{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "cell-00",
      "metadata": {},
      "source": [
        "# CDNOW: from a raw transaction log to lifetime value\n",
        "\n",
        "**Audience.** An analyst who believes CLV is `average ticket x frequency x margin`,\n",
        "is comfortable with pandas, and has never met the buy-till-you-die literature.\n",
        "\n",
        "**Prerequisites.** `uv sync` in the repo root. No network and no API keys. The\n",
        "CDNOW sample ships with the repository.\n",
        "\n",
        "**Learning goals.** By the end you can:\n",
        "\n",
        "1. Define frequency, recency and T the way this literature defines them, which is\n",
        "   not the way RFM defines two of them.\n",
        "2. Turn a raw transaction log into a `CustomerBase`, and say why `time_unit` and\n",
        "   `collapse` are two arguments rather than one.\n",
        "3. Reproduce the published Fader, Hardie & Lee (2005) BG/NBD estimates on CDNOW,\n",
        "   then check the fit against 39 weeks it never saw.\n",
        "4. Compose a transaction model and a spend model into discounted lifetime value,\n",
        "   and get the numbers back out as a DataFrame.\n",
        "\n",
        "Nothing below assumes you've read a BTYD paper. Section 0 builds the vocabulary\n",
        "from scratch, and every term it defines gets drawn on real customers in section 2."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-01",
      "metadata": {},
      "source": [
        "## Outline\n",
        "\n",
        "0. The vocabulary, and the word that trips everyone\n",
        "1. Load the raw log\n",
        "2. Four customers, drawn\n",
        "3. `CustomerBase`, and its two questions about time\n",
        "4. Reproducing the published estimates\n",
        "5. What happens if you get recency wrong\n",
        "6. Predicted against actual, on the holdout\n",
        "7. Lifetime value, plotted and exported\n",
        "8. What `time_unit=\"W\"` does on its own\n",
        "9. Exercise"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-02",
      "metadata": {},
      "source": [
        "## 0. The vocabulary, and the word that trips everyone\n",
        "\n",
        "### What the models are actually looking at\n",
        "\n",
        "Every model in `clvkit` reads one thing: a **transaction log**. That's a table\n",
        "with one row per purchase, carrying who bought, when, and optionally how much.\n",
        "\n",
        "```\n",
        "customer_id        date   amount\n",
        "          1  1997-01-01    11.77\n",
        "          1  1997-01-18    89.00\n",
        "          2  1997-01-01    45.55\n",
        "```\n",
        "\n",
        "That's all. No demographics, no channel, no campaign. The claim these models make\n",
        "is that the timing of somebody's past purchases predicts their future ones, and\n",
        "they'd rather be judged on that than on features they can't get.\n",
        "\n",
        "From the log, `clvkit` computes four numbers per customer. Those four numbers are\n",
        "the whole interface between your data and the maths.\n",
        "\n",
        "### The glossary\n",
        "\n",
        "| Term | What it means here | Watch out |\n",
        "| --- | --- | --- |\n",
        "| **Transaction log** | One row per purchase: who, when, how much. | Not one row per SKU. A basket is one purchase. |\n",
        "| **Purchase event** | One shopping trip, after same-period purchases are merged. | Two orders on the same day are one event. |\n",
        "| **`frequency`** | The number of **repeat** purchases. Total purchases minus one. | A customer who bought 3 times has `frequency = 2`. Someone who bought once has `0`, not `1`. |\n",
        "| **`recency`** (`t_x`) | Time from the customer's **first** purchase to their **last** one. | This is *not* \"time since the last purchase\". See below. |\n",
        "| **`T`** | The customer's **age**: time from their first purchase to the end of the observation window. | Per customer, not a calendar constant. Two customers observed on the same day have different `T` if they arrived in different months. |\n",
        "| **`monetary_value`** | Average amount per **repeat** purchase. | The first purchase is excluded, which surprises people. |\n",
        "| **BTYD** | \"Buy till you die\", the family of models here. | A name nobody outside the field uses. |\n",
        "| **`P(alive)`** | Probability the customer hasn't silently churned by the end of the window. | Nobody cancels anything in this world, so churn is never observed. It's inferred from silence. |\n",
        "| **Calibration / holdout** | Fit on the first slice of time, test on the rest. | The holdout is behaviour the fit never saw. |\n",
        "| **CLV** | Discounted expected revenue (or margin) over a future horizon. | Residual value, from now forward. It doesn't include what the customer already spent. |"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-03",
      "metadata": {},
      "source": [
        "### The word that trips everyone: recency\n",
        "\n",
        "Ask anyone what recency means and you'll get: *today minus the last purchase*. A\n",
        "customer who bought 10 days ago has a recency of 10. That's the RFM definition,\n",
        "it's a perfectly good definition, and it isn't the one used here.\n",
        "\n",
        "In BTYD, `recency` is `t_x`: the time from the customer's **first** purchase to\n",
        "their **last** one. The two definitions point at opposite ends of the same\n",
        "timeline, which is easier to see drawn than described."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "id": "cell-04",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAABA8AAAFUCAYAAABLMRVIAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjExLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlcelbwAAAAlwSFlzAAAPYQAAD2EBqD+naQAAVp1JREFUeJzt3Xd8U9X/x/F30qaTDlpWKbPs0SJLQECQPQRRRHGBgAIKguur/kRBEb+4laGyBFEZMpQh4mApqIDMMgVahuxCB6UjbZr8/gDytdKmTWmb0r6ejwePB0nOPfdzbtJzbj4591yDxZJuEwAAAAAAQDaMrg4AAAAAAAAUbSQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPAAAAAACAQyQPnJSRkaERI0aoSpUqCggIVGRkpDp27KR3333X1aEVqPj4eAUEBGrHjp2uDgUAUADGjXtNvXvf5eowClW9evU1b948V4cBQFK5cuW1Zs2aG6ojKipaXbp0VfnyFRQeHn5DdVksFgUEBGrjxo03VI+UP20DigKSB05atWqVvv9+tbZu3aqEhHhFRETkS7333NNXY8a8ki91AQDgamPGvKJ77unr6jAAlCBvvfWWypUrpxMnjmvPnj2uDgcodtxdHcDN5siRKFWvXl0VKlSwP7d2LZlEAAAAwJWio6PVo0cPeXp63nBd7u7uSkiIv/GggGKEmQdOePDBB/X6669r+/btCggIzPTvn5ct9OzZU6NHP62BAx9VpUqVddddfSRJP/30k9q2vV0hIRXVpElTTZkyRVarVY8/PlRr167V1KlT7fVFRUVft/+ePXtq1KjRGjBgoMLCaqhatep6/fXxslqt9jItW7bSJ598kmm7Rx4ZoKeffibH+C5evKgRI0aoRo2aqlKligYPHqKYmJhMdW3b9qe6dOmq0NBKaty4iZYvX57p9erVwxQQEKjAwNJq2LChxox5RampqfbXbTabJk6cqAYNGqpixVB17twl03Swixcv6oknnlD16mGqXLmK7ryzl3bv3p3LdwgAkF8++eQT+5hUqVJlde3aTVu3bs1UZvfu3erZs6dCQyupQYOGGjNmjJKTk/XBBx9q6tSpWrt2rb2O77777rp9PP/8f3T33ffo+ef/o/r1G6hixVANHz5cSUlJ9jKDBg3WyJFPZdpu4sSJ6tq123X1vPDCC6pePUy1a9eRJJnNZo0b95oaNGiokJCK6tPnbh0+fDhTXUePHlW/fvepatVqqlevviZNmpzp9Z49e9rbULt2HQ0dOlQXLlzIVOarr75Ss2bNVaFCiFq1uk2LFi2yv2Y2m/Xqq2NVt249hYRUVLt27fXTTz/l5i0AipycztOuXeY6Y8YMh+eLBw8etF9e0Lz5rfr887k57js6+qjuu+9+hYRUVOXKVTRw4KM6d+6c/fWAgEBt27ZN48ePV0BAoMaNe+26OkaMGKmhQ4faH7///gcKCAjUhg2/2J9r2rSZZs+ec91lC/nZtpza4kycQGEieeCE+fPn65VXxqhZs2ZKSIi3/2vWrNl1ZefOnasOHTro4MEDWr58mRISEvTIIwM0cOBARUUd0TffLFVMzAUdOHBAM2fOUMeOHTVy5Eh7nTVqhGUZw9y5c9W8eTPt3LlDc+d+rs8//1wzZsxwui3/ji89PV29e/dWdHS0Vq5coT179qhHjx5avnxFpu0++eRTTZz4Xx069JceffRRDRs2PFNnd/RotBIS4nXx4gUtWLBA69ev0zvv/C+x8u23yzRz5ix99dVXioo6orfeeksLF34t6cq1ZX363C2DwaBNmzZq37696ty5s+66q891SQwAQMF68skn7WPS3r171KVLF/Xrd59iY2PtZQYNGqxGjRpp3769WrdurSpVqqR169bp2Wef0ciRI9WxY0d7HXfeeWeW+1m3bp0sFos2bvxVP//8k7Zt267/+7+XnY533bp18vb20fbt23Xo0F+SpMcfH6qVK1dq1qyZOnToL40ePdo+5lwzY8YMDR36uPbv36f33ntX48aN0+bNm+2vr1q1SgkJ8YqPj9NPP/2o2Ng4PfXU/5IZ+/fv1+jRT+vNNyfo2LGjmjv3c61bt86eAHnyyRHatWuXli9fpqioI3rmmWc0YMBA7dy5y+k2Aq7kzHmao/PF9PR03X9/f1WtWlV79kRq/vx5+uyzz2Q2mx3uu1+/fjKZTNqxY7s2bvxVMTHn9cgjA+xlEhLidcstt2jcuHFKSIjX66+/dl09bdu20caNm+yPN23apODgYHuC4Ny5czpy5Ijatm2TbSw32rbctCU/4gQKAsmDAtK5c2cNHDhApUqVkiTFxMQoNTVVPXv2kI+Pj6pVq6bx419XgwYNnKr31ltv1VNPPaWAgAC1a9dOTz31lD7++OMbju+771bp0KHDmjNnjurXr6+AgADde29fPfbYkEzbjRnzspo2bSpfX1899dRI2Ww27dq167r63dzcFB4ermeffU7Lli2zP3/ixAlVrVpVjRvfIm9vbzVt2kQffzxVkrR69WqdPHlSU6ZMUWhoqPz9/TV69ChVrBiilSuv/8UKAFA4AgMD9dxzz6p06dL2X75sNpv+/vtvdezYSYGBgSpfvryeeOKJbJME2QkODtbbb7+l4OBgNWjQQBMmvKF58+YpLi7eqXqqVKmicePGKiiotCTpr7/+0vLlyzV58iS1atVKfn5+uuOO9nr11czrCw0Z8pg6d+4sX19f9ezZUxEREfrjj83X1W8wGOxj9+rVP9i/DJw8eVK+vj6644475OXlpdq1a2vatGny9fVVVFSUlixZok8//UR16tSRj4+P+vS5Sz179tT8+SzUiJuLM+dpjs4Xv/tulWJiYvT++++pXLlyqlWrlt55522H+/7uu1X2fYeEhKhatWqaMmWqtmzZoj/++CPXbWjbtq1Onz6tqKgoWSwWbd26VSNHjrR/Kd+4caPKly+vWrVqZVvHjbYtN23JjziBgkDyoIDUrVs30+OwsDB16NBBHTp01Jgxr+jHH3/MNJ0/txo3viXT4yZNmujEib8zTfHMS3yRkZEKCwvLtJZDVsLCatj/bzQa5e/vr/j4ePtzixcvVtu2tys0tJICAgI1ZMgQnTx50v567969dfLkSXXq1FkffPBhpl9edu7cpdjYWJUtW06lSwepdOkgBQaW1r59+3Xs2DGn2gcAuDFnz57VsGHDVLduPZUuHaSAgEAdPXrU3qcbDAY99tgQDRr0qJ5++hktXfpNpvEgt+rXr5/p+uQmTZrIYrEoOjrKqXrq1Kkjo/F/pzWRkZFyc3NTy5YtHW7375l+gYGBmdqxceNG9ejRQ1WrVlNAQKBatbpNNptNJ0+ekiS1bt1aoaGVdNttrTVhwgRt3LhRFotFkrRz55U7FEVENFJQULB9XFuyZImOHTvuVPsAV3PmPM3R+eJffx1UjRo15O/vby/TuHFjGQyGbPd9ZZswe3JQuvK3GxwcrIMH/8p1G0JDQ1WtWjVt3LhJO3bsUGBgoAYMeEQ7d+5UUlKSNm7cpDZtHP+af6Nty01b8iNOoCCQPCggHh4emR4bjUZ9881STZ8+TZ6ennr99dfVpElTHTp0yKl6HXWs2fnnmgjZxWez2XJVt6Myf/75p0aMGKlRo57Snj17FBcXq3nzvrKfRElSWFh17dixXcOHD9fx48fVt29f9e17rywWi6xWq+rUqaPY2IuKi4tVXFys4uPjlJAQr/HjX3eixQCAGzV8+BOKi4vXihXLdebMaSUkxKtevXqZ+vSJEydq5cqVqlSpkmbNmqnw8AitXbvOqf3k37hmyvQ4P8a1mJgY3X9/f3Xs2Elbt25RbOxF7dixQ5KUkXHlOPj6+uqXXzbozTcnKDk5WSNGjFSbNm0VGxsrq9Uqd3d3nTlz2j62XRvXFi9elO1+gaLImfO0nP72/v26wWCQzWbLtnx2r+X27/yf2rRpo40bN2rjxk1q27aNypQpo7CwMG3evEWbNuX8pfxG25bbttxonEBBIHlQiAwGg26//XaNHfuqfvvtNwUHB2vevPmSJJPJXTbb9SdD/7Zjx85/Pd6hypUrydfXV9L1v5hIV9YhyEmjRo0UFRWVaf0CZ/3555+qXbuW+vXrp6Cg0jIajdq+fcd15fz8/HTvvX01adJH2rBhvdasWaOdO3eqUaNGOnLkiI4ePZbnGAAA+WPr1q0aPHiQateuLS8vL8XGxuno0aPXlWvUqJGef/45rV69Wt27d9P06dMlSSaTKcsv+f+2b9++TNcD79ixQ+7u7vZf97Ia17JaVDiruCwWi7Zs2ZJj2ezs3btPaWlpevbZZ1S+fHm5ublpx47t15Xz8PBQ165d9d///ld//rlV586d08qV39lj2LBhQ55jAIqK/DpPq1Onro4cOaLExET7c9eSctmpW7eeoqKiM13OFB19VLGxsapTp7ZT+2/bto02bdqU6Qt4mzZttHTpkhteRyA3bcttWwoyTiCvSB4Uks2bN+uZZ57V3r17lZqaqsjISJ05c0bVq1eXJFWuXEWRkZE5Xn6wdetWTZkyRQkJCfr11181ZcoUPfnkk/bXW7Vqpa+/XqQjR44oPj5e77zzjvbt259jfD179lDNmjU1ePBgHThwQAkJCVqyZKlmzfos122sXbu2Dh8+ol9//VVJSUlatmy5pk2blqnM++9/oBkzZujvv/9WSkqK1q5dJ5PJpNDQUN15Z0/Vr19fjz76qHbsuDIta+/evXruueczLV4FACh4tWrV0qJFixUfH6/jx49r6NChmS63i42N1UMPPaQtW7YoKSlJR48e0/79B1S9ejVJUuXKlXXkyOHr7kzwbxcvXtRLL/2fLl68qP379+vVV8fqgQceUOnSgZKk225rpXXr1mnLli26fPmyFi1adN3q5lmpU6eOevXqpdGjn9bmzZt1+fJlbdjwiyZMmJDrYxAWVl1Wq1VfffWVkpOT9dtvv123gvuSJUs0YcIEHT58WGazWb///ocuX76s6tWrqU6dOrrnnnv07LPPad269UpKStKRI0f0zjvvaP78+bmOAygK8us87c47eyo4OFjPPvucYmJidOTIEb3wwosOt+nZs4cqVaqkUaNG6ezZszp+/LhGjXpKLVq0UKtWrZxqR9u2bXX27Fn98ssvatu2raQrX8oXLvz6htcRyE3bctuWgowTyCuSB4WkWbNmatiwoYYNG6Zq1arrkUcG6LHHhmjgwCsrqz755JMym9NUo0bNbG/VKEkDBw7U1q1/qnHjJnrkkQF65JFHNHz4cPvrzzzztJo1a6YOHTrottta6+LFi+rSpUuO8Xl4eGjFiuWqVKmSunfvoYiIRvrhhx/Up89duW5jp06dNHr0KA0ePETVq4dpypQpGjlyRKYyAwY8oqioKHXv3l01atTUF198oXnzvlLFihVlMpm0YsUKNW7cWP3791dYWA0NH/6E6tevn+UdLQAABeeTTz7WiRMnVLt2HXXu3EV169ZVeHhD++tBQUF64IEHNHbsONWqVVvdunVTs2bNNHbsWEnSfff1U/36DXTLLY2zvVWjJHXo0EFGo1Ft2rRVp06d1bjxLXrrrYn21/v27atBgwbpwQcfUqNGt+iHH37UwIEDc9WGWbNmqmvXLho48FHVrl1HkyZN0v3335/rY1C1alVNnTpF77zzjqpUqapnnnlWI0Y8manMlXvKe6l//wdUtWo1vfjii3r77bd1++23S5KmT5+mBx98QE8//bSqVauu/v0fkNVqVe/evXMdB1AU5Nd5mslk0tdfL9SxY8fUoEFD9e//gAYPHpxp7ZOstlm8eLFSU1PVuHETtW7dRsHBZfTll1843Y7Q0FBVr15doaEVVbVqVUlXfuW3Wq03fClAbtqW27YUZJxAXhkslvTsLzBCkdKzZ0+1bNnqupWiAQC4GT3//H90+vQpfoUHAOAmwMwDAAAAAADgEMkDAAAAAADgEJctAAAAAAAAh5h5AAAAAAAAHCJ5AAAAAAAAHCJ5AAAAAAAAHCJ5AAAAAAAAHCJ5AAAAAAAAHCJ5AAAAAAAAHCJ5AAAAAAAAHCJ5AAAAAAAAHCJ5AAAAAAAAHCJ5AAAAAAAAHCJ5AAAAAAAAHCJ5UEKkp6Xp9Km/lZ6W5upQAADIN4xvAJxBnwHkHcmDEiI9PV1nzpxSenq6q0MBACDfML4BcAZ9BpB3JA8AAAAAAIBDBosl3ebqIFDwrFarzOZUeXp6yWgkZwQAKB4Y3wA4gz4DyDuSBwAAAAAAwCHSbQAAAAAAwCGSBwAAAAAAwCGSByWE2ZyqqKhDMptTXR0KAAD5hvENgDPoM4C8I3lQQmRYMhQfF6sMS4arQwEAIN8wvgFwBn0GkHckDwAAAAAAgEMkDwAAAAAAgEMkD0oIk8mkkJBQmUwmV4cCAEC+YXwD4Az6DCDvDBZLus3VQQAAAAAAgKKLmQcAAAAAAMAhkgcAAAAAAMAhkgclhMViUVzsRVksFleHAgBAvmF8A+AM+gwg70gelBBpZrOiow8rzWx2dSgAAOQbxjcAzqDPAPKO5AEAAAAAAHCI5AEAAAAAAHCI5EEJYTAa5OXlLYPR4OpQAADIN4xvAJxBnwHkncFiSbe5OggAAAAAAFB0MfMAAAAAAAA4RPIAAAAAAG4iGTar4tKSXR0GShiSByWAzWZz+BjFG+8/ChOfNxQmPm8ASqqDl86p6U/vujoMlDAkD0oAg8GghLQUHUu6qIS0FBkMLBBTkvD+ozDxeUNh4vMGwFnJSUnavm2zkpOSXB0KcNNxd3UAKFibYqL0WfRmbTh/WFbZZJRB7cvV0pCwlmpTtoarw0MB4/1HYeLzhsLE5w0AgMLF3RaKsU8Ob9TbB9dk+/pL9TrriZptCjEiFCbefxQmPm8oTHzeAORVclKSDhzYo3r1wuXj65uvdSekpcjH3UMmo5tSM9Ll5WbK9HpcWrL8TV5yM1yZ/J1hs+pSeqpKe/hct32yJU0mo5tMRjdJUprVIpPBzT7Dal/CGfXaOEPRd46TpOv2Z7PZ7GsieLmZ5OPukWXMFmuGbJJ9P/9mzrDIw+iW5cyunLZF8cPMg2JqU0yUwxMrSXrrwM8KDwjhF5piiPcfhYnPGwoTnzcARVWXXz5R79CGWn3mgM6nJirE21/vNuqjW4OrSpJa/Py+vm3zmBoEhEi6sm7BPxMAXX75RHeFhuvHswcUn5aiGc37y2KzasK+HxV9+aK83Ux6uk57PVq9hX2fnxzeqNlHNyshPUUNAkI0vdn9Ku/lr4T0FHVYP1WSlGpNVxnPUhrfsIc6lK8tSTqfmqindizRjri/5Wl0V9uyNfRm+J0K8rySUPnjwlGN3/eDjiZdlJvBqHsqNdIr9bvK0809x21RfLHmQTFzbbGoz6I356r87KObM22HmxvvPwoTnzcUJj5vAPKDl7e36jeIkJe3d4HUv/rMAc1rOUAHeozRA1Wa6vE/FyghPSXX268/f1jzWw7U7m4vqaxnKQ3aMk8PVW2mvd3/T790HKVTKfH2shk2q86mXtJvHZ9RZNeX5Gl016dHNkmSAj18tKvbi9rV7UUd6D5Gr9Tvqqd3fqP4q7MRZkX/IU+ju/Z0+z/t6vqi+oRGaGf8KUnSiaRYDdk6X0/UbKP93V/Wrx1Ga0/CaU2L2pTjtijemHlQzFxbPGrD+cO5Kr/+3GElpKUowMNbx45G6dKlhAKOEAXFPyBQ1aqF8f6jUPB5Q2Hi8wYgv5lMJtWrH57v9T4W1kpVfYMkScNqtNbnR7doU0y0elZskOvtQ30CJUnfnoxUk9KV9XC15pKkAKO3xtTvmqn8f+p2lKfbla90XSvU1Zpzh66rM9Fi1q1BVVTR21/b406qY/naCvLwUWpGuuLTUlTB219dQ+rZyy89uVu3lK6kNmXCFJ+WIoOkh6s212fRf2h07fYOt0XxRvKgGIpLT5ZVufulxSqb4tOvnFzZZFN6eloBR4eCYrNZJfH+o3DweUNh4vMG4GZR0TvA/n+DwaAKXv46n5qY6+3Le/nZ/3829ZIqX00kZMXNYJSfycv+2MPoLnOGRZKUZDHrpciV+vnsQRllkIfRXYmWVJ1PvSTpSpLCy82k53ctU1x6spoHVdGoWu0U5Omr0ykJ2hZ7wn7ZwzUBHl45bovijeRBMVTa5COjDLk6wTLKoEDTlWlbBhlkMmW9mAqKPsPVxXd4/1EY+LyhMPF5A5DfTCZTzoXy4EhijLpUqCtJSrdm6ERynH0mgbebSckZ6fayZ3NIKlT1DdKPZw5kes5ms+XqtrRzjm7R38lx+r3jM/Yv9Z3WT1XG1Uu53AxGPVq9hR6t3kLp1gw9v2uZxu/7QR816auqvkFqGVxNX7R85Lp957QtijeSB8WMzWZTgIe32perpXXnr5+29G93lK915VcZm03VqrOw1M2O9x+Fic8bChOfNwD5wWxO1cmTJ1SpUpUCqX9W9B+q519BYaWCNSPqN3m5uatd2ZqSpEaBofos+g+V9/JTTOplvXPA8eKv91dpohlRv2v83tV6qFpzXbaY9fHhjZrRvH+OcdhsNqVY0hVjvqzLFrMWntihw5dj7K+P3fu9avmVVZsyYbLabIoxX1aIl78kqX+VJvos+g99+Nd63RUaLqvNpk0XonUs6aJea9jD4bYo3lgwsZi5lokcEtYyV+WHVG+VaTvc3Hj/UZj4vKEw8XkDkB8yLBmKj4tVhiWjQOofUautPj+2RQ/+MVenUhI0r+VA+5oEE8J7KiUjXf1/n6MP/1qvYTVaK+jqbRolKdDknem2h2U9S2lZm8d0NjVRg7bM02t7V+vhqs0kSe4GY6ZtpSu3ZPS/ehnDkLBWahgQokFb5+mRLV9KkjqUq22/nePTtdvrSGKMhmydr8f+XKDKPoF6tcGV9RTKeJbS8jaP63hSrB7bukDDt32tY0kXNapWuxy3RfFmsFjSWYa4mOI+2CUb7z8KE583FCY+bwDyKjkpSQcO7FG9euHy8c3fa/Rb/Py+3m7UW+3L1crXeoGigssWirEna7VVRGBFzT66WevPHZZVNhll0B3la2lw9Zbc/7qY4/1HYeLzhsLE5w0AgMLHzIMSIiEtRfHpKQo0eSvAo2Dua4uii/cfhYnPGwoTnzcAzkhPS1NMzDmVLVteJo/8XUi164ZPNLZhN7UuE5av9QJFBcmDEuDfq7LmdpVWFA+8/yhMfN5QmPi8AQBQeEgeAAAAAAAAh7jbAgAAAAAAcIjkQQlhsVgUF3tRFovF1aEAAJBvGN8AOIM+A8g7kgclRJrZrOjow0ozm10dCgAA+YbxDYAz6DOAvCN5AAAAAAAAHCJ5AAAAAAAAHCJ5UEIYjAZ5eXnLYOQWVgCA4oPxDYAz6DOAvONWjQAAAAAAwCFmHgAAAAAAAIdIHgAAAAAAAIdIHgAAAAAAAIdIHpQQyUlJ2r5ts5KTklwdCgAA+YbxDYAz6DOAvCN5AAAAAAAAHCJ5AAAAAAAAHCJ5AAAAAAAAHDJYLOk2VweBgme1WmU2p8rT00tGIzkjAEDxwPgGwBn0GUDekTwAAAAAAAAOkW4DAAAAAAAOkTwAAAAAAAAOkTwoIczmVEVFHZLZnOrqUAAAyDeMbwCcQZ8B5B3JgxIiw5Kh+LhYZVgyXB0KAAD5hvENgDPoM4C8I3kAAAAAAAAcInkAAAAAAAAcInlQQphMJoWEhMpkMrk6FAAA8g3jGwBn0GcAeWewWNJtrg4CAAAAAAAUXcw8AAAAAAAADpE8AAAAAAAADpE8KCEsFoviYi/KYrG4OhQAAPIN4xsAZ9BnAHlH8qCESDObFR19WGlms6tDAQAg3zC+AXAGfQaQdyQPAAAAAACAQyQPAAAAAACAQyQPSgiD0SAvL28ZjAZXhwIAQL5hfAPgDPoMIO8MFku6zdVBAAAAAACAoouZBwAAAAAAwCGSBwAAAAAAwCGSBwAAAAAAwCGSByVEclKStm/brOSkJFeHAgBAvmF8A+AM+gwg70geAAAAAAAAh0geAAAAAAAAh0geAAAAAAAAhwwWS7rN1UGg4FmtVpnNqfL09JLRSM4IAFA8ML4BcAZ9BpB3/MWUEEajUd7ePnSSwE3o/CWLWk+I0tGYtELZ34PTTmjR1vhC2Rdwoxjfip7kJR/q0tuDXLb/lJXTlfDG/dk+RslGnwHkHX81AFDETf75gu6o56vqZT3sz/X84Jg+WXuxQPY3omOwPl4Tq8upGQVSP4DiIXn5J0p48yFXhwEAKCQkDwCgCDt/yaLvdyeq360BhbbP22r6yMfToOU7EwttnwCQX7x7DVPAq1+7OgwAKHbcXR0ACofZnKqTJ0+oUqUq8vT0cnU4QJbeXhWj7cdS1LlBKS3amqCYRItmDgpV46re+uzXWK3YmagzCekq5+euR1qX1kOtAu3bpmfYNOfXOK3cdUlnEiyqHGTS4Lal1auxvyTJkmHLsY5O7xxVl4alFJ+Uoa3RyUpJt6l5dW+90rucyvi552pfb6w4r53HU/TNU1UztW3OxjjN+iVWa16oLm+P3Odtv9t1SWX83NWosrf9ud4fHdPfsematj5W09bHSpJmPBqqljV9HNa1/3SqBsw4qWe7ltGDV9t98bJF9318Qu3q+Gpsn/KSJIPBoM4N/LR8x6VMxwcoihjfXCNl5XSZf/5SkhQ3spUkyaN1H/k+8OKVArYMpf78lcybv5M17pzcylaW9z2jZKrTLFM96ft+V8oPc5Rx8rAM7h4yRbSVd9+nZfTxy3bfGaeOKPnbqco4sV8yGOUeFiHvu56UW4Vq9tjSdq1zmEDIab/JSz6U5chOeTTtkqs2pP70hSx/H5LBy1ceTTrIu9cwGTx98txGFBz6DCDvmHlQQmRYMhQfF6sMC9OQUbQdPGPWX2fN+mJoJe16o5aah/lo7Dfn9MVv8Xq2WxltfLmGXr2rnD5Ze1HLd1yybzdmyVkt3HKlzIaXquu9/hX0x5FkJV6dep+bOiRp4eZ43RrmrWWjq2r+8Mo6fjFdb6+KyVTG0b4eaBGgI+fStONYir28zWbT0m0J6tXY36nEgSRtPZqiiMqZT25WPF1NlYNMGn5HkCIn1FLkhFo5Jg4kqX5FL73Us6ze/+GCIv9OVYbVphe+PqvgUu56sWfZTGVvqeKlv86aFZ9Mn4GijfHNNbx7DZNn50dkDAlT6al/qPTUP/6XOJCUcfKwMs5EyW/kZAW+uVJu1Roo6bOXZUtJspdJ2/2LLk97Xu41Gytg/Dfye3GOMmJOKvmL1x3u+/LMl2QMKCP/Vxcq4LUl8mzTR+bfluc69tzuN1dt2LlOl6f9R+51b1XA2K/l/9JcGYMrKn3v7zfURhQc+gwg70geAChSvD0MGn93eYUEmiRJ0efT9N3uRI3oGKyO9UvJ19Oo1rV8NahtaX267so1/4fOmvXDnst66c6yuqNeKZXyclONcp76b78K8vNyy1Ud13SP8FOfpgEq5eWmqmU81LeZv/48+r9EQE77qlneU82qeWvR1gT7Nn8cSdaJi+m6r7nzlx4ci0lTxavHIj/c2zxA3cJL6T8Lz+id72N08IxZ7/evIE9T5uEgtLRJNpt07ELhLNIIoHgxePvJ58GXZSxdTgbvUvLuPki25ERZTh22l0lZOV3uNRrJ564nZfQLkluZUPk+/IrS9/4my4mDWdZrs1pljT0rU71bZfQLksG7lEwNbpNP39G5ji23+81tG0yNO8i7+2AZA8vK6B8krzvul0fTTnluIwAUVSQPABQpoaVN8vH8X9e052SqJOn2Or6ZyjWu6q3T8RZdSsnQ/lNXyrQIy/rX99zUcU21Mh6ZygT6uCnhH6/ntC9J6t8yQD/vu6zYJIskadHWBDWv7q2wch7ZbpOdxFSrSnnmb1f9Su9y8nA3aMHmBL1+d3lVDr4+Lt+r+0xMsebrvgGUDMbgEBnc/5f4NPheSZ7akq4kVm3mZFnPHpWpYetM27mVqyyDX2ll/P1XlvUajEZ5tr9PSfMnKmnuazL/tlwZF07nOi5n9ptjG1KTZD1/QqbaTW94XwBwM2DNgxLCZDIpJCRUJlP+/YIJFAQPN0Omx1abTZLU44NjWZY/m2CRLYc6c1OHv7ebJMloyLKIXU77kqQO9UqptO8FLdt+ST0a+emXv5L0Vr8Kudjyen5eRl025+8X+BMX03XukkVuRmV7+8ekq/v08ybHjKKN8a2IMuTQd1iv9KYpyz5WyrKPr385/ny2m/rcM0qerXop/eAWpe/7XcmLP5Bn67vk0+/ZnONyZr85teHagGDIZuC4gTai4NBnAHlH8qCEMHl4qGJoZVeHATitQeiV6/1XPVM1y1/I/2nr0WR1bnD9AlTO1JHbeLLblyS5uxnUr7m/Fv+ZoOQ0mwJ93NShfqk87a96WQ+dikvPYh/281KnXE7N0PMLz6htbV91qOerV745p1uqeKn5v2ZSnIpLl8EgVb3B4wUUNMY31zG4uUnWvF03bvD2lbFcFXm2ulNenR9xenu3kOpyC6ku3dFfaTvWKGn2q/Lq+qiM/kEFut+s6ko/tF2ere8q0H0h/9BnAHnHT0oAirTaFTx1ZyM/vbjorHadSFFymlUXEi36eV+inltwxl6mW3gpvf1djH45eFlJZquiz6fp5SVnlZiakas6nInH0b6uubd5gM5fytCcjXG6p6m/TG45TGnIRosaPtp9IuW65yuVNmnfqVQlpzk3K2Hct+dlk/T63eXU8xZ/3dPUXy8uOqsLiZZM5XadSFXt8h4q7euWp7gBFH/G4FBZ4847ddnAP3nf9YRSfpwr89YfZL2cIJs5WZaje5X0+ThlnDqS5TbW+PO6PONFpR/ZKVtKkqyXE2Q5vFMG3wAZfP0LbL/Z1tVrmNJ3rFXKD5/LmnBB1sRYpW5YpLTta/J9XwDgasw8AFDkvdG3vL74LV6vfXtef8emq7Svm26p4qWh7f/3C9Ob91bQrF9i9daqGMUkZqhKsEmD2pSWn5dbruvIrZz2JUnBpdzVsb6vftp7WffmYaHEa3o28tPkny9o14kU3VLlf7drHNEpWG8sP6/2E6OVmm7L1a0av/o9Tr/+laQvh1VWqauxvtizrPaeMuvFRWc1Y1Co3IwG2Ww2rdl3WQ+2ynvcAIo/j2adlX5gsxLfGSRb8qXMt2rMzfaN2svg6aPUn75Q8tfvSgaD3CrWkNftfWUMCctyG2NgOXm07KnUVbNk+fsvGdzc5Va1vkqNnCSDW+5Oa/Oy32zratxBGuqh1B/nKvXHOTL4BsijcQd5troz3/cFAK5msFjS8zDxFTcbi8WixEsJ8vMPkLs7OSOgMAz57KR8PI2a8nDFG6rn1aVnZbVdSVoUht8PJ+mlxWe16tlqmRIiQFHE+AbAGfQZQN5x2UIJkWY2Kzr6sNLMZleHAhR7NptN6w9c1p9HU/TY7aVvuL6nOpfRhoNJ2S5umN8+XhurER2DSRzgpsD4BsAZ9BlA3pFuA4B8dCklQ23ejJafl1FPdwlWo39capBX5fzd9dsrNfIhutyZN5yFpAAAAJAZyQMAyEf+3m6KnFCr0Pe7dFuCXl+W/W2/xvUpp77NWMMAAAAAeUPyoIQwGA3y8vKWIaeb2AO4KfVtFkByACUS4xsAZ9BnAHnHgokAAAAAAMAhFkwEAAAAAAAOkTwAAAAAAAAOkTwAAAAAAAAOkTwoIZKTkrR922YlJyW5OhQAAPIN4xsAZ9BnAHlH8gAAAAAAADhE8gAAAAAAADhE8gAAAAAAADhksFjSba4OAgXParXKbE6Vp6eXjEZyRgCA4oHxDYAz6DOAvCN5AAAAAAAAHCLdhiwlm61Ks1hdHQYAoIRJSbPKnF4w448t3SxbWmqB1A3ANWzpabKZk10dBlAiuLs6ABQt+0+l6tP1sfrlYJLqhnhq/vDKcnczuDosAEAJcCHRov6f/q2Lly3qfYu/Hm8fpEpBphuuN+PCKaX++LnStqyWwT9Y/i/MltE/OB8iBuBKtgyLEt9/XBknD8nUsI28egyRe5W6rg4LKLZIHpQQZnOqTp48oUqVqsjT0+u61/+ZNLjmyHmzzBYbyQMAQKE4n2jR+UsWSdK3Oy5pxa5LOSYRHI1v/0wayJohSbLFn5c14QLJA6A4SDMr43SUJCl97yal792UYxIhp3NiANljzYMSIjkpSQcO7FG9euHy8fW1P785Kknvr76gv86mZSrv721Ut3A/1avoWdihAgBKsM1RydpwIElmy/9OTwwGqUlVL73au5zCymUel7Ia3yxnjir563eUEbVbsv3jNMfkKVN4W5nqNi+UtgAoeJa//1L69p9lS07M9LxbaC153z1KprrNMj2f3TkxgJyRPCghsuoo45Isaj/xqPgAAABuBl4mg7aMrSGD4X8z4v49vtlsNsU/e4eUbnZhpACKBoMCJq6S0a+0/RmSB0DesWAiAAAAAABwiJkHJUR6WppiYs6pbNnyMnl42J/nsgUAQFGS3WULTat66ZUsLlvIany7ctnCu8qI2sVlC0Axl+1lC5WuXrZQJ/NlC9mdEwPIGckDSLqyYOK09bHa8I8FE92N0sYxNeTryQQVAEDB2386Vf0/+dv+2M0o9W7sr8fb5e2uC1cWTJyrtC3f2xdMlCS/Fz+Xe+U6+RIzANexpSQp/sWumf6+TeFt5dV9MHddAAoAd1uAJKl+qJcmP1wxUxKhVgVPebpzpwUAQOEo5+eu8v7uunDZckNJg2vcyoTK96GX5dV1oD2JYPAPljGgTD5GDcBlPDzlFlpTGX//RdIAKATMPECWktOsMhkNMpE8AAAUopQ0q4wGydOU/7PebOlmyWaTwYPbswHFhS09TbJmyODp7epQgGKP5EEJYbFYlHgpQX7+AXJ3Z8IJAKB4YHwD4Az6DCDvuJi9hEgzmxUdfVhpZm5dBQAoPhjfADiDPgPIO5IHAAAAAADAIZIHAAAAAADAIZIHJYTBaJCXl7cMRhZABAAUH4xvAJxBnwHkHQsmAgAAAAAAh5h5AAAAAAAAHCJ5AAAAAAAAHCJ5AAAAAAAAHCJ5UEIkJyVp+7bNSk5KcnUoAADkG8Y3AM6gzwDyjuQBAAAAAABwiOQBAAAAAABwiOQBAAAAAABwyGCxpNtcHQQKntVqldmcKk9PLxmN5IwAAMUD4xsAZ9BnAHlH8gAAAAAAADhEug0AAAAAADhE8gAAAAAAADhE8qCEMJtTFRV1SGZzqqtDAQAg3zC+AXAGfQaQdyQPSogMS4bi42KVYclwdSgohvbt26shgx912f4PHTqkRx5+KNflXR3vv/07fmfbA5RkN+P4NnnyJM376ktXh1EkTfv0E30+Z3a2j4EbdTP2GUBRQfIAAP6FL+8AAABAZu6uDgAADh44oLfe+q8+n1s0fomrXbu2vvxqnv1xUYsPQMn28dQp8vP314ABA7Mt88UXc/XdyhWSJHd3dwUHl1GHjh119933ZFnmmqCgIE2bPtP+Wo8ePfXooMH211NSUjRs6GNKTU3V1I8/Vbly5fK5dcir1JQUDRjwcLavm0wmzZu/ME91/+f559Srd2/dfnu7vIYHoBggeVBCmEwmhYSEymQyuToUAADyDeNb9sLDw/Xq2NdkSU9XZGSk3nvvHVWoUEGtWt12XZmsVK5cRb/++oseePAheXp6SpJ+/fUXlS9fXsePHy+MJsAJXt7eWrR4qf3xO++8pcDA0ho6dJgLoyp66DOAvCN5UEKYPDxUMbSyq8PATWTsq6+oTZs26tK1m6wZGRoyZJDq1auvF158SZL00UcfqHy58nrgwf9N7/9+1XdavXq1Ll1KUEREIz05YqS8vb119swZjRo1UpLk6empatWqa9CgwQqrUUPxcXEaO/YVSdJ9/fpKkkY//Yxat25zXUx//XVQ8+fN0/HjxxQYGKiHHx6gZs2bX1du9KiRuvuevmrf/g77c3Nmf6aLFy/q+f+8cF35Awf266MPP9CAgY+qdes2OnTokN4Y/7q+/GpevsTnqP25ER8Xp+nTp+nAgf3y8PBQ69Zt9MCDD8nDw+O6svv27dUH77+nu+++Rz/99KPM5jRNnzFTa9b8rO9XfaeYmAuqWDFEDz38iCIiGkmSbDab1q75WatXr1ZMzHnVqFFTQ4cNU0hIRUlyuO28r77UqdOn5FfKT9u3b5PBYFS37t3Vt++9Dut2M7pp9OhRmjJlqsqULWsvO3rUSN3V52517NgpV8cGKArj2+lTp/T8889q9py58vLy0u+//aaPPvpAL7z4kpo1a67k5GQNGfyoJk2aonLly0uSLl26pA/ef0979uyRp6eH7u13nzp16izpyt/V8uXLJEmlS5dWs2bN9ciAgfLy8tLSpUv0yy8bJEnfrVwhT0+vTDOlsuJuMqlJ06YKCQnRsaNHMyUPHClfobyCgoL022+b1KFDR0nSzz/9qK7dumvmjOnZbpeRkaGVK1dow/p1iouLU4MGDfX40GEqXbq0w7blxtatW/T1wgU6d+6cQkJC1P+BB9W0abMsy06ePEnu7u5KTk7Sgf371blzF/Xoeae++vIL7d69W+npaWrU6BYNGjxE/v7+kqSEhATN++pLRUbulsVi0R13dND9/R+Qu7u7Ll265HDbIYMfVe+7+mjrls06ceJvhYSEaOiw4apZs6bDupcsXqTo6Gi9POYVe+z79u3V229N1PQZs/T996u08ddf9NGkKbk6Rvnt/ffe1fHjxzR1ymRNnTJZderU1RsT3sy2/MWLF/XCf57X4CFD7GPl1KlTFBd7UWNeGSuj0bVXTReFPgO4WbHmAYAshUdEKDIyUpIUFR2lUqVK6cCB/crIyJDNZtO+vXsVHhFhL5+YmKhTp07pvxPf0vvvf6jTp09p9ervJUkVQkK0aPFSLVq8VDNmfqYWLVro3XffUXp6ugJLl9b48RPk4+NjL5PVF/NTp05q/Ouvq0WLFpr68ad64YWXtGPH9ixjb9f+Dm1Yv97+2GKxaNOmTWp/xx3Xld20caM++vADPf3Ms1nuNz/ic9T+3FiyZLFsNpsmT56q997/UOXKl9eBA/uzLZ+YmKjjx49rwoT/Xkkc/PyTvlm6REOHPaFZn81W37799P577+r8uXOSrnwZ+PrrrzVg4EDNmDFLfe+9V3/88Yck5bitJG3ftl0NGjbU1I8/1TPPPqulSxYrOirKYd3lypdXRES41qxdY69nz55IxcfHq/VtrXN1XICiomJoqAICArV//z5JUuSeSIWEhNj70H379qpMmTL2xIEk/f77b2p/xx2aNn2GHh00WJ/NmqXYixclSQ89/IgWLV6qrxct0fg3JujUqVP65psrvyj37Xuv2rVrrzt79daixUtzTBxIkiU9XTu2b9fp02dUq3Ztp9rWpWtX/fzTj5KuJFrN5jQ1upo8zM6irxdq3do1euLJEZo2faY6dOiorVu35Ni2nGRkZGjK5Enq1fsuzZ4zV6OfflZbt2xxuM2mjb/q9rbtNG36TN3f/wG9/dZEWa1WTXzrbU2aPFVubm76eOqVL+XWjAy9/dZEXU66rPHjJ+iDDyfJt1QpHTt6VDabzeG2/9zf8Cee1MyZs1SnTh3NnDEtx7o7de6iPXv2ZOpXf/7pJ93Wuo28vb1zdWwK0nPP/0dVq1bTyKdGadHipQ4TB5IUHBysp0aN0ozp03X69GmtW7tGeyJ3a9ToZ1yeOABwY/gLBpCl8PAI7du3V1arVXsiI3XrrS1UvnwFHT58WMePH1dKSorq1KlrL+/m5qZHBw2Wn5+fypQtqxYtWurY0aPX1evt7a07e/WWm5tRR7N4PTtrfv5Z4eHh6tHzTpUqVUoVQ0M1dNjwLMu2a9deBw8e0LlzZyVJ27dvk5ubUY0bN8lU7pulS/T11ws1btzrqlevfq5juZH48tJ+k8kkDw8PeXl5yd/fX92791CjRrdkW95gMGjwkMfkHxAgSVq9+nvd2+8+1a1bV56enrq1RQtFRDTS5s1/XH19tfr27atGjW6Rl7e3GjYM1z339M3VttKVac+3395OXl5eql+/gapVq65jx47mWHeXrt20ft1aZWRcWfH6559+UuvWbeRVBE6WAWeFh4drz9Vkwd49kbr//ge0J3K3JGlPZKTCwyMylW91W2s1adJUnp6eatmylfz9/XXi7xOZyhgMBpUvX0H33ttP27f96XRMe/bs0X39+urBB/vrrbf+q4cefljNmjXPssy1f0uXLsn0etMmTZWQkKAjR47opx9/VOcuXSSDweF+f/jhBz38yADVqVNX3t7eata8ubp27XbDbTNIcnc3ycfbRx4eHqpUqZKeeHKEw21ubdFCt7ZoIZPJdHX8Oqahw4YrKChIfn5+enTQYO3cuUOXL1/W4SOHdezYUY0cOUrlypeXv7+/+vS5WzVr1cpx22vuvqevKleuIi9vb3Xu3EUnTpyQNSPDYd1lypRR4yaNtWbNz5KuzFDYunWLOl2dgdW3770um3WQV7fc0ljdu3fX229N1Ny5n2v06GcUcHVMAnDz4rKFEsJisSjxUoL8/APk7s7bjpzVqlVLVqtV0VFR2rMnUnf1uVtGo1GRkbvl7e2tevXqZbpe0MfHJ9NjDw8PpaWlSbqyiNPcuZ9rx44dSkiIl9VqlST7r2y5ERMTo4oVK+aqbHBwsMLDI7Rh/Xrd3/8BrV+/Tre3ay83Nzd7mdTUVK1cuUJ97+2niqGhuY4jL/HdaPvvu7+/vv1mqcaNGytPT081btJE3bt1l8fVa5D/zdfXN9OvVefPn9e0Tz/RtE8/yVQuIDDwauznVbFi1scgp20l2ZMU1/zzvXdUd+PGTeTubtKff25V3Tp1tW3bn3pjwn+zLAtkp6iMb+EREVq+7FudPXtWGRkZanXbbZo9e5YuXryoPXsidd/9/TOVD7g61f2af/7dHDx4UAvmz9OJE8eVlJQk6Uof63RMV9czSE1J0ZKlS7RyxQq1a9defn5+15XJjtHNTZ06ddaSJYt08MABDRnymJJTUrItfzkxUSkpydn+3d9I24xubhrzyqtasXyZli5drPIVKqhTx86KaJT9TIgyZcra/3/+/DmlpaXp4YceuK5cTMx5xZyPUWBgYJbx5LRtqVKlJCnTF2QPDw9lZGTIkpHhsG5J6tq1m6ZOmaz77u+v9evWKjQ0VDVr1cr+YOSTDevX6ZNPPrY//uKLr/ItgdulazctX75c9erVU/0GDfKlzvxQVPoM4GbEX0wJkWY2Kzr6sOrVC6ejRK64ubmpXv362rZ9m6Kjj6pevfpyc3PToq8Xytvb57pf0RxZtuxbnTx5Uq+Pf0PBwcEymUwa8eRwZViv/OJsMDr+FUuSypYtq9OnT+d6n3fc0UHz5n2pLl27afeuXXrkkQGZXvfy8tIrr47Vf998UyaT6bpfxf7pRuPLqf058fb21oMPXVlBOzExUR99+IEuJSTokWxWWjf861fBMmXKqu+996pNm7bZxF5Op0+fyvIEPKdtc+KobqPRqM6dO+vnn37SqVOnVKlSZfu1wUBuFZXxLTw8QlOnTNGmjb8qPKKRDAaDGoaH65cN63X69Bk1bBie67ref+9d9erVS//5zwvy8fXV3r179O4779hfNzg59dvL21sPPviQInfv1uLFizR48BCntu/QsZOWLFms1q3bqJSfn8PkQSk/P3l7++j06VOqVKnSda/n1Lac1KxZU88+97wk6dBff2n8+Nf01lvvqFLlrK9h/2dvWLZsWfn4+Oiz2Z9nSiZfk5aWpvj4eCUnJ1/3JT+nbXNStlzZbOuWpIiIRvLx8dGWLZu1du0a9ex5p9P7yIv2d3RQ+zs65FjOmItx8J+sVqumTpms5s2b68CBA/r1lw26vV37PEaZv4pKnwHcjLhsAUC2IsIjtPr77xUWFiZPT0/VrVtPJ06c0IED+zOtd5CT1NRUmUwm+fj4KC0tTYsXL1JMTIz99dKBpZWSkqqzZ85kW0fHTp21Z0+kvl/1nS5fvqzTp09rxvRp2ZZvfuutSk5O0bRPP1aNGjUVGnr9SWz16mEa99rr+vabpVq27Nts67rR+HJqf05mz/5Mf/65VclJSfY1J1JTU3O9ffcePbRwwXzt379PZrNZZ8+c0cIF87Vr505JUtdu3bR06VJF7t6t1JQU7du3134Nck7b5sRR3ZLUoUNHHTx4UKu/X6WOnVgkETevwMBAVapUSStXrlTE1eRqRHiEli9frmrVqmX6td8Rq9WqtDSzvLy85OnlpdOnT+vrhQsylSkdGKhTJ/+2z1TIDaPRqPvuv19r1/zs1Kwv6Urb5i/4WiNGPpWr8l27dtVXX36hQ3/9pdSUFG3b9qd+/PGHXLXNkQsxMZo6dYqOHz+m9PR0WW02ZWRYc30catWspdDQUE2f9qliYs4rNTVV+/fv0wfvv2d/vUqVqpo6dbLOnzunxMRELV++TEcOH85x29zsO7u6pStJ386du+jzObMVFxentv+4JeLSpUv09OjcHfuCEhhYWieOH7dfZpaTxYu+VkJCgkaMfEqjRo/WZ5/N0smTJws4SgAFjXQbgGxFRDTS55/PsScKTCaTatWqraNHo1WtWvVc13NXn7s1dcokPfnEMHl6eqpt29tV+R+/EpUrX17de/TQyy+/pMuXL2d5N4NKlSppzCuvasH8+fr664UKCgrWQw87vp916zZt9NOPP2jY8CeyLVepUiWNHz9Bb7zxulJTUtT/gQevK3Oj8eXU/px07dpVCxbM1ycfT5XRaFR4eITu73/91NnsdOnSVe7u7poze7bOnj2roKAg3XHHHWrQsOHV+q/MupgzZ7YuXLig2rVr6fHHh+Vq25xjz75u6colDy1attTWLZt1e9vbc90moCiKiIjQ99+vUsPw8KuPGyklJdmpZKvRaNQTT4zQvHlfas6cOapQoYJat26tEyf+tpfp2KmzPvrwAw16dKCMRmOuFk2UpGbNmqtKlSpatuxbDR7ymHONc8J99/eXp5eXpk6drISEBDVsGK7HHh+aq7Y5ElymjCLCwzV1yhSdPn1awcFBGjBwYK7vXGN0c9OLL/6fFi5coHFjX1VSUpLCwsLU77777a+/9NL/6csvv9Arr7wsm82m9u3vULXq1XPcNjf7zq7ua9rf0UELFy5Qy1a3ydfXN1f1FpY+d9+t6dM+1apV36lmzVoOF02M3L1b33//vf47caI8PT0VHh6hnnf20ocfvKf/TnzbfttPADcfg8WSbnN1ECh4KSnJio46rLAateTt7fx1kwBQUN577x15e3nn+ldN4J8Y31BcJCQkaPiwoRo37jXVrVfP1eEUW/QZQN5x2UIJ4e3towYNG9FJAihSjh6N1vZt29W9R09Xh4KbFOMbigOr1arly5epSpXKJA4KGH0GkHdctgAAcIl33nlLeyIjdVefPgoLC3N1OADgEpcuXdJjQwapbNmyeubZ510dTo7OnjmjUaNGZvlahw4dNfyJJws5IgCFhcsWAAAAAACAQ1y2AAAAAAAAHCJ5UEIkJyVp+7bNSk5KcnUoAADkG8Y3AM6gzwDyjuQBAAAAAABwiOQBAAAAAABwiOQBAAAAAABwiLstlBBWq1Vmc6o8Pb1kNJIzAgAUD4xvAJxBnwHkHckDAAAAAADgEOk2AAAAAADgEMkDAAAAAADgEMmDEsJsTlVU1CGZzamuDgUAgHzD+AbAGfQZQN6RPCghMiwZio+LVYYlw9WhAACQbxjfADiDPgPIO5IHAAAAAADAIZIHAAAAAADAIZIHJYTJZFJISKhMJpOrQwEAIN8wvgFwBn0GkHcGiyXd5uogAAAAAABA0cXMAwAAAAAA4BDJAwAAAAAA4BDJgxLCYrEoLvaiLBaLq0MBACDfML4BcAZ9BpB3JA9KiDSzWdHRh5VmNrs6FAAA8g3jGwBn0GcAeUfyAAAAAAAAOETyAAAAAAAAOETyoIQwGA3y8vKWwWhwdSgAAOQbxjcAzqDPAPLOYLGk21wdBAAAAAAAKLqYeQAAAAAAABwieQAAAAAAABwieQAAAAAAABwieVBCJCclafu2zUpOSnJ1KAAA5BvGNwDOoM8A8o7kAQAAAAAAcIjkAQAAAAAAcIjkAQAAAAAAcMhgsaTbXB0ECp7VapXZnCpPTy8ZjeSMAADFA+MbAGfQZwB5R/IAAAAAAAA4RLoNAAAAAAA4RPIAAAAAAAA4RPKghDCbUxUVdUhmc6qrQwEAIN8wvgFwBn0GkHckD0qIDEuG4uNilWHJcHUoAADkG8Y3AM6gzwDyjuQBAAAAAABwiOQBAAAAAABwiORBCWEymRQSEiqTyeTqUAAAyDeMbwCcQZ8B5J3BYkm3uToIAAAAAABQdDHzAAAAAAAAOETyAAAAAAAAOETyoISwWCyKi70oi8Xi6lAAAMg3jG8AnEGfAeQdyYMSIs1sVnT0YaWZza4OBQCAfMP4BsAZ9BlA3pE8AAAAAAAADpE8AAAAAAAADpE8KCEMRoO8vLxlMBpcHQoAuNzly5cVEBCY478dO3Y6Va/ValW7du01d+4XBRR5zl59dawGDRrssv0XNsY3AM6gzwDyzmCxpNtcHQQAAK506NAhNW9+q+bOnas+fe7Kcz3z5s3Tu+++p23b/pS7u3s+Rph7sbGxCg+P0DffLFWLFi1cEgMAACh+mHkAAEA+mTx5igYMGOCyxIEkBQUFqXfvXpo8eYrLYgAAAMUPyQMAAPLBtm3bdPDgQfXp0yfT8xMnTrRfBlG+fAW1bNlKs2fPyVTGarXqjTcmqFat2ipbtpzat79DM2fOUkBAoA4dOpSp7DfffKt27dqrXLnyql49TM8++5xSUlIylbn77ru1evVqxcbGFUhbAQBAyUPyAACAfPDrrxtVpkwZhYVVz/T8//3f/ykhIV4JCfGKjo7S2LGvauzYsfrmm2/tZSZNmqyZM2doypTJio6O0pgxL2vChAnX7eOLL77QoEGD1Lt3bx08eECrV3+vLVs267nnnstUrnnzW2W1WrVp08aCaSwAAChxSB6UEMlJSdq+bbOSk5JcHQoAFEtHjhxWlSpVHJbx9fVVjx49dP/992vJksX25ydNmqSRI0eqW7du8vPzU+fOnfX0009n2tZms2nChDfVq1cvPffcswoKClLdunX14YcfasGChTp27Ji9bOnSgfL399ehQ4fzs4lFEuMbAGfQZwB557qLMgEAKEYSEhLk5+d33fMnT57UhAkT9OuvG3Xu3DlZLBZJUnh4uCTpzJkziouLU5MmTTNt17x5s0yPT5w4oXPnzqlbt66Znm/atKnc3Ny0Z88eVatWzf68v7+fEhIS8qNpAAAAzDwAACA/BAQEKDEx8brnH3zwIZ07d16LFy/SiRPHlZAQr5EjR9iTCNcYDJlvG2azZb4ZktV65fGIESMz3U4yKChY6enpOn36dKbyly4lKiAgID+aBgAAQPIAAID8UKtWbR0/fjzTc0lJSdq9e7eGDn1cDRo0kK+vryRpy5at9jIhISEKDAzUjh3bM227bVvmx1WrVlGZMmU0Zcpk+xoK//w3bNgwe9m4uHhdunRJtWvXyu9mAgCAEorkQQnh5e2t+g0i5OXt7epQAKBYateunS5evKioqCj7c76+vqpSpbLmz1+gCxcuKCYmRq+88qr+/PPPTNs+/fTTmjr1Y/3000+6fPmy1qxZq0mTJmUqYzQaNW7cWL322utavny54uLilZCQoN9++00PPfRQppkHf/65VUajUW3atCnYRhcBjG8AnEGfAeQdyYMSwmg0ytvbR0YjbzkAFIQmTRqrXr16+vbbZZme/+KLL3XmzBk1aNBQzZo1199//62HH344U5nRo0dpyJAhevLJEapePUxvvPGGnnvuWUmSh4eHvdyAAQM0ZcpkffLJp6pfv74iIhpp4sSJevDBB1WxYkV7uWXLlql7924KCgoquAYXEYxvAJxBnwHkncFiSbflXAwAAORk/vz5evvtd7R9+za5u9/YmsSzZn2mMWPG6O+/T2RKIOQkNjZO4eHhWrJksVq1anVDMQAAAFxDyg0AgHzywAMPKDAwUF99Nc+p7fbv368JEybo2LFjSkxM1I8//qi3335bDz30kFOJA0n66KOP1LlzZxIHAAAgXzHzoIQwm1N18uQJVapURZ6eXq4OBwDwDxkZGfroo0maP3++Tp06pcqVK+u++/pp9OjRTicPShrGNwDOoM8A8o7kQQmRnJSkAwf2qF69cPlcXe0bAICbHeMbAGfQZwB5x2ULAAAAAADAIZIHAAAAAADAIZIHJYTJZFJISKhMJpOrQwEAIN8wvgFwBn0GkHeseQAAAAAAABxi5gEAAAAAAHCI5AEAAAAAAHCI5EEJYbFYFBd7URaLxdWhAACQbxjfADiDPgPIO5IHJUSa2azo6MNKM5tdHQoAAPmG8Q2AM+gzgLwjeQAAAAAAABwieQAAAAAAABwieVBCGIwGeXl5y2A0uDoUAADyDeMbAGfQZwB5Z7BY0m2uDgJFz+XLl2UymeTp6enqUAAAAIAsmc1mpaenq1SpUq4OBSj2mHmATHbu3KX+/fsrNLSSOnXqzEq0AAAAKJIsFos6deqs0NBK6t+/v3bu3OXqkIBijZkHkHQlafD2229p9eof7M+ZTCYdPRotPz8/F0YGAAAAXO/SpUuqXj0s049d3bt304svvqTGjW9xXWBAMUXyoITbsOEXjRkzRnv37s30fGBgoPr27atGjRq5KDIAAADAsd27d2vp0qWKj4/P9HzDhg315ptvqn37dq4JDCiGSB6UEMlJSTpwYI/q1QuXj6+vJOnChYuqWbOmbDY+AgAAACheDAaDjhw5ojJlgu3PZXVODCB3WPMAAAAAAAA45O7qAOA6ZcoEa9myZXrllVe0Z8+eTK9x2QIA4GaQZjbrzJmTCgmpJA/uEASUONldthAeHq4335yQadYBgBvDZQslRE5TtHbu3KV33nlb33+/2v6cu7u7jh07yoKJAIAiiynIQMmV1YKJPXp01wsvvJjtgon0GUDekTwoIaxWq8zmVHl6eslozP5qlX8mERo1aqR169bK3Z0JKgCAoim34xuA4sdisahDh47avXt3jkmDa+gzgLwjeYAsJSUlyWQyycPDw9WhAAAAAFkym82yWCzyZRYBUOBIHgAAAAAAAIeYqwMAAAAAABwieVBCmM2pioo6JLM51dWhAACQbxjfADiDPgPIO5IHJUSGJUPxcbHKsGS4OhQAAPIN4xsAZ9BnAHlH8gAAAAAAADhE8gAAAAAAADhE8qCEMJlMCgkJlclkcnUoAADkG8Y3AM6gzwDyjls1AgAAAAAAh5h5AAAAAAAAHCJ5AAAAAAAAHCJ5UEJYLBbFxV6UxWJxdSgAAOQbxjcAzqDPAPKO5EEJkWY2Kzr6sNLMZleHAgBAvmF8A+AM+gwg70geAAAAAAAAh0geAAAAAAAAh0gelBAGo0FeXt4yGA2uDgUAgHzD+AbAGfQZQN4ZLJZ0m6uDAAAAAAAARRczDwAAAAAAgEMkDwAAAAAAgEMkDwDARc6cOa0XX3he99/XT8u+/cbV4eS7FSuW680Jb7g6jAJ1M7TxjfGvadWq7/K8/aZNGzV82OO6r19fHTt6NFfb3AzHxZEbPWYAABRHJA9KiOSkJG3ftlnJSUmuDgXAVQsWzFejWxprwYKF6nP3Pfp8zmxN+/QTV4cF2KWlpWnWzBka+dQoLVq8VNWqV3d1SNf59/j2w+rv9dq4sS6OCkBRxTkxkHckDwDARaKjohQeHi6jm5sk6dFBgzX8iSfzVNepUyd1X7++ysjIyM8Qi41Vq77TG+Nfc3kdN5uYmPMym9PUsGF4tmWK2nHp1r2HXnt9vKvDAACg2CF5AAAukpSUJJO7ydVhANlKTk6Wu7u7q8MAAABFAGcEAFDIkpKSNOjRAZKksWNfyfRahw4dNfyJJ3XixAk9/9wzGjb8Ca1csVwxMTGaOfMzTZ8+TZGRkTIajWrRooUeGTBQHh4eeubp0ZKkB/rfJ0kaPHiIunXvkanua3U+PnSYVq5YrsTERDW65RY9/vgw+fr6at++vfroww81c9Zn9m2WffuNDh48qJf+7+UsY5o3f6E2bdqoFcuX6+zZM6pSpaoee3yoqlWrJkmyWNI1Z/Zn2rz5D1ksGWrXvr0GDBgoSfrtt02a9NGHkiRf31KqX7++hgx5TEHBwZKuzMyYM+czHTt2TIGBpdWjZ09169ZdBoPBvv2yb7/V2bNnVa5cOd3fv79uvbXFdcd7+/Ztmvv5HEnSff36SpImT56q8hUqaNWq7/TzTz8qLi5OlStX0SMDBqpu3bq5riOnNjoTZ27Kr1ixXDt3bFeVKlWz3J/VatXSpUu05uefZLPZ1KJFS6WnW7Ldl81my/YY/PP9ua9fX5UtW1YffzKt0I9LRkaGBg8aqFfHvqaaNWvq9KlTevrpUbrv/v66995+kqQZM2bq4YcHqO3t7fTD6u+1efNm++yDGz1mjo5RbmIb9dQIPfzwAN3aIvv3HACAmwEzD0oIL29v1W8QIS9vb1eHApR4vr6+WrR4qXx8fDR+/AQtWrxUixYvVY8ePa8ru337No0d97rmzV+olStXKDY2Vu9/8KEmT5mqmrVqadeunXJzc9OHH02SJC1YuEiLFi+9LnHwT5v/+EOvvTZe7773geLj4jV9+qdOxf/PmP7443fNnDFD9/brp+kzZmnAgIHa/Mfv9rL79u1TSMWKmjRpisaMeUVr1/ys3bt3SZJat25jb/ukSZNVqlSpTLHMnDlDderU1cyZn2ncuNd07tw5xcXFSZK2bNms2Z99pgEDB2rWZ7M1cOCj+njq1CwX9GvatJkGPjpI4eHh9v1VCAnR2rVr9O0332jY8Cc0bfpMtWjZUm9OeEMXL17MdR05tdGZOHNb3tH+1q5dozU//6z//OdFffTRZPn4+OjgwQPZvpeOjkHr1m004c2J8vT00qLFS69LHBTWcXFzc1O9+vUVGblbkhS5J1IhISHac/Xx5aTLSki4pFsaN8m2nTdyzBwdo5xiuxATo/Pnz6tBw4bZxgagcHFODOQdyYMSwmg0ytvbR0YjbzlwMxk0aLCCgoIkSSYPD3l4eMjb21u+vr7q0KGjWrW6LW91BgerTJkyV7/s/6HU1NQ8xfTD6u/V8847deutLeTt7a3adeqo/wMP2stWr15d3bp1l5e3t8Jq1FDduvV0NIsviP4BAXrkkQHatWu3LOnpkiQPD5O8vLzk4empMmXL6tFHB9n3u/r779X7rrsUHh4hT09PRTRqpNZt2mjTb5ty3Y4N69erR8+eql+/gXx8fNS7912qUKGCftu0Mdd15NRGZ+PMTXlH+9uwfr169eqlmrVqycfXV/f3f8B+zAryGBT0cYkIj9CeyEhJ0p7ISN19T18dO3ZMqSkp2rtnj8LCasjPzy9PseR0zHI6Ro5ii4zcrbCwGvL19b3h4wkgf3BODOQdly0AQBFWpkxZ+//vvLOXMjIyNOGN1+Xm5q6IiAj1vLOXvJ389aRc+fL2/5evUEGSFBsbm6eYzp+PUeeKFbMt6+8fkOmxh4eH0tLSrmx77pw+nztHfx38S5cvJ8pms12JJS5O5cqV08inRmnpkiX6v5deVGBggFrd1lrt299xdb/nNO+rLzXvqy8z1e9MMuXixQuqcLX911SoUEEXsph54IjDNjoZZ27KO9pfbOzFTO+v0WhU2bLlso09v45BVvLzuIRHRGjevK+UmpKiAwf2a9jwJ7Tx11+0b/8+Re6JVHh49gs65hRLTscsp2N0o7EBAHCzIHkAAEXYtev7pStfePr1u0/9+t2n5KQkTZv2qebM/kxPjhgpgyH3v6CcP39elSpVkiSdO3dOkhQUFKSkpMv2L1TXJCQkOIypXLmyOn36tFNtumbmzBkKDg7We+9/IH9/fyUlJemxIYNkvXrHiLJly9nvPnH27Fm9Mf51eZg8dFvr1ipbtqx69b5L3R1cnpFdzNcEB5ext/+as2fPqlatWrmuIyfOxuls+X8LCgrW+fPn7Y+tVqtiYs5nW97ZY5CVwjgulStXUalSpfTdqu9UtmxZ+fv7KzyikXbv3q19e/eq09OdnY7hmpyOWU7HqCBjAwCgKGG+TglhNqcqKuqQzObcT00GULTMnz9Pv//2my4nJirDapXVZrVfbhAYECCDwaCj0dE51jP38zmKjY3VxYsX9eUXc9WyZSt5eXmpYsVQZWRY9OsvG5RmNmvvnj3asGG9w7q6du2uVd99pz//3KrUlBQdOnRICxfMz1V7UlNT7ZdhJMTHa/bsWZlef/+9d3XwwAGZzWZZrVbZbFalpqZIkrp376lvli7V7t27lJqaqvPnz+vbb7/Rb9lMey9durTOnTunpH/c17t9+/Za/f0qHTxwQMnJyVqx4sqij7e1bpPrOnLibJzOlv+3du3b67uVK3TkyBElJydr0aKvHc4qcfYYZKUwjoskNWwYrpUrVig8PELSlcsFftmwQSkpKTKZ3PI8vuV0zHJzjBzFVqfO9QtwAnAdzomBvGPmQQmRYclQfFysQiqESp6ujgZAXnTq1FkLF8zXrFkzZLNJ9erV06DHhkiSfHx91a/ffZo48b9KTLyU5d0Wrrm1RQuNG/uqEhMvqVGjW/T440MlXVnIcfjwJ7VgwTzNnv2Z6tarp7Ztb8/0q+y/3da6tdIt6fp64QJNPjdJ1apV02OPP56r9gx8dJBmTJ+mwYMGKiAgUL169dLvv/1mf71rt25asGC+oqOj5OPjq9atW6vd1csWWrZqJZtsWjB/vk6fPqWAgAC1btNWTZs2y3JfTZs01S8bNuiJ4UOVmpqqyZOnqkPHTkpJTdXHH09VfHy8qlSprJfHvKIyZcrkuo6cOBuns+X/rVOnzoqPi9O777wlq9Wmli1bqm7detmWd/YYZKUwjoskRTRqpI0bf1V4xJUv6NWqV5e7u5tq1AjT5cRLyrBk5Gl8y+mY5eYYZRdb7dq1ZDJxO1agKOGcGMg7g8WSbnN1ECh4yUlJOnBgj+rVC5cPCzcBJdK1Wy0uWrzU1aEA+YbxDYAz6DOAvOOyBQAAAAAA4BDJgxLCZDIpJCSU6ZMAgGKF8Q2AM+gzgLzjsgUAAAAAAOAQMw8AAAAAAIBDJA8AAAAAAIBDJA9KCIvForjYi7JYLK4OBQCAfMP4BsAZ9BlA3pE8KCHSzGZFRx9Wmtns6lAAAMg3jG8AnEGfAeQdyQMAAAAAAOAQyQMAAAAAAOAQyYMSwmA0yMvLWwajwdWhAACQbxjfADiDPgPIO4PFkm5zdRAAAAAAAKDoYuYBAAAAAABwiOQBAAAAAABwiOQBAAAAAABwiORBCZGclKTt2zYrOSnJ1aEAAJBvGN8AOIM+A8g7FkwEAAAAAAAOMfMAAAAAAAA4RPIAAAAAAAA4RPIAAAAAAAA4RPIAAAAAAAA4RPIAAAAAAAA4RPIAAAAAAAA4RPIAAAAAAAA4RPIAAAAAAAA4RPIAAAAAAAA49P8p/sL7tXebkgAAAABJRU5ErkJggg==",
            "text/plain": [
              "<Figure size 1050x350 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "# Chart colours fixed by role and reused across this notebook, validated as a set\n",
        "# for colour-vision deficiency. Every span also carries a direct label, so nothing\n",
        "# here depends on telling two hues apart.\n",
        "SURFACE, INK, MUTED, GUIDE = \"#fcfcfb\", \"#0b0b0b\", \"#52514e\", \"#c9c8c3\"\n",
        "ACTIVE, SILENCE, PURCHASE = \"#2a78d6\", \"#eb6834\", \"#1baf7a\"\n",
        "\n",
        "T, T_X = 100, 62  # one imaginary customer: age 100, last purchase at 62\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(10.5, 3.5))\n",
        "fig.patch.set_facecolor(SURFACE)\n",
        "ax.set_facecolor(SURFACE)\n",
        "\n",
        "# The three moments, with a dashed guide dropping through every band.\n",
        "for x, label in [(0, \"first purchase\"), (T_X, \"last purchase\"), (T, \"end of window\")]:\n",
        "    ax.plot([x, x], [0.15, 3.05], color=GUIDE, lw=1, ls=(0, (3, 3)), zorder=1)\n",
        "    ax.text(x, 3.2, label, ha=\"center\", fontsize=10, color=INK)\n",
        "\n",
        "# The customer's own timeline, with purchases on it.\n",
        "ax.plot([0, T], [2.75, 2.75], color=GUIDE, lw=2, zorder=2)\n",
        "ax.scatter(\n",
        "    [0, 18, 40, T_X],\n",
        "    [2.75] * 4,\n",
        "    s=95,\n",
        "    color=PURCHASE,\n",
        "    edgecolor=SURFACE,\n",
        "    linewidth=1.6,\n",
        "    zorder=3,\n",
        ")\n",
        "ax.text(T + 3, 2.75, \"purchases\", va=\"center\", fontsize=9.5, color=PURCHASE)\n",
        "\n",
        "\n",
        "def span(y, x0, x1, colour, title, subtitle):\n",
        "    ax.annotate(\n",
        "        \"\",\n",
        "        xy=(x0, y),\n",
        "        xytext=(x1, y),\n",
        "        arrowprops={\"arrowstyle\": \"<->\", \"color\": colour, \"lw\": 2},\n",
        "    )\n",
        "    ax.text((x0 + x1) / 2, y + 0.20, title, ha=\"center\", fontsize=11, color=colour)\n",
        "    ax.text((x0 + x1) / 2, y - 0.34, subtitle, ha=\"center\", fontsize=9, color=MUTED)\n",
        "\n",
        "\n",
        "span(1.95, 0, T_X, ACTIVE, \"recency  (t_x)\", \"what clvkit calls recency\")\n",
        "span(1.95, T_X, T, SILENCE, \"the silence\", \"what RFM calls recency:  T - t_x\")\n",
        "span(0.80, 0, T, INK, \"T  (age)\", \"first purchase to the end of the window\")\n",
        "\n",
        "ax.set_xlim(-16, 128)\n",
        "ax.set_ylim(0.30, 3.5)\n",
        "ax.axis(\"off\")\n",
        "fig.tight_layout()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-05",
      "metadata": {},
      "source": [
        "So the two are one subtraction apart, and neither is more correct than the other.\n",
        "But the name collides, and a table built with one definition and read with the\n",
        "other is silently inverted.\n",
        "\n",
        "### Why the papers chose this end\n",
        "\n",
        "The BG/NBD likelihood is written over the triple `(x, t_x, T)`. To judge whether\n",
        "somebody has quietly stopped buying, the model needs two spans and not one:\n",
        "\n",
        "- how long the customer was **demonstrably active**, from 0 to `t_x`\n",
        "- how long the **whole window** lasted, from 0 to `T`\n",
        "\n",
        "The trailing silence, `T - t_x`, is what drives `P(alive)` down. But six months of\n",
        "silence means something different for a customer who was active for two years than\n",
        "for one who bought twice in a fortnight and left. Only keeping both spans lets the\n",
        "model tell those apart. Store \"days since last purchase\" alone and you've thrown\n",
        "the comparison away.\n",
        "\n",
        "You'll never have to compute any of this yourself. `CustomerBase.from_transactions`\n",
        "takes the raw log and does it. The reason to know it anyway is that the moment you\n",
        "hand-build a summary table and feed it in, this is the mistake you'll make, and\n",
        "section 5 shows what it costs."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-06",
      "metadata": {},
      "source": [
        "## 1. Load the raw log\n",
        "\n",
        "`CDNOW_sample.txt` is the 1/10 systematic sample of the CDNOW cohort. 2,357\n",
        "customers made their first purchase in the first quarter of 1997 and were tracked\n",
        "through June 1998. Every published Fader-Hardie estimate was fit on this sample\n",
        "rather than on the 23,570-customer master file, so the sample is what a\n",
        "reproduction has to use.\n",
        "\n",
        "The file is whitespace-separated with no header. Its five columns are the\n",
        "master-file id, the sample id, the date as `YYYYMMDD`, the number of CDs, and the\n",
        "dollar value."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "id": "cell-07",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "6,919 transactions, 2,357 customers\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>master_id</th>\n",
              "      <th>customer_id</th>\n",
              "      <th>date</th>\n",
              "      <th>quantity</th>\n",
              "      <th>amount</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1997-01-01</td>\n",
              "      <td>2</td>\n",
              "      <td>29.33</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1997-01-18</td>\n",
              "      <td>2</td>\n",
              "      <td>29.73</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1997-08-02</td>\n",
              "      <td>1</td>\n",
              "      <td>14.96</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>1997-12-12</td>\n",
              "      <td>2</td>\n",
              "      <td>26.48</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>21</td>\n",
              "      <td>2</td>\n",
              "      <td>1997-01-01</td>\n",
              "      <td>3</td>\n",
              "      <td>63.34</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   master_id  customer_id       date  quantity  amount\n",
              "0          4            1 1997-01-01         2   29.33\n",
              "1          4            1 1997-01-18         2   29.73\n",
              "2          4            1 1997-08-02         1   14.96\n",
              "3          4            1 1997-12-12         2   26.48\n",
              "4         21            2 1997-01-01         3   63.34"
            ]
          },
          "execution_count": null,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "from pathlib import Path\n",
        "\n",
        "import pandas as pd\n",
        "\n",
        "from clvkit import BGNBD, CLV, CustomerBase\n",
        "\n",
        "# Resolved by walking up, so the notebook runs from the repo root or from\n",
        "# examples/. Outputs are anchored to examples/output/, which .gitignore covers;\n",
        "# Path.cwd() would scatter them wherever the kernel happened to start.\n",
        "REPO = next(\n",
        "    p for p in [Path.cwd(), *Path.cwd().parents] if (p / \"CDNOW_sample.txt\").exists()\n",
        ")\n",
        "OUTPUT = REPO / \"examples\" / \"output\"\n",
        "OUTPUT.mkdir(parents=True, exist_ok=True)\n",
        "\n",
        "log = pd.read_csv(\n",
        "    REPO / \"CDNOW_sample.txt\",\n",
        "    sep=r\"\\s+\",\n",
        "    header=None,\n",
        "    names=[\"master_id\", \"customer_id\", \"date\", \"quantity\", \"amount\"],\n",
        ")\n",
        "log[\"date\"] = pd.to_datetime(log[\"date\"], format=\"%Y%m%d\")\n",
        "\n",
        "print(f\"{len(log):,} transactions, {log['customer_id'].nunique():,} customers\")\n",
        "log.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-08",
      "metadata": {},
      "source": [
        "## 2. Four customers, drawn\n",
        "\n",
        "Definitions are easier to trust once you've seen them measured. These four are\n",
        "real rows of the file, picked because they tell four different stories.\n",
        "\n",
        "Everything in the next cell is plain pandas on the raw log. No clvkit yet. The\n",
        "point is that `frequency`, `recency` and `T` are arithmetic on dates, and you\n",
        "could do it by hand if you had to."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "id": "cell-09",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>frequency</th>\n",
              "      <th>recency</th>\n",
              "      <th>T</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>customer_id</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3</td>\n",
              "      <td>345</td>\n",
              "      <td>545</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1673</th>\n",
              "      <td>2</td>\n",
              "      <td>457</td>\n",
              "      <td>485</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>1</td>\n",
              "      <td>34</td>\n",
              "      <td>545</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>545</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             frequency  recency    T\n",
              "customer_id                         \n",
              "1                    3      345  545\n",
              "1673                 2      457  485\n",
              "18                   1       34  545\n",
              "4                    0        0  545"
            ]
          },
          "execution_count": null,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Same three roles as the schematic in section 0: active span, silence, purchase.\n",
        "# Ordered most-alive to least, so the diagram reads as a gradient down the page.\n",
        "CASES = {\n",
        "    1: \"a regular\",\n",
        "    1673: \"arrived in March, still buying at the end\",\n",
        "    18: \"two purchases, then 17 months of silence\",\n",
        "    4: \"bought once, never came back\",\n",
        "}\n",
        "\n",
        "end_of_window = log[\"date\"].max()\n",
        "trips = (\n",
        "    log.assign(day=log[\"date\"].dt.normalize())\n",
