{
    "cells": [
        {
            "attachments": {},
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "# Probabilistic Paradigm"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "# Initial Set-Up"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 2,
            "metadata": {},
            "outputs": [
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "WARNING (pytensor.configdefaults): g++ not available, if using conda: `conda install m2w64-toolchain`\n",
                        "WARNING (pytensor.configdefaults): g++ not detected!  PyTensor will be unable to compile C-implementations and will default to Python. Performance may be severely degraded. To remove this warning, set PyTensor flags cxx to an empty string.\n",
                        "WARNING (pytensor.tensor.blas): Using NumPy C-API based implementation for BLAS functions.\n"
                    ]
                }
            ],
            "source": [
                "import pymc as pm\n",
                "\n",
                "SAMPLES = 100"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "## Predecessor and Successor"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 2,
            "metadata": {},
            "outputs": [
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "Metropolis: [a]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:04&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 29 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n",
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "Metropolis: [a]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:05&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 28 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n"
                    ]
                }
            ],
            "source": [
                "def predecessor(a: int) -> int:\n",
                "    with pm.Model():\n",
                "        interim_variable = pm.DiracDelta(\"a\", a)\n",
                "        result = pm.Deterministic(\"result\", interim_variable - 1)\n",
                "        data = pm.sample(SAMPLES)\n",
                "    \n",
                "    mean_result = data.posterior.result.mean()    \n",
                "    return int(mean_result.values)    \n",
                "\n",
                "\n",
                "assert predecessor(1) == 0\n",
                "assert predecessor(10) == 9\n"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 3,
            "metadata": {},
            "outputs": [
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "Metropolis: [a]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:04&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 27 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n",
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "Metropolis: [a]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:04&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 27 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n"
                    ]
                }
            ],
            "source": [
                "def successor(a: int) -> int:\n",
                "    with pm.Model():\n",
                "        interim_variable = pm.DiracDelta(\"a\", a)\n",
                "        result = pm.Deterministic(\"result\", interim_variable + 1)\n",
                "        data = pm.sample(SAMPLES)\n",
                "    \n",
                "    mean_result = data.posterior.result.mean()    \n",
                "    return int(mean_result.values) \n",
                "\n",
                "\n",
                "assert successor(0) == 1\n",
                "assert successor(10) == 11\n"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "## Addition"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 4,
            "metadata": {},
            "outputs": [
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "CompoundStep\n",
                        ">Metropolis: [a1]\n",
                        ">Metropolis: [a2]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:06&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 27 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n",
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "CompoundStep\n",
                        ">Metropolis: [a1]\n",
                        ">Metropolis: [a2]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:04&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 27 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n",
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "CompoundStep\n",
                        ">Metropolis: [a1]\n",
                        ">Metropolis: [a2]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:04&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 23 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n",
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "CompoundStep\n",
                        ">Metropolis: [a1]\n",
                        ">Metropolis: [a2]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:04&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 24 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n"
                    ]
                }
            ],
            "source": [
                "def addition(addend_1: int, addend_2: int) -> int:\n",
                "    with pm.Model():\n",
                "        interim_addend_1 = pm.DiracDelta(\"a1\", addend_1)\n",
                "        interim_addend_2 = pm.DiracDelta(\"a2\", addend_2)\n",
                "        result = pm.Deterministic(\"result\", interim_addend_1 + interim_addend_2)\n",
                "        data = pm.sample(SAMPLES)\n",
                "    \n",
                "    mean_result = data.posterior.result.mean()    \n",
                "    return int(mean_result.values) \n",
                "\n",
                "\n",
                "assert addition(0, 0) == 0\n",
                "assert addition(1, 0) == 1\n",
                "assert addition(0, 1) == 1\n",
                "assert addition(10, 10) == 20\n"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "## Multiplication"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 5,
            "metadata": {},
            "outputs": [
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "CompoundStep\n",
                        ">Metropolis: [m1]\n",
                        ">Metropolis: [m2]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:05&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 26 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n",
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "CompoundStep\n",
                        ">Metropolis: [m1]\n",
                        ">Metropolis: [m2]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:04&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 24 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n",
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "CompoundStep\n",
                        ">Metropolis: [m1]\n",
                        ">Metropolis: [m2]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:05&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 25 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n",
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "CompoundStep\n",
                        ">Metropolis: [m1]\n",
                        ">Metropolis: [m2]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:05&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 27 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n"
                    ]
                }
            ],
            "source": [
                "def multiplication(multiplicand: int, multiplier: int) -> int:\n",
                "    with pm.Model():\n",
                "        interim_multiplicand = pm.DiracDelta(\"m1\", multiplicand)\n",
                "        interim_multiplier = pm.DiracDelta(\"m2\", multiplier)\n",
                "        result = pm.Deterministic(\"result\", interim_multiplicand * interim_multiplier)\n",
                "        data = pm.sample(SAMPLES)\n",
                "    \n",
                "    mean_result = data.posterior.result.mean()    \n",
                "    return int(mean_result.values) \n",
                "\n",
                "\n",
                "assert multiplication(0, 0) == 0\n",
                "assert multiplication(2, 0) == 0\n",
                "assert multiplication(0, 2) == 0\n",
                "assert multiplication(10, 10) == 100\n"
            ]
        },
        {
