_colour/case_b_total}\")\n",
        "print(f\"case b: diff_col={case_b_diff_colour}, ratio={case_b_diff_colour/case_b_total}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C0yIaybFQftX",
        "outputId": "d99879b9-e6f8-41a3-86d3-d349abcef817"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Particle 'test' with flavour {'1': 'R', '2': 'R', '3': 'R'}\n",
            "R\n",
            "R\n",
            "['21GR', '21GG', '32RR', '31RR', '31RG', '12GG', '12GG', '31RG', '11RR', '33RR', '21GR', '23GR', '22RR', '32GG', '11GG', '13RR', '21GR', '22GG', '31GG', '22GG', '23RG', '13GG', '33RR', '31RG', '31GG', '13GR', '33RR', '21GG', '23RR', '32RG', '12GR', '23GG', '31GG', '11GG', '12GR', '11GG', '12GG', '23GR', '31RR', '13GG', '31RR', '11RR', '21GG', '32RR', '32RR', '13GR', '31RG', '31GR', '31RR', '31RR', '21RG', '33RR', '13RG', '31RR', '23RR', '11GG', '21GR', '23RG', '13GR', '31RG', '33GG', '23GG', '12RR', '32GR', '11GG', '22RR', '31GG', '33GG', '22GG', '11RR', '12GR', '12RG', '13GG', '33GG', '11RR', '31RR', '31GR', '23RR', '32GG', '22GG', '32RR', '23RG', '22GG', '21GR', '11GG', '22RR', '22GG', '22RR', '21RR', '11GG', '31RG', '13RG', '13GR', '23RG', '11GG', '13GG', '12RG', '12RG', '33RR', '22RR']\n",
            "case a: total=31, same col=31, ratio=1.0\n",
            "case b: total=69, same col=35, ratio=0.5072463768115942\n",
            "case b: diff_col=34, ratio=0.4927536231884058\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "toto je divny. Kde jsou ty tretiny v case b, co mermin slibuje?"
      ],
      "metadata": {
        "id": "x1n9L_KgQ8bP"
      }
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "5Y03PdpjQ7aL"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "m_LcAEdbQ7TF"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "p9-MWBuFQ7LA"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Dalsi vyhodnoceni od ChatGPT, asi neni dulezite..."
      ],
      "metadata": {
        "id": "n91LhgRMRLdE"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "\n",
        "def analyze_and_visualize(data):\n",
        "    # Parse into DataFrame\n",
        "    df = pd.DataFrame(data, columns=[\"raw\"])\n",
        "    df[\"code\"] = df[\"raw\"].str[:2]   # numeric part\n",
        "    df[\"label\"] = df[\"raw\"].str[2:]  # letters part\n",
        "\n",
        "    print(\"🔹 First few rows of parsed data:\")\n",
        "    print(df.head(), \"\\n\")\n",
        "\n",
        "    # Frequency counts\n",
        "    code_counts = df[\"code\"].value_counts().sort_index()\n",
        "    label_counts = df[\"label\"].value_counts().sort_index()\n",
        "    cross_tab = pd.crosstab(df[\"code\"], df[\"label\"])\n",
        "\n",
        "    print(\"🔹 Counts by code:\")\n",
        "    print(code_counts, \"\\n\")\n",
        "    print(\"🔹 Counts by label:\")\n",
        "    print(label_counts, \"\\n\")\n",
        "    print(\"🔹 Cross-tabulation:\")\n",
        "    print(cross_tab, \"\\n\")\n",
        "\n",
        "    # --- Visualization ---\n",
        "    fig, axes = plt.subplots(1, 3, figsize=(15, 5))\n",
        "\n",
        "    # Code frequency\n",
        "    sns.barplot(x=code_counts.index, y=code_counts.values, ax=axes[0])\n",
        "    axes[0].set_title(\"Frequency by Code\")\n",
        "    axes[0].set_xlabel(\"Code\")\n",
        "    axes[0].set_ylabel(\"Count\")\n",
        "\n",
        "    # Label frequency\n",
        "    sns.barplot(x=label_counts.index, y=label_counts.values, ax=axes[1])\n",
        "    axes[1].set_title(\"Frequency by Label\")\n",
        "    axes[1].set_xlabel(\"Label\")\n",
        "    axes[1].set_ylabel(\"Count\")\n",
        "\n",
        "    # Heatmap of cross-tab\n",
        "    sns.heatmap(cross_tab, annot=True, fmt=\"d\", cmap=\"Blues\", ax=axes[2])\n",
        "    axes[2].set_title(\"Code vs Label\")\n",
        "\n",
        "    plt.tight_layout()\n",
        "    plt.show()\n",
        "    return df, code_counts, label_counts, cross_tab\n",
        "\n",
        "\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "\n",
        "def analyze_mermin(data):\n",