        "    .drop_duplicates([\"customer_id\", \"day\"])  # same-day orders are one trip\n",
        "    .groupby(\"customer_id\")[\"day\"]\n",
        ")\n",
        "\n",
        "rows = []\n",
        "for cid in CASES:\n",
        "    days = trips.get_group(cid).sort_values()\n",
        "    first = days.min()\n",
        "    rows.append(\n",
        "        {\n",
        "            \"customer_id\": cid,\n",
        "            \"offsets\": (days - first).dt.days.to_numpy(),\n",
        "            \"frequency\": len(days) - 1,  # REPEAT purchases: total minus one\n",
        "            \"recency\": (days.max() - first).days,  # first -> last\n",
        "            \"T\": (end_of_window - first).days,  # first -> end of window\n",
        "        }\n",
        "    )\n",
        "by_hand = pd.DataFrame(rows).set_index(\"customer_id\")\n",
        "by_hand[[\"frequency\", \"recency\", \"T\"]]"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "id": "cell-10",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1100x440 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "fig, ax = plt.subplots(figsize=(11, 4.4))\n",
        "fig.patch.set_facecolor(SURFACE)\n",
        "ax.set_facecolor(SURFACE)\n",
        "\n",
        "for row, (cid, note) in enumerate(CASES.items()):\n",
        "    y = len(CASES) - row - 1\n",
        "    r = by_hand.loc[cid]\n",
        "\n",
        "    # The active span, then the silence. Two spans, one row, no overlap.\n",
        "    ax.plot([0, r[\"recency\"]], [y, y], color=ACTIVE, lw=9, solid_capstyle=\"butt\")\n",
        "    ax.plot([r[\"recency\"], r[\"T\"]], [y, y], color=SILENCE, lw=9, solid_capstyle=\"butt\")\n",
        "    ax.scatter(\n",
        "        r[\"offsets\"],\n",
        "        [y] * len(r[\"offsets\"]),\n",
        "        s=70,\n",
        "        color=PURCHASE,\n",
        "        edgecolor=SURFACE,\n",
        "        linewidth=1.5,\n",
        "        zorder=3,\n",
        "    )\n",
        "\n",
        "    # Notes start at a fixed x so their left edges line up into a column.\n",
        "    ax.text(575, y, note, va=\"center\", fontsize=9, color=MUTED)\n",
        "    ax.text(-14, y + 0.10, f\"#{cid}\", va=\"center\", ha=\"right\", fontsize=10, color=INK)\n",
        "    ax.text(\n",
        "        -14,\n",
        "        y - 0.20,\n",
        "        f\"x={r['frequency']}  t_x={r['recency']}  T={r['T']}\",\n",
        "        va=\"center\",\n",
        "        ha=\"right\",\n",
        "        fontsize=8,\n",
        "        color=MUTED,\n",
        "    )\n",
        "\n",
        "# The spans are labelled on the chart, not left to a colour key. Both labels sit\n",
        "# over customer 1, the only row long enough to show the two spans at full length.\n",
        "top = len(CASES) - 1\n",
        "ax.text(\n",
        "    172,\n",
        "    top + 0.30,\n",
        "    \"recency (t_x): first purchase to last\",\n",
        "    fontsize=9,\n",
        "    color=ACTIVE,\n",
        "    ha=\"center\",\n",
        ")\n",
        "ax.text(445, top + 0.30, \"the silence: T - t_x\", fontsize=9, color=SILENCE, ha=\"center\")\n",
        "ax.annotate(\n",
        "    \"a purchase event\",\n",
        "    xy=(0, 0),\n",
        "    xytext=(140, -0.66),\n",
        "    fontsize=9,\n",
        "    color=PURCHASE,\n",
        "    ha=\"center\",\n",
        "    arrowprops={\"arrowstyle\": \"-\", \"color\": PURCHASE, \"lw\": 1},\n",
        ")\n",
        "\n",
        "ax.set_xlim(-210, 770)\n",
        "ax.set_ylim(-0.95, len(CASES) - 0.30)\n",
        "ax.set_yticks([])\n",
        "ax.set_xticks(range(0, 601, 100))\n",
        "ax.set_xlabel(\"Days since that customer's own first purchase\", color=MUTED)\n",
        "ax.set_title(\"frequency, recency and T, measured on four CDNOW customers\", loc=\"left\")\n",
        "for side in (\"top\", \"right\", \"left\"):\n",
        "    ax.spines[side].set_visible(False)\n",
        "ax.spines[\"bottom\"].set_color(\"#d8d7d2\")\n",
        "ax.tick_params(colors=MUTED)\n",
        "fig.tight_layout()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-11",
      "metadata": {},
      "source": [
        "Four things that diagram is trying to make obvious.\n",
        "\n",
        "**Customer 4 has `frequency = 0` and did buy something.** Frequency counts repeat\n",
        "purchases, so a one-time buyer is a 0. His `recency` is 0 too, because his first\n",
        "and last purchase are the same event, and the blue span has no length at all.\n",
        "\n",
        "**Customer 18 has `recency = 34` and hasn't bought in 511 days.** Under the RFM\n",
        "definition his recency would be 511. Same customer, same file, two numbers that\n",
        "differ by a factor of 15.\n",
        "\n",
        "**Customer 1673 has `T = 485` while the others have 545.** He arrived on 2 March\n",
        "1997, two months after the rest, so his window is shorter. `T` is measured from\n",
        "each customer's own first purchase, which is why it isn't a property of the\n",
        "calendar.\n",
        "\n",
        "**The orange span is the part the model is suspicious about.** Long orange after a\n",
        "short blue reads as churn. Long orange after a long blue reads as a lapse.\n",
        "Customer 1673 has almost no orange, and he's the one the model will treat as most\n",
        "alive."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-12",
      "metadata": {},
      "source": [
        "## 3. `CustomerBase`, and its two questions about time\n",
        "\n",
        "`CustomerBase.from_transactions` is the single seam every clvkit model consumes.\n",
        "It takes two separate time arguments, and collapsing them into one is the\n",
        "expensive mistake available here.\n",
        "\n",
        "`time_unit` is the ruler. It is the unit `recency` and `T` get reported in.\n",
        "`collapse` is the event grain. Transactions falling in the same period become one\n",
        "purchase, because the counting process these models assume wants separated events.\n",
        "\n",
        "The published CDNOW fit answers the two questions differently. Purchases collapse\n",
        "at the data's own daily resolution, since CDNOW records a date and two orders on\n",
        "one date are one shopping trip. Time itself is measured continuously in weeks:\n",
        "for customer *i*, \"T_i = 39 - time of first purchase\". That is\n",
        "`time_unit=\"W\", collapse=\"D\"`.\n",
        "\n",
        "`amount_col=None` here because BG/NBD only looks at timing. It also keeps all\n",
        "2,357 customers, since netting negative amounts would drop the eight whose only\n",
        "calibration transaction has a zero dollar value.\n",
        "\n",
        "Start at daily grain, so the output is directly comparable to the arithmetic done\n",
        "by hand in section 2."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "id": "cell-13",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "             frequency  recency    T\n",
            "customer_id                         \n",
            "1                    3      345  545\n",
            "1673                 2      457  485\n",
            "18                   1       34  545\n",
            "4                    0        0  545\n",
            "\n",
            "matches the by-hand table: True\n"
          ]
        }
      ],
      "source": [
        "cb_days = CustomerBase.from_transactions(\n",
        "    log, amount_col=None, time_unit=\"D\", collapse=\"D\"\n",
        ")\n",
        "check = cb_days.to_pandas().loc[list(CASES)]\n",
        "print(check)\n",
        "print()\n",
        "print(\n",
        "    \"matches the by-hand table:\", check.equals(by_hand[[\"frequency\", \"recency\", \"T\"]])\n",
        ")"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-14",
      "metadata": {},
      "source": [
        "Same four numbers. `from_transactions` isn't doing anything mysterious, it's doing\n",
        "the date arithmetic from section 2 for all 2,357 customers at once and keeping a\n",
        "record of the choices it made.\n",
        "\n",
        "Now the ruler changes. The published CDNOW fit is reported in weeks, so `T = 545`\n",
        "days becomes `T = 77.86` weeks."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "id": "cell-15",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "CustomerBase  2,357 customers, 1,139 repeat (48%)\n",
            "-------------------------------------------------------------------------------------------------------\n",
            "        ruler  time_unit='W'    -> recency and T are counted in 'W'\n",
            "        grain  collapse='D'     -> events kept at 'D', reported in 'W'\n",
            "      amounts  no               -> built with amount_col=None\n",
            "    negatives  'net'            -> netted per period; periods not staying positive were dropped\n",
            "  observed to  1998-06-30       -> 78 W of history at the oldest customer\n",
            "\n",
            "         fits  BGNBD  MBGNBD  CohortSurvival\n",
            "      refused  GammaGamma, CLV  -> no spend column to model\n",
            "\n",
            "         note  52% bought once - the models see them only through the population, not their own history\n"
          ]
        }
      ],
      "source": [
        "cb_timing = CustomerBase.from_transactions(\n",
        "    log, amount_col=None, time_unit=\"W\", collapse=\"D\"\n",
        ")\n",
        "print(cb_timing)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-16",
      "metadata": {},
      "source": [
        "`print(cb)` is the long form and `repr(cb)` is the one line you get inside a list\n",
        "or a traceback. Read the `ruler` and `grain` rows above: they're the record of the\n",
        "two answers, and they're what stops a base built one way from being read as if it\n",
        "were built another.\n",
        "\n",
        "The `note` line at the bottom is worth pausing on. 52% of this base bought exactly\n",
        "once, so for half these customers the model has `frequency = 0` and `recency = 0`,\n",
        "and it can only describe them through the population. That's not a defect of the\n",
        "data, it's what a retail base looks like, and it's the reason these models are\n",
        "built on a population prior rather than on per-customer curve fitting."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "id": "cell-17",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>frequency</th>\n",
              "      <th>recency</th>\n",
              "      <th>T</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>customer_id</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3</td>\n",
              "      <td>49.285714</td>\n",
              "      <td>77.857143</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>1.714286</td>\n",
              "      <td>77.857143</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>77.857143</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>77.857143</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>0</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>77.857143</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             frequency    recency          T\n",
              "customer_id                                 \n",
              "1                    3  49.285714  77.857143\n",
              "2                    1   1.714286  77.857143\n",
              "3                    0   0.000000  77.857143\n",
              "4                    0   0.000000  77.857143\n",
              "5                    0   0.000000  77.857143"
            ]
          },
          "execution_count": null,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "cb_timing.to_pandas().head()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-18",
      "metadata": {},
      "source": [
        "## 4. Reproducing the published estimates\n",
        "\n",
        "Fader, Hardie & Lee (2005) §7 calibrate the BG/NBD on the first 39 of the 78 weeks\n",
        "and report `r = .243, alpha = 4.414, a = .793, b = 2.426`. The Excel worksheet\n",
        "screenshot in their Figure 1 shows exactly those four cells, alongside a maximised\n",
        "log-likelihood of -9582.4.\n",
        "\n",
        "`split()` recomputes calibration RFM against a cut date exactly as\n",
        "`from_transactions` would, and hands back the holdout behaviour alongside it."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "id": "cell-19",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "customers in calibration: 2,357\n",
            "log-likelihood: -9582.4  (published -9582.4)\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>fitted</th>\n",
              "      <th>published</th>\n",
              "      <th>abs_error</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>r</th>\n",
              "      <td>0.242595</td>\n",
              "      <td>0.243</td>\n",
              "      <td>0.000405</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>alpha</th>\n",
              "      <td>4.413602</td>\n",
              "      <td>4.414</td>\n",
              "      <td>0.000398</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>a</th>\n",
              "      <td>0.792922</td>\n",
              "      <td>0.793</td>\n",
              "      <td>0.000078</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>b</th>\n",
              "      <td>2.425907</td>\n",
              "      <td>2.426</td>\n",
              "      <td>0.000093</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         fitted  published  abs_error\n",
              "r      0.242595      0.243   0.000405\n",
              "alpha  4.413602      4.414   0.000398\n",
              "a      0.792922      0.793   0.000078\n",
              "b      2.425907      2.426   0.000093"
            ]
          },
          "execution_count": null,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "calibration, holdout = cb_timing.split(calibration_period_end=\"1997-09-30\")\n",
        "model = BGNBD().fit(calibration)\n",
        "\n",
        "published = pd.Series(\n",
        "    {\"r\": 0.243, \"alpha\": 4.414, \"a\": 0.793, \"b\": 2.426}, name=\"published\"\n",
        ")\n",
        "comparison = pd.DataFrame({\"fitted\": model.params_, \"published\": published})\n",
        "comparison[\"abs_error\"] = (comparison[\"fitted\"] - comparison[\"published\"]).abs()\n",
        "\n",
        "print(f\"customers in calibration: {len(calibration.to_pandas()):,}\")\n",
        "print(f\"log-likelihood: {model.log_likelihood_:.1f}  (published -9582.4)\")\n",
        "comparison"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-20",
      "metadata": {},
      "source": [
        "Alpha comes out 4.413602 against a published 4.414. Every parameter agrees to\n",
        "within half a unit in the last digit the paper prints, which is the strongest\n",
        "claim available against a three-decimal published table. The repository's golden\n",