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "## Exponentiation"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 6,
            "metadata": {},
            "outputs": [
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "CompoundStep\n",
                        ">Metropolis: [b]\n",
                        ">Metropolis: [e]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:05&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 27 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n",
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "CompoundStep\n",
                        ">Metropolis: [b]\n",
                        ">Metropolis: [e]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:05&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 27 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n",
                        "Only 100 samples in chain.\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "CompoundStep\n",
                        ">Metropolis: [b]\n",
                        ">Metropolis: [e]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='4400' class='' max='4400' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [4400/4400 00:04&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 100 draw iterations (4_000 + 400 draws total) took 28 seconds.\n",
                        "w:\\Repositories\\understanding-programming-paradigms\\.venv\\lib\\site-packages\\arviz\\stats\\diagnostics.py:584: RuntimeWarning: invalid value encountered in scalar divide\n",
                        "  (between_chain_variance / within_chain_variance + num_samples - 1) / (num_samples)\n"
                    ]
                }
            ],
            "source": [
                "def exponentiation(base: int, exponent: int) -> float:\n",
                "    with pm.Model():\n",
                "        interim_base = pm.DiracDelta(\"b\", base)\n",
                "        interim_exponent = pm.DiracDelta(\"e\", exponent)\n",
                "        result = pm.Deterministic(\"result\", interim_base ** interim_exponent)\n",
                "        data = pm.sample(SAMPLES)\n",
                "    \n",
                "    mean_result = data.posterior.result.mean()    \n",
                "    return int(mean_result.values) \n",
                "\n",
                "\n",
                "assert exponentiation(1, 0) == 1\n",
                "assert exponentiation(0, 1) == 0\n",
                "assert exponentiation(3, 3) == 27\n"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": null,
            "metadata": {},
            "outputs": [],
            "source": []
        },
        {
            "attachments": {},
            "cell_type": "markdown",
            "metadata": {},
            "source": [
                "What is particular of the Probabilistic Paradigm?"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 22,
            "metadata": {},
            "outputs": [],
            "source": [
                "men_height_mean = 175.768\n",
                "men_height_std = 6.7564\n",
                "\n",
                "women_height_mean = 163.322\n",
                "women_height_std = 6.5532"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 24,
            "metadata": {},
            "outputs": [
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Auto-assigning NUTS sampler...\n",
                        "Initializing NUTS using jitter+adapt_diag...\n",
                        "Multiprocess sampling (4 chains in 4 jobs)\n",
                        "NUTS: [Height_{man}, Height_{woman}]\n"
                    ]
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "<style>\n",
                            "    /* Turns off some styling */\n",
                            "    progress {\n",
                            "        /* gets rid of default border in Firefox and Opera. */\n",
                            "        border: none;\n",
                            "        /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
                            "        background-size: auto;\n",
                            "    }\n",
                            "    progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
                            "        background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
                            "    }\n",
                            "    .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
                            "        background: #F44336;\n",
                            "    }\n",
                            "</style>\n"
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "data": {
                        "text/html": [
                            "\n",
                            "    <div>\n",
                            "      <progress value='8000' class='' max='8000' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
                            "      100.00% [8000/8000 00:10&lt;00:00 Sampling 4 chains, 0 divergences]\n",
                            "    </div>\n",
                            "    "
                        ],
                        "text/plain": [
                            "<IPython.core.display.HTML object>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                },
                {
                    "name": "stderr",
                    "output_type": "stream",
                    "text": [
                        "Sampling 4 chains for 1_000 tune and 1_000 draw iterations (4_000 + 4_000 draws total) took 34 seconds.\n"
                    ]
                }
            ],
            "source": [
                "with pm.Model():\n",
                "    men_height = pm.Normal(\"Height_{man}\", mu=men_height_mean, sigma=men_height_std)\n",
                "    women_height = pm.Normal(\"Height_{woman}\", mu=women_height_mean, sigma=women_height_std)\n",
                "    difference = pm.Deterministic(\"result\", men_height - women_height)\n",
                "    data = pm.sample()\n",
                "\n",
                "mean_result = data.posterior.result.mean()"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 14,
            "metadata": {},
            "outputs": [],
            "source": [
                "import arviz as az\n",
                "az.style.use(\"bmh\")"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 15,
            "metadata": {},
            "outputs": [
                {
                    "data": {
                        "text/plain": [
                            "array([[<AxesSubplot: title={'center': 'result'}>,\n",
                            "        <AxesSubplot: title={'center': 'result'}>]], dtype=object)"
                        ]
                    },
                    "execution_count": 15,
                    "metadata": {},
                    "output_type": "execute_result"
                },
                {
                    "data": {
                        "image/png": "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",
                        "text/plain": [
                            "<Figure size 1200x200 with 2 Axes>"
                        ]
                    },
                    "metadata": {},
                    "output_type": "display_data"
                }
            ],
            "source": [
                "az.plot_trace(data, var_names=[\"result\"])"
            ]
        },
        {
            "cell_type": "code",
            "execution_count": 41,
            "metadata": {},
            "outputs": [
                {
                    "name": "stdout",
                    "output_type": "stream",
                    "text": [
                        "There is a probability of 0.62% that the difference is larger than 35cm\n"
                    ]
                }
            ],
            "source": [
                "threshold = 35\n",
                "probability = float((data.posterior.result > threshold).mean().values)\n",
                "print(f\"There is a probability of {probability:.2%} that the difference is larger than {threshold}cm\")"
            ]
        }
    ],
    "metadata": {
        "kernelspec": {
            "display_name": "Python 3.10.4 ('.venv': venv)",
            "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": "3.9.13 (tags/v3.9.13:6de2ca5, May 17 2022, 16:36:42) [MSC v.1929 64 bit (AMD64)]"
        },
        "orig_nbformat": 4,
        "vscode": {
            "interpreter": {
                "hash": "a3349956cd5f31eb09cfa8ab65ba9774bc4eb05174a2fc2b960d18a93fca386a"
            }
        }
    },
    "nbformat": 4,
    "nbformat_minor": 2
}