        "    # Parse\n",
        "    df = pd.DataFrame(data, columns=[\"raw\"])\n",
        "    df[\"a\"] = df[\"raw\"].str[0]    # Alice's setting\n",
        "    df[\"b\"] = df[\"raw\"].str[1]    # Bob's setting\n",
        "    df[\"Aout\"] = df[\"raw\"].str[2] # Alice's outcome\n",
        "    df[\"Bout\"] = df[\"raw\"].str[3] # Bob's outcome\n",
        "\n",
        "    # Map outcomes: R -> +1, G -> -1\n",
        "    outcome_map = {\"R\": 1, \"G\": -1}\n",
        "    df[\"Aval\"] = df[\"Aout\"].map(outcome_map)\n",
        "    df[\"Bval\"] = df[\"Bout\"].map(outcome_map)\n",
        "\n",
        "    # Compute correlation E(a,b)\n",
        "    correlations = df.groupby([\"a\", \"b\"]).apply(\n",
        "        lambda g: (g[\"Aval\"] * g[\"Bval\"]).mean()\n",
        "    ).unstack()\n",
        "\n",
        "    # Joint frequency table\n",
        "    joint_counts = pd.crosstab([df[\"a\"], df[\"b\"]], [df[\"Aout\"], df[\"Bout\"]])\n",
        "\n",
        "    # --- Visualization ---\n",
        "    fig, axes = plt.subplots(1, 2, figsize=(12, 5))\n",
        "\n",
        "    # Heatmap of correlations\n",
        "    sns.heatmap(correlations.astype(float), annot=True, fmt=\".2f\", cmap=\"coolwarm\", center=0, ax=axes[0])\n",
        "    axes[0].set_title(\"Correlation E(a,b)\")\n",
        "\n",
        "    # Heatmap of joint frequencies\n",
        "    sns.heatmap(joint_counts, annot=True, fmt=\"d\", cmap=\"Blues\", ax=axes[1])\n",
        "    axes[1].set_title(\"Joint outcome frequencies\")\n",
        "\n",
        "    plt.tight_layout()\n",
        "    plt.show()\n",
        "\n",
        "    return df, correlations, joint_counts\n"
      ],
      "metadata": {
        "id": "TOc6OuQZQtkz"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "analyze_and_visualize(results)\n",
        "df, correlations, joint_counts = analyze_mermin(results)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "Cc4-1sn4Qfoz",
        "outputId": "de0e1b19-0106-4986-b056-cbd0784867a8"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "🔹 First few rows of parsed data:\n",
            "    raw code label\n",
            "0  21GR   21    GR\n",
            "1  21GG   21    GG\n",
            "2  32RR   32    RR\n",
            "3  31RR   31    RR\n",
            "4  31RG   31    RG \n",
            "\n",
            "🔹 Counts by code:\n",
            "code\n",
            "11    12\n",
            "12    10\n",
            "13    11\n",
            "21    10\n",
            "22    11\n",
            "23    11\n",
            "31    19\n",
            "32     8\n",
            "33     8\n",
            "Name: count, dtype: int64 \n",
            "\n",
            "🔹 Counts by label:\n",
            "label\n",
            "GG    35\n",
            "GR    17\n",
            "RG    17\n",
            "RR    31\n",
            "Name: count, dtype: int64 \n",
            "\n",
            "🔹 Cross-tabulation:\n",
            "label  GG  GR  RG  RR\n",
            "code                 \n",
            "11      8   0   0   4\n",
            "12      3   3   3   1\n",
            "13      4   4   2   1\n",
            "21      3   5   1   1\n",
            "22      6   0   0   5\n",
            "23      2   2   4   3\n",
            "31      4   2   6   7\n",
            "32      2   1   1   4\n",
            "33      3   0   0   5 \n",
            "\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1500x500 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/tmp/ipython-input-848585950.py:68: DeprecationWarning: DataFrameGroupBy.apply operated on the grouping columns. This behavior is deprecated, and in a future version of pandas the grouping columns will be excluded from the operation. Either pass `include_groups=False` to exclude the groupings or explicitly select the grouping columns after groupby to silence this warning.\n",
            "  correlations = df.groupby([\"a\", \"b\"]).apply(\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1200x500 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
 }
No results found