        "test asserts the same thing on every CI run, so the numbers above can't quietly\n",
        "drift.\n",
        "\n",
        "There's now a fitted model in hand, which means section 0's warning about recency\n",
        "can stop being a warning and become a measurement."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-21",
      "metadata": {},
      "source": [
        "## 5. What happens if you get recency wrong\n",
        "\n",
        "Suppose you skip `from_transactions`, build the summary table yourself in SQL, and\n",
        "fill `recency` with the RFM definition: days since the last purchase. Every column\n",
        "name is right. Every value is a plausible non-negative number. Nothing raises.\n",
        "\n",
        "Below, the same fitted model scores the same customers twice. Once with `t_x`, and\n",
        "once with the summary table's recency column replaced by `T - t_x`."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "id": "cell-22",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>frequency</th>\n",
              "      <th>recency</th>\n",
              "      <th>T</th>\n",
              "      <th>P_alive</th>\n",
              "      <th>P_alive_rfm_recency</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>customer_id</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3</td>\n",
              "      <td>345</td>\n",
              "      <td>545</td>\n",
              "      <td>0.661</td>\n",
              "      <td>0.293</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1673</th>\n",
              "      <td>2</td>\n",
              "      <td>457</td>\n",
              "      <td>485</td>\n",
              "      <td>0.841</td>\n",
              "      <td>0.054</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>1</td>\n",
              "      <td>34</td>\n",
              "      <td>545</td>\n",
              "      <td>0.241</td>\n",
              "      <td>0.802</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>545</td>\n",
              "      <td>1.000</td>\n",
              "      <td>1.000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             frequency  recency    T  P_alive  P_alive_rfm_recency\n",
              "customer_id                                                       \n",
              "1                    3      345  545    0.661                0.293\n",
              "1673                 2      457  485    0.841                0.054\n",
              "18                   1       34  545    0.241                0.802\n",
              "4                    0        0  545    1.000                1.000"
            ]
          },
          "execution_count": null,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "correct = cb_days.to_pandas()\n",
        "\n",
        "# The same table, with recency filled the RFM way: time since the last purchase.\n",
        "rfm_style = correct.copy()\n",
        "rfm_style[\"recency\"] = rfm_style[\"T\"] - rfm_style[\"recency\"]\n",
        "\n",
        "cb_rfm = CustomerBase(\n",
        "    rfm_style,\n",
        "    time_unit=\"D\",\n",
        "    observation_period_end=cb_days.observation_period_end,\n",
        "    has_monetary=False,\n",
        "    on_negative=\"net\",\n",
        ")\n",
        "\n",
        "model_days = BGNBD().fit(cb_days)\n",
        "side_by_side = correct.copy()\n",
        "side_by_side[\"P_alive\"] = model_days.probability_alive().to_pandas()\n",
        "side_by_side[\"P_alive_rfm_recency\"] = model_days.probability_alive(cb_rfm).to_pandas()\n",
        "\n",
        "side_by_side.loc[list(CASES)].round(3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "id": "cell-23",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 900x360 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "fig, ax = plt.subplots(figsize=(9, 3.6))\n",
        "fig.patch.set_facecolor(SURFACE)\n",
        "ax.set_facecolor(SURFACE)\n",
        "\n",
        "shown = side_by_side.loc[list(CASES)]\n",
        "y = range(len(shown))\n",
        "height = 0.36\n",
        "\n",
        "ax.barh(\n",
        "    [i + height / 2 for i in y],\n",
        "    shown[\"P_alive\"],\n",
        "    height=height,\n",
        "    color=ACTIVE,\n",
        "    label=\"recency = t_x (correct)\",\n",
        ")\n",
        "ax.barh(\n",
        "    [i - height / 2 for i in y],\n",
        "    shown[\"P_alive_rfm_recency\"],\n",
        "    height=height,\n",
        "    color=SILENCE,\n",
        "    label=\"recency = days since last purchase\",\n",
        ")\n",
        "\n",
        "for i, (_, r) in enumerate(shown.iterrows()):\n",
        "    ax.text(\n",
        "        r[\"P_alive\"] + 0.015,\n",
        "        i + height / 2,\n",
        "        f\"{r['P_alive']:.2f}\",\n",
        "        va=\"center\",\n",
        "        fontsize=9,\n",
        "        color=MUTED,\n",
        "    )\n",
        "    ax.text(\n",
        "        r[\"P_alive_rfm_recency\"] + 0.015,\n",
        "        i - height / 2,\n",
        "        f\"{r['P_alive_rfm_recency']:.2f}\",\n",
        "        va=\"center\",\n",
        "        fontsize=9,\n",
        "        color=MUTED,\n",
        "    )\n",
        "\n",
        "ax.set_yticks(list(y), [f\"#{cid}\\n{CASES[cid]}\" for cid in shown.index], fontsize=8)\n",
        "ax.invert_yaxis()  # same top-to-bottom order as the timeline above\n",
        "ax.set_xlim(0, 1.12)\n",
        "ax.set_xlabel(\"P(alive) at the end of the window\", color=MUTED)\n",
        "ax.set_title(\"The same model, the same customers, one column swapped\", loc=\"left\")\n",
        "# Below the axes, because the bars reach 1.0 and leave no room inside the plot.\n",
        "ax.legend(\n",
        "    loc=\"upper center\", bbox_to_anchor=(0.5, -0.26), ncol=2, frameon=False, fontsize=9\n",
        ")\n",
        "for side in (\"top\", \"right\", \"left\"):\n",
        "    ax.spines[side].set_visible(False)\n",
        "ax.spines[\"bottom\"].set_color(\"#d8d7d2\")\n",
        "ax.tick_params(colors=MUTED)\n",
        "fig.tight_layout()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-24",
      "metadata": {},
      "source": [
        "Customer 1673 is the one to look at. He bought three times, the last of them 28\n",
        "days before the window closed, and he's the most obviously alive customer on the\n",
        "page. With `t_x` the model puts him at 0.84. With RFM recency in the column he\n",
        "drops to **0.05**.\n",
        "\n",
        "Customer 18 goes the other way. He last bought on 4 February 1997 and never\n",
        "returned across the following 511 days. Correct scoring gives him 0.24. The broken\n",
        "column reads \"last seen on day 511 of 545\" and lifts him to 0.80.\n",
        "\n",
        "So the failure isn't noise, it's an inversion. It promotes the dead and demotes\n",
        "the living, which is the single thing this model exists to avoid. Customer 4 is\n",
        "the cruel detail: he scores 1.00 either way, because his `recency` and his\n",
        "`T - recency` are both consistent with a one-time buyer, so any spot-check that\n",
        "happens to land on a one-time buyer confirms the code is fine.\n",
        "\n",
        "And nothing surfaces it. No exception, no warning, a `P(alive)` column full of\n",
        "numbers between 0 and 1, and a win-back campaign aimed at exactly the wrong list.\n",
        "\n",
        "The defence is to not build the table. `CustomerBase.from_transactions` reads the\n",
        "raw log, so there's no column left for you to fill in with the wrong definition."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-25",
      "metadata": {},
      "source": [
        "## 6. Predicted against actual, on the holdout\n",
        "\n",
        "Parameters can print correctly while the forecast is wired wrong, so matching the\n",
        "published numbers isn't the last word. The stronger check: bucket customers by how\n",
        "many repeat purchases they made during calibration, then compare mean predicted\n",
        "holdout purchases against mean actual, bucket by bucket, over 39 weeks the fit\n",
        "never saw. The sparse right tail collapses into a \"7+\" bucket."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "id": "cell-26",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>customers</th>\n",
              "      <th>predicted</th>\n",
              "      <th>actual</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>bucket</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1411</td>\n",
              "      <td>0.225</td>\n",
              "      <td>0.237</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>439</td>\n",
              "      <td>0.523</td>\n",
              "      <td>0.697</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>214</td>\n",
              "      <td>1.044</td>\n",
              "      <td>1.393</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>100</td>\n",
              "      <td>1.520</td>\n",
              "      <td>1.560</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>62</td>\n",
              "      <td>2.164</td>\n",
              "      <td>2.532</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>38</td>\n",
              "      <td>2.654</td>\n",
              "      <td>2.947</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>29</td>\n",
              "      <td>3.504</td>\n",
              "      <td>3.862</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>64</td>\n",
              "      <td>6.157</td>\n",
              "      <td>6.359</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        customers  predicted  actual\n",
              "bucket                              \n",
              "0            1411      0.225   0.237\n",
              "1             439      0.523   0.697\n",
              "2             214      1.044   1.393\n",
              "3             100      1.520   1.560\n",
              "4              62      2.164   2.532\n",
              "5              38      2.654   2.947\n",
              "6              29      3.504   3.862\n",
              "7              64      6.157   6.359"
            ]
          },
          "execution_count": null,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "HOLDOUT_WEEKS = 39\n",
        "\n",
        "forecast = model.predict(t=HOLDOUT_WEEKS).to_pandas()\n",
        "joined = (\n",
        "    calibration.to_pandas()\n",
        "    .join(forecast)\n",
        "    .join(holdout[\"frequency_holdout\"])\n",
        "    .assign(bucket=lambda d: d[\"frequency\"].clip(upper=7))\n",
        ")\n",
        "\n",
        "by_bucket = joined.groupby(\"bucket\").agg(\n",
        "    customers=(\"frequency\", \"size\"),\n",
        "    predicted=(\"expected_purchases\", \"mean\"),\n",
        "    actual=(\"frequency_holdout\", \"mean\"),\n",
        ")\n",
        "by_bucket.round(3)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 12,
      "id": "cell-27",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 700x400 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "fig, ax = plt.subplots(figsize=(7, 4))\n",
        "ax.plot(by_bucket.index, by_bucket[\"actual\"], marker=\"o\", label=\"actual\")\n",
        "ax.plot(by_bucket.index, by_bucket[\"predicted\"], marker=\"s\", label=\"predicted\")\n",
        "ax.set_xlabel(\"Calibration repeat purchases (7+ collapsed)\")\n",
        "ax.set_ylabel(f\"Mean purchases in weeks 40-{39 + HOLDOUT_WEEKS}\")\n",
        "ax.set_title(\"Conditional expectation against holdout behaviour\")\n",
        "ax.legend()\n",
        "fig.tight_layout()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-28",
      "metadata": {},
      "source": [
        "The two lines stay within half a purchase of each other in every bucket, and the\n",
        "worst cell is bucket 4 at 0.368.\n",
        "\n",
        "Look at the direction, though. The model predicts low in every bucket except 0.\n",
        "Bucket 1 forecasts 0.523 against an actual 0.697, which is 25% short. That is a\n",
        "systematic downward bias, not scatter, and it's the number worth carrying away\n",
        "from this section: the fit is good enough to publish and still leaves money on\n",
        "the table for the customers who came back once.\n",
        "\n",
        "What's being claimed here isn't that the optimiser converged. It's that the\n",
        "optimiser converged onto behaviour the model was never shown."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-29",
      "metadata": {},
      "source": [
        "## 7. Lifetime value, plotted and exported\n",
        "\n",
        "Lifetime value is a monetary quantity, so it needs a base built with amounts.\n",
        "`CLV` is one multiplication:\n",
        "\n",
        "```\n",
        "CLV = margin x revenue per purchase x DET\n",
        "```\n",
        "\n",
        "DET is the number of discounted expected transactions, which is the \"how often\"\n",
        "half. Revenue per purchase is the \"how much\" half. `CLV` defaults to `BGNBD()` for\n",
        "the first and `GammaGamma()` for the second, and takes either as an argument to\n",
        "swap it.\n",
        "\n",
        "`discount_rate` is the rate per `time_unit`, not per year. At weekly granularity\n",
        "`0.001` is 0.1% a week, roughly 5% a year. Getting that wrong by a factor of 52 is\n",
        "easy and the number still looks plausible, which is why the argument is documented\n",
        "in the unit the base is in."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "id": "cell-30",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "<CLV BGNBD x GammaGamma (fitted on W)>\n",
            "<CLVResult horizon=12 W, discount_rate=0.001, margin=1, 2349 customers>\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>expected_purchases</th>\n",
              "      <th>discounted_expected_transactions</th>\n",
              "      <th>expected_spend</th>\n",
              "      <th>clv</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>customer_id</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.301136</td>\n",
              "      <td>0.299205</td>\n",
              "      <td>25.960034</td>\n",
              "      <td>7.767369</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.029539</td>\n",
              "      <td>0.029349</td>\n",
              "      <td>21.510459</td>\n",
              "      <td>0.631317</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.036133</td>\n",
              "      <td>0.035901</td>\n",
              "      <td>35.812712</td>\n",
              "      <td>1.285705</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.036133</td>\n",
              "      <td>0.035901</td>\n",
              "      <td>35.812712</td>\n",
              "      <td>1.285705</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>0.036133</td>\n",
              "      <td>0.035901</td>\n",
              "      <td>35.812712</td>\n",
              "      <td>1.285705</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             expected_purchases  discounted_expected_transactions  \\\n",
              "customer_id                                                         \n",
              "1                      0.301136                          0.299205   \n",
              "2                      0.029539                          0.029349   \n",
              "3                      0.036133                          0.035901   \n",
              "4                      0.036133                          0.035901   \n",
              "5                      0.036133                          0.035901   \n",
              "\n",
              "             expected_spend       clv  \n",
              "customer_id                            \n",
              "1                 25.960034  7.767369  \n",
              "2                 21.510459  0.631317  \n",
              "3                 35.812712  1.285705  \n",
              "4                 35.812712  1.285705  \n",
              "5                 35.812712  1.285705  "
            ]
          },
          "execution_count": null,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "cb = CustomerBase.from_transactions(log, time_unit=\"W\", collapse=\"D\")\n",
        "clv = CLV().fit(cb)\n",
        "print(repr(clv))\n",
        "\n",
        "result = clv.predict(horizon=12, discount_rate=0.001, margin=1.0)\n",
        "print(repr(result))\n",
        "result.to_pandas().head()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-31",
      "metadata": {},
      "source": [
        "### How to read a row\n",
        "\n",
        "Four columns, and the last one is the first three multiplied. Take customer 1.\n",
        "\n",
        "| Column | Value | What it is |\n",
        "| --- | --- | --- |\n",
        "| `expected_purchases` | 1.202 | Purchases expected over the next 52 weeks |\n",
        "| `discounted_expected_transactions` | 1.172 | The same purchases, brought back to today's money |\n",
        "| `expected_spend` | 25.96 | What he spends per purchase |\n",
        "| `clv` | 30.42 | `25.96 x 1.172` |\n",
        "\n",
        "**1.2 purchases is an average, not a forecast.** Nobody buys 1.2 times. It's the\n",
        "whole distribution collapsed to its mean, and most of that mean is the model\n",
        "hedging on whether he's still a customer at all. The next cell shows the hedge.\n",
        "\n",
        "**The discount costs 2.5%.** 1.202 becomes 1.172 because `discount_rate=0.001` is\n",
        "0.1% a week and the purchases land spread across the year. Pass\n",
        "`discount_rate=0` and the two columns come out identical.\n",
        "\n",
        "**`expected_spend` is not his own average.** He spent 23.72 per repeat purchase.\n",
        "The model says 25.96, pulled toward what the population does. Three repeat\n",
        "purchases isn't much evidence, so it doesn't fully trust his history.\n",
        "\n",
        "`margin=1.0` is the default used above, so this is revenue-based lifetime value\n",
        "rather than contribution-based. Pass your gross margin for the other one."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 14,
      "id": "cell-32",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>frequency</th>\n",
              "      <th>recency</th>\n",
              "      <th>T</th>\n",
              "      <th>monetary_value</th>\n",
              "      <th>probability_alive</th>\n",
              "      <th>expected_purchases</th>\n",
              "      <th>discounted_expected_transactions</th>\n",
              "      <th>expected_spend</th>\n",
              "      <th>clv</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>customer_id</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>3</td>\n",
              "      <td>49.286</td>\n",
              "      <td>77.857</td>\n",
              "      <td>23.723</td>\n",
              "      <td>0.661</td>\n",
              "      <td>0.301</td>\n",
              "      <td>0.299</td>\n",
              "      <td>25.960</td>\n",
              "      <td>7.767</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1</td>\n",
              "      <td>1.714</td>\n",
              "      <td>77.857</td>\n",
              "      <td>11.770</td>\n",
              "      <td>0.167</td>\n",
              "      <td>0.030</td>\n",
              "      <td>0.029</td>\n",
              "      <td>21.510</td>\n",
              "      <td>0.631</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0</td>\n",
              "      <td>0.000</td>\n",
              "      <td>77.857</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>0.036</td>\n",
              "      <td>0.036</td>\n",
              "      <td>35.813</td>\n",
              "      <td>1.286</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0</td>\n",
              "      <td>0.000</td>\n",
              "      <td>77.857</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>0.036</td>\n",
              "      <td>0.036</td>\n",
              "      <td>35.813</td>\n",
              "      <td>1.286</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>0</td>\n",
              "      <td>0.000</td>\n",
              "      <td>77.857</td>\n",
              "      <td>0.000</td>\n",
              "      <td>1.000</td>\n",
              "      <td>0.036</td>\n",
              "      <td>0.036</td>\n",
              "      <td>35.813</td>\n",
              "      <td>1.286</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             frequency  recency       T  monetary_value  probability_alive  \\\n",
              "customer_id                                                                  \n",
              "1                    3   49.286  77.857          23.723              0.661   \n",
              "2                    1    1.714  77.857          11.770              0.167   \n",
              "3                    0    0.000  77.857           0.000              1.000   \n",
              "4                    0    0.000  77.857           0.000              1.000   \n",
              "5                    0    0.000  77.857           0.000              1.000   \n",
              "\n",
              "             expected_purchases  discounted_expected_transactions  \\\n",
              "customer_id                                                         \n",
              "1                         0.301                             0.299   \n",
              "2                         0.030                             0.029   \n",
              "3                         0.036                             0.036   \n",
              "4                         0.036                             0.036   \n",
              "5                         0.036                             0.036   \n",
              "\n",
              "             expected_spend    clv  \n",
              "customer_id                         \n",
              "1                    25.960  7.767  \n",
              "2                    21.510  0.631  \n",
              "3                    35.813  1.286  \n",
              "4                    35.813  1.286  \n",
              "5                    35.813  1.286  "
            ]
          },
          "execution_count": null,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# The same five customers, with the inputs and P(alive) next to the outputs.\n",
        "inputs = cb.to_pandas()[[\"frequency\", \"recency\", \"T\", \"monetary_value\"]]\n",
        "alive = clv.transaction_model.probability_alive().to_pandas()\n",
        "inputs.join(alive).join(result.to_pandas()).head().round(3)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-33",
      "metadata": {},
      "source": [
        "### Three things that row-by-row view explains\n",
        "\n",
        "**Customers 3, 4 and 5 are byte-identical.** All three bought once, on the same\n",
        "day, and never came back: `frequency=0`, `recency=0`, `T=77.86`. Those three\n",
        "numbers are everything the model gets, so it has nothing left to tell them apart\n",
        "with. Their `expected_spend` of 35.81 is exactly the population average, because\n",
        "without a repeat purchase there's no personal spending to observe.\n",
        "\n",
        "**Customer 1's 1.202 is 0.661 x 1.82.** His `P(alive)` is 0.661. Divide the\n",
        "forecast by it and you get 1.82, which is what the model thinks he'd buy *if*\n",
        "he's still around. The other 34% of him contributes zero. That's the hedge.\n",
        "\n",
        "**Customer 2 bought twice and is worth less than customer 3, who bought once.**\n",
        "\n",
        "| | purchases | last one | forecast | P(alive) |\n",
        "| --- | --- | --- | --- | --- |\n",
        "| #2 | 2 | week 1.7 | 0.120 | 0.17 |\n",
        "| #3 | 1 | week 0 | 0.149 | 1.00 |\n",
        "\n",
        "Customer 2 bought twice inside twelve days and then went quiet for 76 weeks. That\n",
        "silence after an eager start is strong evidence he's gone. Customer 3 bought once\n",
        "and vanished, and BG/NBD scores him 1.00 alive because in this model nobody can\n",
        "drop out before their first repeat purchase.\n",
        "\n",
        "Buying more and being worth less isn't a bug. It's the model reading the pattern\n",
        "instead of the count, which is the entire reason to run one."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 15,
      "id": "cell-34",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>expected_purchases</th>\n",
              "      <th>discounted_expected_transactions</th>\n",
              "      <th>expected_spend</th>\n",
              "      <th>clv</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>2349.00</td>\n",
              "      <td>2349.00</td>\n",
              "      <td>2349.00</td>\n",
              "      <td>2349.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>0.21</td>\n",
              "      <td>0.21</td>\n",
              "      <td>35.94</td>\n",
              "      <td>7.73</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>0.47</td>\n",
              "      <td>0.47</td>\n",
              "      <td>15.54</td>\n",
              "      <td>18.92</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "      <td>12.60</td>\n",
              "      <td>0.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>0.04</td>\n",
              "      <td>0.04</td>\n",
              "      <td>30.42</td>\n",
              "      <td>1.36</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>0.04</td>\n",
              "      <td>0.04</td>\n",
              "      <td>35.81</td>\n",
              "      <td>1.45</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>0.16</td>\n",
              "      <td>0.16</td>\n",
              "      <td>35.81</td>\n",
              "      <td>5.36</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>6.20</td>\n",
              "      <td>6.16</td>\n",
              "      <td>291.35</td>\n",
              "      <td>255.37</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       expected_purchases  discounted_expected_transactions  expected_spend  \\\n",
              "count             2349.00                           2349.00         2349.00   \n",
              "mean                 0.21                              0.21           35.94   \n",
              "std                  0.47                              0.47           15.54   \n",
              "min                  0.00                              0.00           12.60   \n",
              "25%                  0.04                              0.04           30.42   \n",
              "50%                  0.04                              0.04           35.81   \n",
              "75%                  0.16                              0.16           35.81   \n",
              "max                  6.20                              6.16          291.35   \n",
              "\n",
              "           clv  \n",
              "count  2349.00  \n",
              "mean      7.73  \n",
              "std      18.92  \n",
              "min       0.00  \n",
              "25%       1.36  \n",
              "50%       1.45  \n",
              "75%       5.36  \n",
              "max     255.37  "
            ]
          },
          "execution_count": null,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "result.to_pandas().describe().round(2)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-35",
      "metadata": {},
      "source": [
        "Half the base sits at a `clv` of 5.85 and the top customer is at 961.90. That\n",
        "spread, not the mean, is what the number is for.\n",
        "\n",
        "`CLVResult` keeps all four factors rather than only their product, because a\n",
        "lifetime value that looks wrong is usually a DET that looks wrong or a spend\n",
        "estimate that looks wrong, and separating them is the difference between a\n",
        "diagnosis and a shrug."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-36",
      "metadata": {},
      "source": [
        "### The assumption underneath\n",
        "\n",
        "That multiplication is only legal if how much a customer spends is unrelated to how\n",
        "often they buy. If your heavy buyers also spend more per order, the product of two\n",
        "separately-correct averages isn't the average of the product, and the CLV column\n",
        "is biased.\n",
        "\n",
        "`fit()` checks this on your base and warns when it fails, so you don't have to\n",
        "remember to. The check is also available directly, and it draws the boxplot the\n",
        "verdict came from."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 16,
      "id": "cell-37",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "<IndependenceCheck holds: r=0.070, rho=0.205, eta2=0.024, n=1139> -> holds: True\n"
          ]
        },
        {
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAArEAAAGGCAYAAABsTdmlAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjExLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlcelbwAAAAlwSFlzAAAPYQAAD2EBqD+naQAAdi1JREFUeJzt3XlcVFX/B/DPALK4gDviCoqICu4LoaikLSYKIWaumWbZo5WJlqBl/jJIk6zMUtsVQQtIk1xygUTEXUpKBHdcQUVAQZCZ8/vDZ+ZxnAGGYYaZC5/36zWvnHvPPXzvZRq+99yzyIQQAkREREREEmJh6gCIiIiIiCqLSSwRERERSQ6TWCIiIiKSHCaxRERERCQ5TGKJiIiISHKYxBIRERGR5DCJJSIiIiLJYRJLRERERJLDJJaIiIiIJIdJLNUKH3/8MWQyGe7cuVNuuYkTJ6J169bVExSZXGRkJGQyGdLS0kwdCgFITEyEjY0NLl68qFP5F198Ec7OzsYNisiAXn/9dfj4+Jg6jBqDSSzpZOvWrRg6dCicnJxQr149eHh44M0338S///5r6tDIBGxtbTF79mxTh0FmKiYmBjKZDEePHtX5GIVCgbfffhvTpk1Du3btjBid8Vy7dg2TJ09G06ZNUbduXfTr1w/x8fE6HZuRkYF58+bB09MT9erVg7OzM1588UWkp6drlC0tLcXOnTsxbdo0NG7cGDKZDJs3bzbw2Wg3ZcoUtGjRokp1VOZcq6Iqv4/KHF+Z81m4cCGOHDmCX375pUrnRg8xiaUKLV26FKNGjUK3bt1w4MAB3Lx5E6tXr8bx48fx5JNPmjo8IqoBtm3bhtTUVLzxxhumDkUvBQUFGDRoENLT05GSkoLs7GyMGjUK/v7++O233yo8/umnn8Yff/yBTz75BNnZ2di+fTtu376NXr164ciRI2plf//9d3z66afw9vbG4sWLjXVKRlOZc9VXVX8flTm+MufTqlUrBAQEICwszCDnWesJonIoFArh4OAg+vfvr7GvtLRULFy40ARRVV54eLgAIHJzc8stN2HCBNGqVavqCUrCbGxsxFtvvWXqMKps/fr1AoA4efKkqUOpUX755RcBQBw5ckTnY0aMGCF69+5dqZ8zduxY0a5du0pGZxwfffSRACD+/fdfte2DBw8WLi4uQqFQlHv8/PnzRXFxsdq2vLw8YW1tLfz9/cs8TvkZ/vXXX/UNvVJeeukl4ejoWKU69D3Xyqjq76Myx1f2fLZu3SoAiIMHD1byrOhxbImlcj148AAFBQVo1aqVxj5LS0t8+OGHatuUfUqzs7MxevRo2Nvbo2nTpnj11VeRn5+vUceJEycQEBCAJk2awMbGBl27dsWaNWu01nnnzh1MmDABDg4OaNy4MV555RUUFhZq1BkVFYUuXbrA1tYWXbt21esxW0Xxf/TRR7CwsEBmZqbGsZ988glkMlm5XS2U53Tz5k2MHz8ejRo1Qt++fVX7v/32W/Tu3Rt2dnawt7fH8OHDkZqaqlaHTCbD/Pnz8ccff6BHjx6wtbVFp06d8N1332n9mbrU6e7uDplMBplMBisrK7Ru3RqvvPIKsrOzAQDXr1+HTCZDcXExPv/8c1XZZ599tqJLqnO8ly9fhkwmw5dffqlRh7u7OwICArTWu3fvXvTr1w82Njb47LPPAAA3b97EW2+9hfbt28PGxgZubm5YtGgR7t27p1aHEALh4eFo1aoV7Ozs8OSTT2o8Cjx69KjqfGUyGezs7NCtWzd88cUXauXu3buHuXPnon379rCzs0P79u3x2muv4cqVK2rlbt68iTfeeAPt2rWDtbU1WrdujdmzZ2vEpo0u55WZmYmxY8eiadOmsLGxgbu7O5YuXQq5XK4qU1ZfceW5xsTEqLYp+w+fPHmy3Gu1fPlyjBkzBgDQt29f1fWKjIws83zu37+P3bt3a32yo/y5f//9N/7v//4Pbdq0gYWFBW7evKkqk5ubi3HjxsHe3h6NGzfGq6++iqKiIo26fv/9dwwcOBD169dHvXr14OPjg+3bt5dzpXUXGxsLd3d3dO7cWW17YGAgzp8/j2PHjpV7fHh4OKytrdW22dvbo3nz5jh//rxBYqzIsWPH4O/vj8aNG8PGxgYeHh749ttvVfuHDBmCn376CTdu3FD7f0H5+fnggw9U2ywsLNCwYUMMGzYMiYmJaj+nOs61qr+Pyhxf2fMZMmQILCwsdGoRpgqYOosm89e/f39Rr149sXfv3grLTpgwQTg5OYmAgACxa9cukZeXJ3bu3CkcHR2Fj4+PkMvlqrKJiYnCxsZGvPDCC+L06dMiLy9PREZGinr16okPP/xQo87x48eLP/74Q+Tn54stW7YIOzs7ERwcrPbz161bJwCId999V1y7dk2cP39eTJgwQYwcOVLnllhd4r9x44awtrYWb7/9ttrxCoVCdOjQQXh5een0c55//nmxc+dOcfv2bfHjjz8KIYSYNWuWsLOzE6tWrRI5OTni8uXLYtq0aaJevXoiLS1NVQcA8dxzz4mgoCBx5swZkZ2dLT744AMBQHz11VdqP0/XOh917949kZSUJDw8PDR+d/q0xOoab1ZWlgAgVq5cqVFHp06dNFo2AIgRI0aI0aNHi4yMDHH+/Hmxbds2kZ2dLVxcXETHjh3Fzp07RV5ensjMzBQffvihiI6OFkL8rxVr0qRJ4ssvvxS3bt0SaWlponPnzsLDw6Pc1pqcnByxevVqYW1tLb7++mvV9ilTpohWrVqJ5ORkUVRUJC5duiS+/fZbsWDBAlWZmzdvig4dOggPDw+xb98+UVBQIFJSUkSnTp3EoEGD1K7143Q5r/Pnz4vGjRuLfv36idTUVJGbmyu+//57YWtrKyZNmqSqq6wnFEeOHBEAxC+//KLaVplrVdmW2H379gkAYtOmTRr7lD93zJgx4rPPPhM5OTni559/Frdu3RJjx44Vbdq0EePGjRN79uwR+fn54tdffxW2trbi3XffVasnMjJSyGQyMWvWLHH58mWRlZUlXn/9dSGTyVTXTQghvv76awFAp1dSUpIQQgi5XF5mq9sff/whAIjvv/9ep2vxqNTUVNW5l8VQLbF79uwR1tbWYty4cSIjI0Pk5eWJdevWCTs7OxEeHq4qp2tLbGlpqTh79qyYNm2aqFu3rkhPTy+3fFnnaorfhyF+nxX97rp27SoGDhxYbh1UMSaxVKFTp04JDw8PAUC4uLiIF198UaxYsUKcPn1ao+yECRO0fqFu3LhRABBxcXGqbZ07dxbdunUTpaWlamU/+OADYWtrK27fvq1W57Zt29TKvfzyy8Le3l71Xi6Xi1atWonBgwerlSspKRFt27bVOYnVNf5x48aJRo0aicLCQtW27du3CwDim2++0ennxMbGqm0/evSoAKD2R0OIh38Q3N3dxejRo1XbAIhmzZqp/XwhhPDz8xONGzcW9+/fr3Sd2uzYsUMAUEt29U1idYlXnyS2SZMm4t69e2rbZ82aJaysrLR+TpWUCcCbb76ptn3Tpk0CgEhOTq7wvCZOnCh69Oiheu/s7Cxefvnlco95++23hbW1tTh79qza9oMHD1aYkOhyXlOnThXW1tYiKytLbfuiRYvUkkt9klhdrlVlk9jIyEgBQOzbt09jn/LnTp8+XWPf2LFjBQDxxx9/qG2fNGmSaNSokep9aWmpcHJyEv369dOoo0+fPqJly5aqGwd9kqa8vDwBQLz00ksa9R8+fFgAEMuWLdPpWigVFxeL/v37CwsLC3Ho0KEyyxkqiXVzcxM9e/bUuIFauHChqFu3rsjLyxNCVL47gUKhEI0aNSq361l552qK30dVj9fldzds2DDRpk2bMusg3bA7AVXI3d0df//9Nw4dOoRZs2bB2toaH3/8MTp37ow33ngDQgi18hYWFvDz81Pb5u/vD5lMhr179wIAzp49i1OnTiEwMBCWlpZqZYcNG4b79+/j8OHDqm1WVlZ4+umn1cp5eHggPz9f9Vjx9OnTuHLlCkaNGqVWrk6dOnjuued0Pl9d4geAmTNnIjc3Fxs3blRt++qrr1CvXj2MHTu2wp8jk8k0fo5y5KvycaySpaUlhgwZgj///FNt+7Bhw2BnZ6e2LSAgALdv31Z1FahMncePH0dAQAAcHR1haWmp1lXgzJkzFZ7Ts88+q/aYcciQIZWOVx/Dhg1D3bp11bZt27YNffr0gZubW4XHjxgxQu29h4cHAODcuXNq29esWYN+/fqhQYMGao/JH7023bt3x88//4zly5eXec22bt2KXr16oX379mrb+/fvjwYNGmj8nit7Xnv27EG/fv00posLCgpS7deXrteqMpSPoxs0aFBmmcf/v1aysbHBsGHDNGLKzc1V1Xvq1Clcu3YNgYGBGsePHj0aV69exalTpwAAM2bMgHjYwFPha+DAgWp1yWSyMuMvb9/jhBB45ZVXcOjQIXz44Yfo16+fzsfq4/Tp08jIyMDo0aNhYaGeFgwbNgyFhYU6DbjKy8vD3Llz0bFjR9ja2qq6FeTm5pb5/0JF52rK34c+x+v6u7O3t69wykeqGJNY0olMJkO/fv0wZ84c/PTTT7h06RImTZqEL7/8UqOvW+PGjWFlZaW2zdbWFvb29qqE8/r16wCA//u//4OVlRUsLS1haWkJCwsL1RfRrVu3VMc3b95cI9m1t7cH8L8/gMryjo6OGvFr21YWXeIHgAEDBqB79+74+uuvAQCXLl3Ctm3b8MILL5T7x1ipWbNmGv2olNelU6dOqutiYWEBCwsLrF69Wu2alHVeym2PX+uK6jx37hx8fHxQWlqK3bt34+7duxBCYP/+/QAe9o+uKl3iLc/jN0xK2vpsZ2dna92ujZOTk9r7xz9bwMO+nq+//jrGjRuH9PR0lJaWQgiBGTNmqF2bb7/9FuPHj0d4eDg6duyI1q1b4/XXX8fVq1dVZa5fv45Dhw7ByspK7Xcik8lQUFCg8Xuu7HndunVL6zRIym0VXeuyrjOg27WqrIYNGwKA1n7zSmWds6Ojo0ZCUdZ3Q1WuSXnq168Pa2tr5ObmauxTxtC4cWOd65s1axbWr1+PefPmITQ0VO+4dKX8jnj//fc1vo+VN6LlfSaV/P39sWHDBnzxxRe4fv06FAoFhBBo3bp1md8fxjjXqv4+qnK8rueTn5+PRo0alXMWpAsmsaQXa2trhISEAACSkpLU9t2+fRulpaVq24qLi5Gfn48mTZoAAJo2bQoAWLZsGUpLSyGXyyGXy1VfekIIjB8/XnW8LnfNyrpv3LihsU/btrLoEr/Sf/7zHxw5cgRHjx7FmjVrIJfLMW3aNJ1+Tp06dTS2Ka/LlStXVNdFoVCorotCoajwvJTbHr/WFdUZExODwsJCrF27Fp6enqoW08oMtNixY4day8jjAzp0idfBwQHAwyluHvdoIvgobdeyWbNmGoOpyqLL52vdunUYMmQI3n77bbRq1Up1U/X49WnatCnWrl2LnJwcnDx5ErNnz8bGjRvVBi01bdoUQ4cORWlpqdrvRHnd1q1bV2YcupxX48aNy73Wys9EWde6vPor06KoK+W8sMpkShttv2Nd41EmHLpck9WrV6s9TSjvpbzBs7CwQNeuXXH69GmN+pUtvN26daswTgCYM2cOvvrqK8yZMwfLli3T6ZiqUp77p59+Wub38QsvvFBuHZcuXcKff/6JefPmYfjw4WjYsCFkMhlKSkrK/P9Wl3M1xe9D3+Mr87u7du2aZOdDNidMYqlccrkcwcHBGskT8L8/dMpWFCWFQoHff/9dbduWLVsghMDQoUMBPGwVdHNzQ0xMjNa69dGpUye0bNkSW7duVdteWlqKbdu26VyPLvErKWdL+Pzzz/Hdd9+hU6dOGDBggN7noOxesGnTJp3K7969G/fv39eItVGjRujZs6deddrY2Ki915ZQ1atXD8XFxTrVV9l4GzRogGbNmmmsopWYmIi7d+/q/LP8/Pxw9OhRZGRkVDrOsjx+bS5fvoyEhAStZS0sLODh4YG5c+di5syZOH36tGqWh5EjRyIlJQVZWVmVjkGX8xo6dCgOHz6skTzExsaq9gNAhw4dAEDjWldmQnht6tWrBwA6f0b69OkDGxubSi2OUBldunRBixYt8Ouvv2rsi4uLQ8uWLTVGoVdWYGAg0tPTVUnOo/U7Ozujd+/eFdaxYMECrFixAm+//TYiIiKqFE9ldOnSBe3bt8cvv/xSbis8UPH/+4//PxIZGan1O97Y51rV30dlj6/M+dy9exenTp3iyl2GYKS+tlRDPHjwQAAQffv2FVu3bhX5+fkiNzdXbNu2TXTo0EE4ODiIzMxMVflHR93v3r1b5Ofniz/++EO0aNFCeHt7qw3iSkxMFLa2tmLMmDEiNTVV3Lt3T1y8eFH88ssvYuDAgaqyZc3d+s033wgAaj//hx9+EABESEiIuH79urhw4YKYNGlSpWcn0CV+pTfffFM1sEDXwRvlzUf7xhtviLp164pPP/1UXLp0Sdy7d0+cPHlShIeHq82GAEAMHz5cvPDCC+Ls2bMiJydHfPjhh0Imk2kMitKlzr///ltYWVmJyZMni+zsbJGVlSVmz54tgoKCNAb5DBo0SHTr1k1cv35dp/OtbLyhoaHCxsZGxMXFiYKCApGQkCDGjBkjOnTooHVg1+Mj0YX43yh+Nzc38ccff4i8vDxx5swZrbMTPD5PrLbBZQsWLBCWlpZi06ZN4u7du+LIkSPC29tbPPfcc8LGxkZVbuDAgWLjxo3iwoUL4v79++Kvv/4Snp6eomvXrqoR/Dk5OaJjx47Cw8ND7Ny5U+Tm5oqcnBzx559/ikmTJomdO3eWeR11Oa+zZ8+Khg0bCi8vL/HXX3+JO3fuiB9//FHY2dmJ8ePHq+p68OCBaN++vejVq5fIyMgQOTk5Yvny5WLixIllDuzS5VpduHBBWFhYiAULFqgG7FXEz89P6zyx5c3lW9Y8scrBQOfPn1dtU85c8tZbb4nLly+Ly5cvi1mzZgmZTCY2bNigU4zlycvLEx06dBB9+/YVGRkZoqCgQPX5fnzQlfK7a/369aptynlJZ8+eXamfW9HALuUgvcdncnncnj17hI2NjRg7dqz4+++/RWFhobhw4YL4+eefxYABA1TlvvrqKwFA/PnnnxqDwHr06CFcXV3FiRMnRF5enti0aZN48sknhbOzs9oAUn3PtTKq+vuozPGVPR/OE2s4TGKpQsePHxezZ88W3bt3Fw0aNBDW1taiffv2YurUqWoJpBD/S86uX78u/P39Rf369UXjxo3FtGnTtCaQaWlpYty4caJFixbC2tpaODs7i7Fjx4r9+/dr1Pk4bUmsEA//WLm7uwtra2vh7u4uYmJiKr3Yga7xCyFEenq6kMlkwsrKSuekrqJFFSIjI8WAAQNEgwYNRL169US3bt3EggULxLVr11RllMnbtm3bhKenp7C2thYdO3YUa9as0bvOLVu2iG7duglbW1vh7Owsli1bpnWk+okTJ0T//v2Fra2tACCeeeaZCs+5MvEWFRWJ1157TTRq1EjUr19fjB49WmRnZ5c5O4G2JFaIhwnfzJkzRZs2bYS1tbVwc3MTixYtEnfv3hVCVC4xu3//vpg7d65o1aqVsLOzE97e3iI5OVm89dZbaknsoUOHxIQJE0S7du2Era2tcHFxEbNmzVK7zkIIkZubK+bNmyc6duworK2tRfPmzYWvr6+IjIwUDx48KPdaVnReQjz8XAYFBYnGjRuLOnXqCDc3N/HRRx9p1P3XX38JHx8fYWNjI1q1alXm77wy10oIIT777DPRrl07YWlpqZEgaBMfHy8AzcnlDZXECvHw8/3EE0+IunXrirp16wpvb28RHx9fblyVcfnyZTFhwgTRuHFjYWNjI/r06SO2bNmiUU5b0lSvXr0yR90/PhvAtWvXyh2lX1BQoCqbkJAgAIiIiIgK4//777/Fiy++qPo+Vs5Gk5KSoipz//591TnKZDK179Xz58+LUaNGiYYNG4qGDRuKF198UWRnZ4sOHTqoJbGVOdeqqMrvozLHV/Z8xo4dqzajCelPJkQFzw6IKmHixIlITEzE5cuXTR1Ktbly5Qratm2LUaNGaX1caSwymQzvvvsuPv7442r7mVUhtXipeikUCvTp0wdeXl746quvTB1OjbF48WJ89dVXOHv2LOrXr2/qcGq9K1euoEOHDli/fr3GjDFUeewTS1RFyn6906dPN3UoRJJlYWGBTz/9FN999x0uXrxo6nBqjD179iA0NJQJrJn46KOP0KdPHyawBmJVcREiKsulS5fwxRdfoGfPnhg+fLipwyGStCFDhug1YJDKtm/fPlOHQI/gUwbDYksskZ7c3d3h6uoKR0dHREdHG2XqISIiItKOfWKJiIiISHLYEktEREREksMkloiIiIgkhwO7/kuhUODq1ato0KAB+zYSERERmYgQAgUFBWjZsiUsLMpub2US+19Xr15FmzZtTB0GEREREQHIyspC69aty9zPJPa/GjRoAODhBbO3tzdxNERERES1U35+Ptq0aaPKzcrCJPa/lF0I7O3tmcQSERERmVhF3Ts5sIuIiIiIJIdJLBERERFJDpNYIiIiIpIcJrFEREREJDlMYomIiIhIcpjEEhEREZHkmDSJLSgoQFhYGAYPHozevXtj8uTJ+OuvvzTKbdmyBUOHDoWHhwfGjRuHM2fO6FWGiIiIqDaRy+VITExEdHQ0EhMTIZfLTR2SwZg0iX399dehUCjw8ccfY/Xq1ahTpw4GDhyIs2fPqsrEx8cjKCgII0eOxHfffQchBHx8fHD79u1KlSEiIiKqTeLi4uDq6gpfX1+MHz8evr6+cHV1RVxcnKlDMwiZEEKY6oeXlpbCyup/6y3I5XLY2tpizZo1mDp1KgCgT58+8PT0xA8//AAAKCkpgZOTE+bMmYMFCxboXKYi+fn5cHBwQF5eHhc7ICIiIkmLi4tDUFAQ/Pz8EBoaCg8PD6SlpSEsLAzx8fGIiYlBYGCgqcPUSteczKQtsY8msMDDLgEymQz9+vUD8LC7wfHjx/HMM8+oylhbW2PYsGFITEzUuQwRERFRbSGXyxEcHAw/Pz9s3rwZXl5eqF+/Pry8vLB582b4+flh7ty5ku9aYPKBXUlJSXB1dUXz5s0xbdo0/Pbbb/Dw8AAAXL58GUIItGjRQu2YFi1a4PLlyzqX0aa4uBj5+flqLyIiIiKpS0pKwoULFxAaGgoLC/VUz8LCAiEhITh//jySkpJMFKFhmDyJ7d27N3bs2IHffvsNzz//PCZNmoSMjAwAgEKhAADUqVNH7Rhra2vV3YMuZbQJDw+Hg4OD6tWmTRuDnRMRERGRqVy7dg0AVI2Cj1NuV5aTKpMnsXXr1oWrqyu8vLzw3XffoVmzZvjyyy8BAE2bNgUA3Lp1S+2YmzdvqvbpUkabkJAQ5OXlqV5ZWVkGOyciIiIiU3FycgIApKWlad2v3K4sJ1UmT2IfJZPJYG9vj4KCAgCAo6Mj2rRpg5SUFLVyycnJ6NOnj85ltLGxsYG9vb3ai4iIiEjqfHx84OzsjLCwMNUTayWFQoHw8HC4uLjAx8fHRBEahsmS2MLCQoSEhCA3NxfAw4v6/fff4/Dhw3j++edV5WbMmIFvv/0W6enpAIC1a9fi3LlzmD59eqXKEBEREdUGlpaWiIiIQHx8PAICApCSkoKCggKkpKQgICAA8fHxWL58OSwtLU0dapVYVVzEOOzs7ODo6AgPDw8IIZCfnw8nJyf8+OOPGDVqlKrcu+++i8uXL6N79+6oV68eLC0tERkZCU9Pz0qVISIiIqotAgMDERMTg+DgYHh7e6u2u7i4mPX0WpVh0nlilW7cuIH69eujXr16ZZa5d+8ebt++DScnJ42puSpTpiycJ5aIiIhqGrlcjqSkJFy7dg1OTk7w8fEx+xZYXXMys0hizQGTWCIiIiLTk8RiB0RERERE+mASS0RERESSwySWiIiIiCSHSSwRERERSQ6TWCIiIiKSHCaxRERERCQ5TGKJiIiISHKYxBIRERGR5DCJJSIiIiLJYRJLRERERJLDJJaIiIiIJIdJLBERERFJDpNYIiIiIpIcJrFEREREJDlMYomIiIhIcpjEEhEREZHkMIklIiIiIslhEktEREREksMkloiIiIgkh0ksEREREUkOk1giIiIikhwmsUREREQkOUxiiYiIiEhymMQSERERkeQwiSUiIiIiyWESS0RERESSwySWiIiIiCSHSSwRERERSQ6TWCIiIiKSHCaxRERERCQ5TGKJiIiISHKYxBIRERGR5DCJJSIiIiLJYRJLRERERJJjZeoAAODevXuwsrKCjY2Nxr47d+7g7t27atvq1KkDR0dHjbIlJSXIz89HkyZNIJPJjBYvEREREZmWSVtiN2zYgJ49e8LJyQkNGzZE//79ceTIEbUy8+fPh5ubG7y8vFSvCRMmqJVRKBQIDg5Gw4YN4ezsjNatW+PXX3+tzlMhIiIiompksiRWLpdj+/bt+PHHH5GXl4c7d+6ge/fuGD58OG7fvq1W9rnnnsPly5dVr927d6vtj4iIwA8//IADBw4gPz8f7777LsaOHYtTp05V5ykRERERUTUxWRJraWmJyMhIdO/eHTKZDDY2Nvjggw9w69YtjdZYALh9+zZKSkq01rVq1Sq88sor6NGjBywsLPDmm2+ibdu2WLt2rbFPg4iIiIhMwKwGdp05cwYA0KJFC7Xtv/76K1xcXFC/fn14e3vj+PHjqn3Z2dm4ePEiBgwYoHbMwIEDcfjwYeMHTURERETVrspJbFFRkSHiQGFhId588034+vqie/fuqu19+vTB0aNHkZeXh5ycHLRv3x5PPfUUrl+/DgDIyckBADRt2lStvqZNm6r2aVNcXIz8/Hy1FxERERFJg15J7IMHD7Bw4UI4OTmhbt26qu0zZsxAZmZmpesrKSlBUFAQCgsLERUVpbbvlVdeQc+ePQEADg4O+Oabb1BcXIyYmJiHJ2Dx8BRKS0s1YrS0tCzzZ4aHh8PBwUH1atOmTaXjJiIiIiLT0CuJXbJkCbZs2YJVq1apbff19cXixYsrVZcygc3IyMDevXs1uhI8zs7ODk5OTrhw4QIAoFWrVgCgaplVunHjhmqfNiEhIcjLy1O9srKyKhW3qcjlciQmJiI6OhqJiYmQy+WmDomIiIio2umVxK5btw5RUVEIDAxU2z548GD8/vvvOtfz4MEDjBkzBv/++y8SExPRunVrjTJCCLX3165dw8WLF9G+fXsAgL29PXr06IFdu3apypSWlmLPnj0YNGhQmT/bxsYG9vb2ai9zFxcXB1dXV/j6+mL8+PHw9fWFq6sr4uLiTB0aERERUbXSK4m9evUqOnToAABqiwpYWlri/v37OtWhUCjw4osv4uDBg4iMjAQA1RRahYWFAB72W/X29kZcXBxOnz6NXbt2YeTIkWjdujUmTpyoquu9997DTz/9hDVr1uDvv//G1KlTAQCvv/66PqdnluLi4hAUFARPT0+kpKSgoKAAKSkp8PT0RFBQEBNZIiIiqlVk4vGmTh1069YNy5cvx9NPPw1LS0vVI+2IiAj88ssvOHjwYIV15ObmwtPTU+u+Tz75BOPGjQMApKamYtmyZThx4gQaNWoEHx8fzJ8/H40aNVI7ZtOmTfj8889x48YNeHp6IiwsDF26dNH5nPLz8+Hg4IC8vDyza5WVy+VwdXWFp6cnNm/erOoHDDy8GQgICEBaWhoyMzPL7QdMREREZO50zcn0SmI3bdqEN954A6GhoQgODkZUVBR27NiB9evXIzY2Fv7+/lUK3hTMOYlNTEyEr68vUlJS4OXlpbE/JSUF3t7eSEhIwJAhQ6o/QCIiIiID0TUns9Kn8rFjx8LCwgJLlixRdQvo0qULoqOjJZnAmrtr164BADw8PLTuV25XliMiIiKq6fRKYgFgzJgxGDNmDEpKSiCEgI2NjSHjokc4OTkBANLS0rS2xKalpamVIyIiIqrpqrzYgbW1NRNYI/Px8YGzszPCwsKgUCjU9ikUCoSHh8PFxQU+Pj4mipCIiIioeunVEqutNfBRugzsIt1ZWloiIiICQUFBCAgIQEhICDw8PJCWlobw8HDEx8cjJiaGg7qIiIio1tArifXz81N7r1AokJmZiY0bN2LGjBkGCYzUBQYGIiYmBsHBwfD29lZtd3FxQUxMjMacvUREREQ1mV6zE5Rlw4YN+O2337Bp0yZDVVltzHl2gkfJ5XIkJSXh2rVrcHJygo+PD1tgiYiIqMYw6hRbZbl79y7at2+P7OxsQ1VZbaSSxBIRERHVZLrmZFUe2PWo1NRUtYn4iYiIiIiMQa8+sVOmTNHYlpubi127duGNN96oakxEREREROXSK4ktLS3V2Na2bVv8+OOPGDNmTJWDIiIiIiIqj15JbGRkpKHjIB1xYBcRERGRgfvEknHFxcXB1dUVvr6+GD9+PHx9feHq6oq4uDhTh0ZERERUrXRuiR0yZIjOlSYmJuoRCpUnLi4OQUFB8PPzQ3R0tGqxg7CwMAQFBXGuWCIiIqpVdJ5ia+HChTpXumTJEr0DMhVznmJLLpfD1dUVnp6e2Lx5s9oMEAqFAgEBAUhLS0NmZia7FhAREZGkmWSeWCkz5yQ2MTERvr6+SElJ0brkb0pKCry9vZGQkFCpFnMiIiIic2OSeWLJOK5duwYA8PDw0LpfuV1ZjoiIiKim02t2AgDIycnBrl27cOnSJY0ptyrT9YAq5uTkBABIS0vT2hKblpamVo6IiIioptOrO0FycjJGjBiBFi1a4PTp0+jevTsyMjJQVFSEXr164dixY8aI1ajMuTsB+8QSERFRbWHU7gTz5s3D+++/j/T0dAAPl5u9cuUK/Pz84Ovrq1/EVCZLS0tEREQgPj4eAQEBSElJQUFBAVJSUhAQEID4+HgsX76cCawByOVyJCYmIjo6GomJiZDL5aYOiYiIiLTQqyW2QYMGuHLlCuzt7WFpaYnCwkLY2Njg4sWLeOKJJ3D16lVjxGpU5twSqxQXF4fg4GBcuHBBtc3FxQXLly/n9FoGoO36Ojs7IyIigteXiIiomhi1Jfbu3buqSh0dHXHx4kUAgL29PXJzc/WpknQQGBiIM2fOICEhAVFRUUhISEBmZiYTLANQzsPr6emp1tLt6emJoKAgLihBRERkZvRqiZXJZFAeNmnSJNy+fRvTp0/Hhg0bcPXqVSQnJxs8UGOTQkssGQf7HBMREZkPo7bErly5UvXvZcuWQaFQYMqUKcjKysLatWv1qZLIZJKSknDhwgWEhoaqJbAAYGFhgZCQEJw/fx5JSUkmipCIiIgep9cUW7NmzVL928nJCdu3bzdYQETVjfPwEhERSY9eLbGurq5YtGgRMjMzDR0PUbV7dB5ebTgPLxERkfnRK4mdOnUqYmJi4Obmhv79+2PlypXIyckxdGxE1cLHxwfOzs4ICwuDQqFQ26dQKBAeHg4XFxf4+PiYKEIiIiJ6nF5JbGhoKP755x+cOHECgwYNwtKlS9GyZUuMGDEC0dHRho6RyKg4Dy8REZH06DU7weMUCgUSExMxZ84c/PXXXzBAldWOsxMQ5+ElIiIyPV1zMr0Gdj3q33//xYYNGxAVFYWsrCw89dRTVa2SyCQCAwPh7++PpKQkXLt2DU5OTvDx8WELLBERkRnSK4m9cuUKoqOjsWHDBqSmpqJXr1548803MW7cOLRo0cLQMRJVG0tLSwwZMsTUYRAREVEF9Epi27Zti3bt2mH8+PGIjo6Gu7u7oeMiIiIiIiqTXklsUlISvL29DR0LEREREZFO9JqdgAksEREREZmSXkksEREREZEpVXl2Aqpecrmco+eJiIio1jN5Env79m0cO3YMVlZW6NGjBxo1aqRRRqFQ4NChQ7hx4wY8PDzg6uqqVxmp0zaPqbOzMyIiIjiPKREREdUqJutOIITA66+/Dk9PT3zyySd477330K5dO3z//fdq5fLy8jBgwAAEBQVh1apV6N69O+bPn1/pMlIXFxeHoKAgeHp6qq0o5enpiaCgIMTFxZk6RKIKyeVyJCYmIjo6GomJiZDL5aYOiYiIpEro4fbt22L27Nmib9++okOHDhovXcjlcvH111+L4uJi1bavvvpKWFlZiYsXL6q2zZw5U7i5uYk7d+4IIYRISkoSAMTu3bsrVaYieXl5AoDIy8vT+ZjqUlpaKpydncXIkSOFXC5X2yeXy8XIkSOFi4uLKC0tNVGERBWLjY0Vzs7OAoDq5ezsLGJjY00dGhERmRFdczK9lp0NCgrCP//8g8mTJ2t9/D9jxgy9Eurs7Gw4Ojri999/x3PPPQchBBo3boyQkBC88847qnL9+/dHly5d8MMPP+hURhfmvOxsYmIifH19kZKSAi8vL439KSkp8Pb2RkJCAifqJ7OkfJLg5+eH0NBQeHh4IC0tDWFhYYiPj0dMTAy7xBAREQAjLzu7a9cunDhxAu3bt9c7QG12794NmUyGLl26AAAuX76MO3fuwMPDQ62cp6cnUlNTdS6jTXFxMYqLi1Xv8/PzDXMSRnDt2jUA0DhHJeV2ZTkicyKXyxEcHAw/Pz9s3rwZFhYPezF5eXlh8+bNCAgIwNy5c+Hv789BikREpDO9+sQ2bNhQawtsVVy4cAGzZ8/G66+/DmdnZwAP+7oCQOPGjdXKNmnSBHfu3NG5jDbh4eFwcHBQvdq0aWOYEzECJycnAEBaWprW/crtynJE5iQpKQkXLlxAaGioKoFVsrCwQEhICM6fP4+kpCQTRUhUMfbnJjI/eiWxY8aMwWeffWawIK5evYqnnnoK/fv3V6vXxsYGAHD37l218nfv3oWtra3OZbQJCQlBXl6e6pWVlWWIUzEKHx8fODs7IywsDAqFQm2fQqFAeHg4XFxc4OPjY6IIicrGJwkkdXFxcXB1dYWvry/Gjx8PX19fuLq6ckAtkYnplcSmpqbi//7v/+Du7o4RI0bAz89P7VUZV69eha+vL9zc3BATE4M6deqo9rVt2xZWVla4dOmS2jEXL15UdWXQpYw2NjY2sLe3V3uZK0tLS0RERCA+Ph4BAQFqsxMEBAQgPj4ey5cv56NYMkt8kkBSxplhiMyXXgO7Zs+eXe5+XVtpr127hiFDhqjuaJWtqo969tlnIZPJsH37dgDAzZs30bZtW6xYsQKvvfaazmUqYs4Du5S0zRPr4uKC5cuXc1AMmS25XA5XV1d4enqq9YkFHj5JCAgIQFpaGjIzM3kjRmaFn10i09A1J9MriTWE4uJi9OzZE7du3cKyZcvUEth+/fqpWlFTU1MxcOBAPP/883jiiSfw7bffQiaT4cCBA6pjdClTESkksQBX7CJpenR2gpCQENXsBOHh4ZydgMwWZ4YhMg2jzk5gCCUlJejWrRsAqFpQlRwdHVVJbI8ePXD8+HF8++23OHToEMaPH4/XX39dLTnVpQwRmU5gYCBiYmIQHBwMb29v1XYXFxcmsGS22J+byLzp3RK7a9cuLF26FKdOnYIQAl26dMG7776Lp556ytAxVgsptMRy2VmSOj5JIClhSyyRaeiak+k1sOunn37Cc889hxYtWiAkJAQLFixAixYt8Nxzz2HdunV6B01l4+ACqgksLS0xZMgQjBs3DkOGDGECS2aNM8MQmTe9WmI7deqEBQsWYPLkyWrbf/rpJ4SFheH06dMGC7C6mHNL7KODC2JjY5GcnKxqyRowYABGjx7NwQVEREbA/txE1c+oLbHnz59HQECAxvbnn38e58+f16dKKodysnhvb2+4ubmpzVXo5uaGJ554gpPFExEZgbI/98mTJ+Ht7Q17e3t4e3sjLS2NCSyRiemVxLZq1Qp//vmnxvaEhAS0bt26ykGROuWggdDQUK3dCRYsWKBWjoiIDCcwMBBnzpxBQkICoqKikJCQgMzMTCawRCam1+wEb731FiZMmID//Oc/6NevHwDg0KFD+Oqrr/Dhhx8aNEACmjdvDgAYMGCA1rXnBw8ejP3796vKERGRYSn7cxOR+dAriZ09ezYaN26MZcuW4dNPPwUAuLm5YdWqVRr9ZMn4TDTVLxEREZHJ6NWdAAAmT56MtLQ0FBUVoaioCGlpaUxgjSQ7OxsAsH//fq3LziYnJ6uVIyIiIqrp9E5ilSwtLTki3siUa8qHh4drHVwQFhamVo6IiIioptO5O4GyL1BiYmKF/YISExOrEBI9TjlX4YEDB5CRkaF1ii3OVWgYnIyfiIhIGnROYgcOHKj132R8lpaWiIiIQFBQEEaPHo2QkBD4+fkhLS0No0ePVs1VyGSrargiGhERkXTonMQuWbJE9e8ePXogKChIa7mYmJiqR0UalHMVzpkzR23teWdnZ85VaACPTmgeHR2tmtA8LCwMQUFBvMZERERmRq8+sWPGjNFrH1WdTCYzdQg1jlwuR3BwMPz8/LB582Z4eXmhfv36qinM/Pz8MHfuXMjlclOHSkRERP9V5YFdj8rOzoaDg4Mhq6T/UrYUalvsICgoCHFxcaYOUbKUK6KFhoaq5uBVsrCwQEhICFdEIyIiMjOVmid22LBhWv8NAAqFAhkZGZwM2ggebyl8fLGDgIAAzJ07F/7+/uwXqwflSmceHh5a9yu3c0U0IiIi81GpJLZPnz4AgD179qj+rVSnTh1MmDABL774ouGiIwD/aymMjo4us6XQ29sbSUlJvInQg3JqsrS0NPTt21djdoK0tDS1ckRERGR6lUpiP/74YwBA06ZNMXfuXKMERJrYUmhcyinM3njjDdy8eVNjdoKmTZtyCjMiIiIzo1efWCaw1evRlkK5XI7ExERER0cjMTERcrmcLYVVZGlpiTFjxuDo0aMoKirC2rVrcfXqVaxduxZFRUU4evQogoKC2FWDiIjIjMiEEKKyB0VHR2Pr1q2IiopS2z5u3DgEBARg7NixBguwuuTn58PBwQF5eXmwt7c3dThq5HI5XF1d0bRpU+Tk5ODixYuqfe3atUOzZs1w69YtZGZmMtHSQ3nXV9kSy+tLRERUPXTNyfRqif2///s/LFq0SGP7Bx98gA8//FCfKqkcj7YU3r9/X62l8P79+2wprCJln+OVK1fi7NmzSEhIQFRUFBISEnDmzBl88cUXnJ2AiIjIzOjVEmtra4vr16+jYcOGattzc3Ph5OSE+/fvGyq+aiOVltjH+2y6uLigSZMmbCmsgujoaIwfPx4FBQWoX7++xv6CggLY29sjKioK48aNM0GEREREtYdRW2I7d+6sdWWun3/+GZ06ddKnSirHoy2FZ86cUWspzMzMZEthFT3a51gb9jkmIiIyP5WanUBp/vz5eOmll/D3339j0KBBEEJg3759WLt2LX766SdDx1jrPTo7gaWlpcY0WpydoGqUsxOEhYWpzcMLPJz/ODw8nLMTEBERmRm9ktixY8dCoVBgyZIlWLlyJQCgS5cu+PHHHzlPrBFwHlPjsrS0REREBIKCghAQEICQkBB4eHggLS0N4eHhiI+PR0xMDLtqEBERmRG9+sQ+qri4GDKZDNbW1oaKySSk0ieWsxMYT1xcHIKDgzX6HC9fvhyBgYGmC4yIiKgWMWqf2EfZ2NhIPoE1d5ydoHoEBgZq7XPMBJaIiMj86N0Sm5OTg127duHSpUsoLS1V27dw4UKDBFedpNISy9kJiIiIqCbTNSfTq09scnIyRowYgRYtWuD06dPo3r07MjIyUFRUhF69ekkyiTVnytkJoqOjtfaJPXz4MLy9vZGUlKQx6IuIiIioJtIriZ03bx7ef/99zJkzBzKZDKmpqcjNzcXkyZM5xZYRcHaC6iOXyzVuEti6TUREZH706hN78uRJvPLKKw8rsLBAcXExGjVqhC+//FJjKVqqOs5jWj3i4uLg6uoKX19fjB8/Hr6+vnB1dUVcXJypQyMiIqLH6JXE3r17V9VHwdHRUTVa3t7eHrm5uYaLjgCoz2OqUCjU9nEeU8OIi4tDUFAQPD09kZKSgoKCAqSkpMDT0xNBQUFMZImIiMyMXgO7ZDIZlIdNmjQJt2/fxvTp07FhwwZcvXoVycnJBg/U2Mx5YBfwvyTLz8+vzHlMOYpeP8qBc56enloXOwgICEBaWhoHzhEREVUDo06xpVzgAACWLVsGhUKBKVOmICsrC2vXrtWnSqpAYGAgYmJicPLkSXh7e8Pe3h7e3t5IS0tjAltFyoFzoaGhagks8LC7TEhICJf1JSIiMjN6DeyaNWuW6t9OTk7Yvn27wQKisgUGBsLf358Djwzs0YFz2nDgnOFw4BwRERmKXi2x9+7dQ2xsrOr99u3b8eyzz2LmzJm4d++ewYIjTcrZCcaNG4chQ4YwATAADpyrHhw4R0REhqRXErtgwQJVq1Rubi7Gjh0LJycnHDx4EPPmzTNogETGxoFzxseBc0REZGh6Dexq3bo1jh8/jubNm2Pjxo1YvXo1EhMTcfbsWQwaNAhXrlzRua68vDysX78e6enpeOutt9CxY0e1/Zs2bdLoi9iyZUuEhoaqbbt27RoiIyNx48YNeHp6Yvz48ahTp47OcZj7wC4yLg6cMx4OnCMiosow6sCuvLw8WFtbAwD27t2LZ555BgDQokUL5OXl6VzP2rVr0blzZ6SkpGDVqlVak9+EhAQkJyfD3d1d9XJxcVErk5GRAU9PTyQkJKBBgwb46KOP8PTTT0Mul+tzelQLceCc8XDgHBERGYNeA7t69eqFRYsW4ZlnnsGmTZuwb98+AEBqaip69uypcz1PPPEEMjMzkZubW+4iCR06dFAbTPa4d999Fx4eHvj9998hk8kwbdo0dOjQARs2bMDkyZN1PzGq1Thwzjg4cI6IiIxBryT2s88+w/jx4/HNN9/gzTffRPfu3QEAS5cuxTvvvKNzPZ6engBQ4QIJZ86cwTvvvAMHBwf4+Phg0KBBqn0PHjzA9u3b8cUXX0AmkwF42N1hyJAh2LJlC5NYqhRty/pS1Tw6cM7Ly0tjPwfOkRRwZg0i86NXEtuzZ0+cOnVKY/v69evh4OBQ5aAeJZPJ4OjoiIYNG+LKlSt49tln8dJLL+Hrr78GAFy6dAnFxcUaXQzat29f7qILxcXFKC4uVr3Pz883aNxE9NCjA+e09YnlwDkyd3FxcQgODsaFCxdU25ydnREREcGuRkQmpFef2LIYOoEFgNDQUOzcuROhoaFYtWoVtm/fjtWrV2PXrl0AgKKiIgBAgwYN1I6zt7dHYWFhmfWGh4fDwcFB9WrTpo3BYyeih63bERERiI+PR0BAgNrsBAEBAYiPj8fy5cvZqkVmiTNrEJkvvZLY3NxcvP322+jXrx9cXV01Xob0eHI5ePBgtGnTBvv37wcA1K9fHwBw584djRjLG9EWEhKCvLw81SsrK8ugcRPR/3DgHEmRXC5HcHAw/Pz8sHnzZnh5eaF+/frw8vLC5s2b4efnh7lz53IQMZGJ6NWdYPr06fjnn38wefJkNGrUyNAxVaikpASlpaUAgLZt26J+/fo4deoUnn32WVWZU6dOoUuXLmXWYWNjAxsbG6PHSkQPceAcSY1yZo3o6OgyZ9bw9vZGUlIS+9ITmYBeSeyuXbtw4sQJtG/f3tDxqCktLcWff/6JoUOHqratW7cON27cUE3rZWFhgTFjxuDHH3/EjBkzYGdnh9TUVBw4cKBSg8yIyPg4cI6khDNrEJk3vZLYhg0bGqQF9tChQ1i/fr1qqdrPP/8cMTExeO655/Dcc89BJpMhIiICISEh6Nq1Ky5duoSUlBSEh4erzVDw8ccfY8iQIejduzd69OiBHTt2YMqUKRg5cmSVYyQikgKOnjc8zqxBZN70WrFr7ty5qFevHhYvXlylH56eno7du3drbO/Xrx/69eunev/333/jxIkTaNSoEfr27av1C6O4uBg7d+5Urdil7QunPFyxi4ikiqPnjYOrzVFNIMUbXF1zMr2S2GHDhmHPnj3o1KkTOnTooJqfVSk+Pr7yEZsYk1gikqJHl0wODQ1VLZkcFhbGJZMNgEtSk5RJ9QbXqEns7Nmzy93/2WefVbZKk2MSS0RSw5bC6qEtEXBxccHy5cvNOhGg2k3KN7hGTWJrIiaxBEjzsYuU8PoaVmJiInx9fZGSkqK1C1VKSgq8vb2RkJDAAXVVxM8uSYnUb3B1zcn0GthFVBNJ9bGLVPD6Gh5Hz1cfzqxBUvLo9HBCCCQmJqrdgNWU6eH0XrErJycHUVFR+Pjjj7FkyRK1F5HUcFUe4+L1NY5HR89rw9HzRLWT8sb17NmzcHV1ha+vL8aPHw9fX1+4urri3LlzauWkSq/uBMnJyRgxYgRatGiB06dPo3v37sjIyEBRURF69eqFY8eOGSNWo2J3gtpL6o9dzB2vr/Hw2hKRNsquRjKZrMw+sUIIs+1qpHNOJvTwxBNPiIiICPHfBFgIIcTt27eFn5+fCA4O1qdKk8vLyxMARF5enqlDoWqWkJAgAIiUlBSt+w8cOCAAiISEhOoNrIbg9TWu2NhYIZPJxMiRI8WBAwdEfn6+OHDggBg5cqSQyWQiNjbW1CESUTUrLi4WVlZWwtHRUTx48EBt34MHD4Sjo6OwsrISxcXFJoqwfLrmZHp1Jzh58iReeeUVAA9XzCouLkajRo3w5ZdfIioqSp8qiUyG/QqNi9fXuAIDAxETE4OTJ0/C29sb9vb28Pb2RlpamlmPPiYi4zlw4ABKS0tx48YNBAYGqnXjCgwMxI0bN1BaWooDBw6YOtQq0SuJvXv3rqp519HRERcvXgQA2NvbIzc313DREVUD9is0Ll5f4wsMDMSZM2eQkJCAqKgoJCQkIDMzkwksUS2lbBSIjIzUeoMbGRmpVk6q9OoTK5PJoDxs0qRJuH37NqZPn44NGzbg6tWrSE5ONnigxiaVPrGc5sXw2K/QuHh9iYiq16PT7/Xt21cjbzh8+LBZT7+na06mV0vsypUrVf9etmwZFAoFpkyZgqysLKxdu1afKkkHcXFxWkcZcmR31VhaWiIiIgLx8fEICAhQe+wSEBCA+Ph4LF++nAmWnnh9iYiql4+PD5ydnREWFgaFQqG2T6FQIDw8HC4uLvDx8TFRhAZSHR10pcDcB3Y9OngjJSVFFBQUiJSUFA7eMKDY2Fjh7OwsAKheLi4uvLYGwutLRFR9lHmDnZ2d2veunZ2d2ecNuuZkenUnaNasGXJycgyWSJsDc+5O8Ojj2NjYWCQnJ6seCwwYMACjR4/m41gDYXcN4+L1NR5eWyJ6VFxcHEaPHg07OzsUFRWpttetWxeFhYWIjY01237zRl12tmnTpsjIyEDjxo2rFKQ5MeckVtm3JTw8HGvWrNFY8ejVV19FaGio2fZtISLj4mpoRPQoZeNX06ZNkZ2djUuXLqn2tW3bFs2bN8etW7fMtvHLqH1iJ06ciKVLl2r0syDjUI4eVE5WvGrVKnz//fdYtWoVPDw8sGDBArVyRFR7cDW06iGXy5GYmIjo6GgkJiZCLpebOiSiMimXnT169Ci6d++u9t3QvXt3HD16FOfPn0dSUpKpQ60SK30OSktLw549exAdHQ13d3dYW1ur7Y+PjzdIcPRQ8+bNAQCdOnXCyZMn1a5vu3bt0KlTJ6Snp6vKEVHtIJfLERwcDD8/P7WZH7y8vLB582YEBARg7ty58Pf3N8vWFqlgSzdJzZUrVwAAw4cP1/rd4Ofnh+3bt6vKSZVeSayHh0eZE5eT8aSnp8POzk5tW3Z2tlpfFyKqPZStLdHR0WpTlwEPF6IJCQmBt7c3kpKS2NVIT8qWbj8/P0RHR6st3RkUFMQFJcgsKcctBQYGav1uCAgIwPbt2yU/vkmvJHbgwIEICgrSui8mJqZKAZGm69evq/5tb2+Pzz//HH5+foiPj8d7772nSmIfLUdENR9XQzMutnSTVDVr1gzAw5uwqVOnaszPvXnzZrVyUqVXn9gxY8botY/0c+PGDQAPuw7Y2tri1VdfRcuWLfHqq6/Czs4O7dq1UytHRLUDV0MzLmVLd2hoaJkt3TWhXyHVPK1atQIA7NixQ+v83Dt27FArJ1V6tcSWJTs7Gw4ODoaskgDcunULwMM/RPv27dOYYmvQoEG4ePGiqhwR1Q6PTmiubTW0GjOhuYmwpZukSvnd0LRpU9Wys0ouLi7o3bs3bt26JfnvhkolscOGDdP6b+DhF2ZGRgb7XRmB8g/TwYMHERgYiGeffRZ2dnb4559/EBERgYMHD6qVI6LaQbkaWlBQEAICAhASEqLqsxkeHo74+HjExMTwUbeeHm3p9vLy0tjPlm4yV49+N4wYMQJz585VzRe7Y8cO/P777zXiu6FS88TOnz8fALB06VK8++67avvq1KkDZ2dnvPjii6hXr55ho6wG5jxP7J49ezBs2DC0atUK169fV5vaxdLSEi1atMCVK1ewe/duDB061ISREpEpaBs97+LiguXLl3PQURU8utCMtpbugIAALjRDZk2q3w1GXexg+fLlmDt3bpUCNDfmnMTK5XI0adIEeXl5aNasGXx9fVUrbiQkJCAnJwcODg64desWv0iJaimu2GUcj85OUFZLtzknA0RS/G4wahJbE5l7Euvk5KR1KgyZTAYhBJo3b46rV6+a/QeTajcpfpkSSbU1i0iqmMRWkjknscplZwGUuQYyAC47S2aNE8aTlPEGjKj6GHXZWapej668kZeXh4SEBERFRSEhIQF37tzB8OHD1coRmRsujUpSZ2lpiSFDhmDcuHEYMmQIE1giM2DQKbbIOB5deaNOnToara01ZeUNqpk4YTwRERlDlVtiueSp8T268oZCoVDbV5NW3jAHJSUl+Oyzz/DGG2/gs88+Q0lJialDkjxOGE9ERMagVxL74MEDLFy4EE5OTqhbt65q+4wZM5CZmWmw4Oih2rLyhqm98847qFevHt5++218+eWXePvtt1GvXj288847pg5N0jhhPBERGYNeSeySJUuwZcsWrFq1Sm27r68vFi9ebJDA6H+UK2/07t1btfKGvb09vL29kZaWht69e3NVnip655138Mknn2ht6f7kk0+YyFYBl0YlIiJj0Gt2AhcXF/z222/w9PRUTfEEANevX0fnzp2Rm5tr8ECNzZxnJwD+NzBmxIgRqhW7Hl95gyO89VNSUgI7OzsoFAqMGDECCxcuVM0FuWTJEvz++++wsLBAUVERrK2tTR2u5HDCeCIiqgydczKhB2tra3Hv3j0hhBAymUy1PTs7W9ja2upTpcnl5eUJACIvL8/UoZQpNjZWODs7CwCql4uLi4iNjTV1aJK2fPlyAUB069ZNyOVytX1yuVx4enoKAGL58uUmilD6YmNjhUwmEyNHjhQHDhwQ+fn54sCBA2LkyJFCJpPxM0xERCq65mR6dSfo1KkT9u/fD+DhZPtK69atQ/fu3fWpknQQGBiIM2fOqE2xlZmZyRbYKlJ+lsPCwiCEQGJiIqKjo5GYmAghBJYsWaJWjiovMDAQMTExWrvD8CkCEZHxyOVytb9rjy5dL3V6TbG1YMECTJw4EaGhoQCATZs2YceOHVi/fj1iY2MNGiCpk8vlSE1NxdmzZ9GhQwd4e3vzEWwV1a9fHwDw66+/YtasWRqT8T/55JNq5Ug/gYGB8Pf354TxRETVpMYvMqNvU+/PP/8sunXrpnqs3aVLF/Hzzz/rW53JSaE7wbx584SlpaVadwJLS0sxb948U4cmaTt37lRdzxEjRoiUlBRRUFAgUlJSxIgRI1T7du7caepQiYiIdPJoN65H/65JoRuXrjlZlZedLSkpgRACNjY2VUqmTc3cB3YpR89bWFiojaBXvp83bx6WLVtmwgilq6SkBLa2thBCoHnz5pg4cSI6dOiAs2fPIjIyEtnZ2ZDJZLh//z4HdhERkdmT+oBaXXOyKiexVSGEwJ49e7B69Wqkp6fjhx9+QN++fTXK7dixAytXrsSNGzfg6emJRYsWwdnZudJlymPOSeyjo+etra3VJuBXvufoef0lJibC19e3wnIJCQkaq6URmQu5XM6uGkQE4H9/11JSUuDl5aWxPyUlBd7e3mb7d03XnEyvgV1eXl5lvgYPHoyXX34Zf/75Z4X1zJ8/H+Hh4RgyZAj++ecf3Lt3T6PMjh07MHLkSPj4+CAiIgJ37tzBwIEDcefOnUqVkbKVK1eqWl8fPHigtk/5XqFQYOXKldUeW02gnGRfJpPBzs5ObZ+dnZ1q8CIn4ydzFRcXB1dXV/j6+mL8+PHw9fWFq6sr4uLiTB0aEZlAbVlkRq8ktm/fvjh8+DAcHBzg6+uLJ598Evb29jh06BDc3Nxw584dPPnkk9i6dWu59XzwwQfYs2cPAgICyiyzaNEivPjii5g/fz4GDx6MjRs34t69e1i9enWlykjZo8txNm/eHN988w2uXbuGb775Bs2bN9dajnSnvIYDBgzA7du3sWLFCsyaNQsrVqzA7du3MWDAALVyROZEOYe0p6en2mp+np6eCAoKYiJLVAvVmkVm9OlwO2rUKLF69WqN7V999ZXw9/cXQgixatUq0bt3b53qy8rKEgBEQkKC2vaCggIhk8lEVFSU2vagoCDx9NNP61xGF+Y8sGvo0KECgKhbt6548OCB2r4HDx6IunXrCgBi6NChJopQ2nbv3i0ACHd3d9GuXTu1gXPt2rUT7u7uAoDYvXu3qUMlUlNaWiqcnZ3FyJEjRUlJiUhISBBRUVEiISFBlJSUiJEjRwoXFxdRWlpq6lCJqBo9+t2gbf5zc/9uMOo8sfv378e4ceM0to8fP141l+aLL76I06dP61O9ypUrVyCE0LhTcHJyQlZWls5ltCkuLkZ+fr7ay1wJHbst61qO1GVnZwMA0tPTcf/+faxduxZXr17F2rVrcf/+faSnp6uVI/3V5PkKTSEpKQkXLlyAt7c33Nzc1LoTuLm54YknnsD58+f5lIaolrG0tERERATi4+MREBCg9pQmICAA8fHxWL58ueT7zeuVxMpkMhw+fFhj+6FDh1T9B/Py8tCyZcsqBVdaWgoAGoOVbGxsVH1BdSmjTXh4OBwcHFSvNm3aVClWY2rQoAEAoLCwEK1bt1ZLslq3bo3CwkK1clQ5ym4C7u7usLOzw6uvvoqWLVvi1VdfRd26deHu7q5WjvTDfpuGp+zPFhISorU7gXIub6n3eyOiyqsNi8zotdjBjBkz8MILL+CNN95Anz59IITAsWPH8MUXX2DmzJkAgLVr12L69OlVCq5JkyYAgFu3bqltv3XrFpo2bapzGW1CQkIwZ84c1fv8/HyzTWR9fHywZcsWAMCNGzfw2muvlVmO9Ne0aVPs3bsXycnJqhHeAwYMUC12QPpT9tscMWIE5s2bBzs7OxQVFWH79u0ICgqqMV+o1U15YzVw4EC1aXS8vLywefNmDBo0CMnJybwBI6qlAgMD4efnh6+++kq1SNJ//vOfmjOTkT59FRQKhfj666+Fu7u7sLKyElZWVsLd3V18/fXXQqFQCCGEuHjxourfFSmrT6wQQrRs2VIsWLBAbZu7u7uYOXNmpcpUxJz7xBYXFwsLCwu1vpqPvywsLERxcbGpQ5WkqKgoAUA1KfSBAwdEfn6+OHDggGpSaAAa/a5JN8q+WX369NHa57hPnz5m3TfLnCn7cw8cOFBrv7eBAweyPzdRLRYbGyucnZ3VvnednZ3NeqEDIYzcJ1Ymk2HGjBk4deoUioqKcP/+fZw6dQozZsxQdSdo27at6t9V8dprr+Hbb7/FuXPnAADr1q1DRkYGXnnllUqVkTJra2uMHDmy3DIjR46sOXdW1UzZnzosLEzrY5ePPvpIrRxVjrLf5tGjR9GtWze1R97dunXD0aNH2W9TT8p+2snJyVr7vSUnJ6uVI6Lao1bMXFJNSbVWmzdvFl27dhWdOnVS3R107dpVrFq1SlXmwYMH4uWXXxZ16tQRTk5OwsHBQfz4449q9ehSpiLm3BJbWloqmjdvXm5LbPPmzdmSpSeO8DauyMhIAUAMHz5c6/UdPny4ACAiIyNNHarkJCQkCAAiPDxco7XFxcVFhIWFlfmUi4hqrtoyO4FefWIBICcnB7t27cKlS5dUg6uUFi5cqFMdyjldH/do/y0rKyt8//33iIiIwM2bN9G2bVuNJW51KSNliYmJqpaU5557Dh07dkRRURHs7OyQmZmJbdu2ITs7G4mJiRg6dKiJo5Ue5SjOoKAgjB49GiEhIfDz80NaWhpGjx6N+Ph4xMTESH4Up6nk5OQAAJydneHm5oYLFy6o9jk7O+PZZ59VK0e68/HxgbOzMw4cOICMjAyN/tyjR4+Gi4sL+8sT1TLKJ2DR0dEQQiAxMVFtNb+QkBB4e3sjKSnJLFfs0pk+GfL+/fuFg4ODqgW1e/fuws7OTgAQvXr10qdKkzPnltjQ0FABQDzxxBNa76i8vLwEABEaGmqiCGsGbX2HXFxczL7vkLlTtsQCEH5+fiIlJUUUFBSIlJQU4efnp9rHllj9xMbGltufm59fotpHOdYjMjJS61gE5feyuY71MGpL7Lx58/D+++9jzpw5kMlkSE1NRW5uLiZPnoxOnTpVLasmDZcuXQLwcB5e5ehjJQsLC4wbNw4HDx5UlSP91PhRnCbSokUL1b/Ff2cy+ffff1FUVKQ2t/Gj5Uh3yml0goOD4e3trdru4uLCWR+IainlGI6JEydqLKeenZ2NiRMnqpWTLH0y5Pr166uyYwsLC3H//n0hhBAXLlwQTk5O+lRpcmyJJamO4jR3yhH0rVq1EpaWlmrX18rKSrRq1Yoj6A2gtLRUrb+xufZ1IyLje3RWo7KegJnzrEZGnZ3g7t27sLe3BwA4Ojri4sWLAAB7e3vk5uZWJacmLZTzlKakpMDf319tlKG/vz8OHjyoVo4qr1aM4jQRZX/uK1euoHHjxnjhhRfw8ssv44UXXkCjRo1w5coVtXKkH0tLSwwZMgTjxo3DkCFD2IebqBZLSkqCQqFQvRdCqF5KCoVC8rPCyISo/FqlMplMdSEmTZqE27dvY/r06diwYQOuXr2qmtZFSvLz8+Hg4IC8vDxVgm4u5HI5WrZsiezsbNja2uL+/fuqfcpJ45s3b46rV6/yD5ce5HI5XF1d4enpqTZhPPDwf/KAgACkpaUhMzOT11cPe/bswbBhw9CqVSvcuHFDbSColZUVHB0dceXKFezevZsDE6tALpcjKSlJbfAGP69EtdN7772HJUuW4IMPPsCPP/6oNqDWxcUFkydPxuLFi7Fw4UJ8+OGHpgu0DLrmZHq1xK5cuVL172XLlkGhUGDKlCnIysrC2rVr9amSymFpaYmvv/4aAFBcXKy2T5nQfv311/yDpSflKM7Q0FCtfY5DQkI4j6kBXLlyBc888wxWrVqF77//HqtWrcIzzzyjaokl/XFJXyLSxsfHB6dPn8aKFSswa9YsrFixAunp6Rg4cKCpQzMIvQZ2zZo1S/VvJycnbN++3WABUdlkMhlsbW1RVFSk2vZ4yyxVnnJdeQ8PD637ldu5/rx+rl+/rvZe22MtbeVIN8quMH5+foiOjoaHhwfS0tIQFhbGJX2JaqkhQ4ZgyZIlmDlzJgoLC9UGfq9YsQJ169ZVlZMyvVpimzVrZug4qBxyuRzBwcHo3bu3xrVv1qwZevfujblz50Iul5soQmlTjs5MS0vTul+5XfKjOE1EOf/rM888g507d2LWrFmYNm0aZs2ahZ07d+Kpp55SK0e6U343+Pn5ITY2Fvfv38fWrVtx//59xMbGws/Pj98NRLXQkCFD4ODggPT0dGRlZanty8rKQnp6OhwcHCSfxOrVEiuEwO3bt9G4cWNDx0NaKB93X7hwAba2tmr7srOzVXdYkp+02ESUE8aHhYUhNjZWY8L48PBwThhfBcobr507d2LEiBF47rnnVH25t23bht9//12tHOlO+d3w2muvaV1I4tVXX8XWrVv53VCOwsJCpKen61S2qKgIFy5cgLOzs8a0RWVxd3dXtXrVRry+pqN82vX4U6+ytkuRXknsxIkTsXTpUoSHh2v0ISTDe7TPoEwmU9v36Hv2LdSPcsWu0aNHw97eXq17hrK7RmxsLPsc6+nx+V/L6k7AeWIrT9nFJTQ0VGt3ggULFqiVI03p6eno3bu30eo/duwYevXqZbT6zR2vr2kkJiYiPz8fwP8GgCsp3+fn50t+pU+9kti0tDTs2bMH0dHRcHd315gMPj4+3iDB0UOP9hUcOnQoFixYoPpD9dFHH6muN/sUVt3j/YvZ39hwWrVqhR07dqhaXoGHNxCtWrXiDZielEt0DxgwQPUUYevWrXByckJsbCyefPJJ7N+/X20pb1Ln7u6OY8eO6VT21KlTmDhxIiIjI9G5c2ed66/NeH1NY+/evQAALy8v7Nu3T+MJ46BBg3Dw4EHs3bu39iWxHh4eZQ6CIcO7efMmAKBRo0b49ddfYWX18Nfm5eWFX3/9Fc2bN0dubq6qHFWOXC7HpEmTAADW1tYICgpCnz59cPToUcTExKCkpASTJk2Cv78/W2P18Og8sY6Ojpg4cSLat2+Pc+fOITIykvPEGsDNmze1did4vPsRaapbt26lW/I6d+7M1j8d8fqahrIf7IQJE1CnTh2N7kTKlT4f7y8rNXolsZ999pmBw6DyXL58GQBw584dBAYGIiQkRNUSGx4ejjt37qiVo8rZtWsXCgsLUadOHRQUFKg9Wfjhhx9Qv359FBYWYteuXXj22WdNGKk0KVsB3d3dcf/+fURERKj2ubi4wN3dHenp6Wwt1IMy8U9PT9fo2nXp0iXVZOe8QSCqXdq0aQMA2LBhA/7zn/9ozH8eHR2tVk6qqtyh9dF+FmQcbdu2BQB07NgRJ0+ehLe3N+zt7eHt7Y20tDR07NhRrRxVzqeffgoAeOONNzS6xlhbW2PmzJlq5Ug/TZs2RUZGBhISEhAVFYWEhAScPn0aTZo0MXVokvVo4v/4Z9fGxkZrOSKq+ZQreB48eLBGr/SpV0vsgwcPsHjxYnz33Xe4fv26aoDGjBkzEBwcrEqqSHfljeBU3illZGRg4MCBeP7551FQUIAGDRrgyJEj2L9/v6rc8ePHtdZR20dwlnd9lS3Yjo6OOH78uMYIWWUCcPnyZV5fPShbAffv34/Ro0cjJCQEfn5+SEtLw+jRo1Ur/LG1sPKULa2NGjXCtWvXkJKSour39sQTT8DJyQm5ublqy08SUc03ZMgQNGvWDDk5OdizZ4/aWCXlzA/NmzeX/qwlQg/vv/++8PDwELGxseLRKjZu3CgmTJigT5Uml5eXJwCIvLw8k/z8Y8eOCQBGex07dswk52UueH1NJyEhQQAQ4eHhwtnZWe26ubi4iLCwMAFAJCQkmDpUyVm4cKHqWo4cOVIcOHBA5OfniwMHDoiRI0eq9i1cuNDUodYIyu8R/v9uHLy+hqXM0ezs7NS+d+vWrSsAiNjYWFOHWCZdczK9WmLXrVuH3377DZ6enmrbBw8ejBkzZuhTZa1X0QjOvXv34p133oG1tbXa0rM2NjYoKSnBsmXLyn0sUNtHcJZ3fYuKilRL8A0cOBBDhw7F4sWLsWjRIuzZs0fV0r1///4y5y6s7de3PMp5eA8cOICMjAyNUbKjR4/mPLxVpFwf3dvbW7XNxcUFixYtwuLFi00YGREZS0Vz8Do7O+OTTz7Bp59+qtb1s2HDhli8eDGcnZ3LfLoISOQJoz4ZsrW1tbh3754QQgiZTKbanp2dLWxtbfWp0uRM3RKri9jYWK0tWeZ8NyUV/v7+5ba0+vv7mzpESYuNjRUymUz4+fmJL7/8Unz33Xfiyy+/FH5+fkImk/EzrKfdu3cLAGLgwIGipKREJCQkiKioKJGQkCBKSkrEwIEDBQCxe/duU4daI7Cl0Lh4fSunJj9hNGpLbKdOnbB//348/fTTapPtr1u3Dt27d9enStJBYGAg/P398d133+G1117DmjVrMG3aNE77ZACbN29GQEAAtmzZorHP398fmzdvrv6gapDAwEDMnTsXK1asUOubZWVlhblz5yIwMNCE0UmXst/b/v37ERgYqFr0IC0tDYGBgao5YiXf742INHAOXj0Hdi1YsAATJ05EaGgoAGDTpk3YsWMH1q9fj9jYWIMGSOosLS3Rp08fAECfPn2YwBrQ5s2bUVRUhJdeegm//PILxowZg59++knn5Q+pbHFxcVi+fDlGjBiB4cOHq1aM2b59O5YvXw4vLy8msnqwtLTE6tWrMXr0aI3BG8rHgF9//TW/J4hqIM7Bq2cSO3bsWFhYWGDJkiVQKBR48cUX0aVLF0RHR8Pf39/QMRJVGzs7O8yfPx+//PIL5s+fzwTWAORyOYKDg+Hn54fNmzerzVc4Y8YMBAQEYO7cuVxMQk+BgYGYN2+exhRwxcXFmDdvHm8OiKjG0iuJBYAxY8ZgzJgxKCkpgRBCbU5CIiKlpKQkXLhwAdHR0RoT8ltYWCAkJATe3t5ISkriY289sJWbiGorvZJYV1dXTJgwARMnTuScsERUrmvXrgFAmUtVK7cry5Hu2MpNRLWZXit2TZ06FTExMXBzc0P//v2xcuVK5OTkGDo2IqoBnJycAABpaWla9yu3K8uR7pSt3KGhoWW2cp8/fx5JSUkmipCIyHj0SmJDQ0Pxzz//4MSJExg0aBCWLl2Kli1bYsSIEar1eImIgP/NExsWFqaxcpRCoUB4eDjnidUTW7mJqDbTu08sAPTo0QM9evTA0qVLkZiYiDlz5mD8+PEYN26coeIjIgmoaNLtmTNn4p133oGvry/Gjx8PS0tLyOVyREVFISkpCcuWLcNff/1V5vGSmHTbBB5t5fby8tLYz1ZuIqrJqpTEAsC///6LDRs2ICoqCllZWXjqqacMERcRSUh6ejp69+5dYbl9+/Zh3759GtvnzZtX7nHHjh2rUdPCGMqjrdyP94llKzcR1XR6JbFXrlxBdHQ0NmzYgNTUVPTq1Qtvvvkmxo0bhxYtWhg6RiIyc7pOui2Xy7F582aEhYUhNDQUAQEBOg04ksKk26ZgaWmJiIgIBAUFISAgACEhIfDw8EBaWhrCw8MRHx+PmJgYDuoiohpJryS2bdu2aNeuHcaPH4/o6Gj+gSGq5Soz6balpSXCwsIwevRotq4aQGBgIGJiYhAcHAxvb2/VdhcXF8TExHB6LSKqsfRKYpOSktS+LJWEEEhMTISvr2+VAyMioor7GwOAs7Mzfv75Zxw8eBDp6elwd3eHl5cXLC0tcfz48XKPZX9jIpIqvZLYxxPYq1ev4scff8R3332Hc+fOQQhhkOCIiGo7Xfsb64v9jYlIqvQe2FVaWorff/8d3377LbZv3w4vLy/MmDEDzz//vCHjIyKq1XTtbwwAp06dwsSJExEZGYnOnTvrXD8RkRRVOonNyMjA999/j59++gmlpaV46aWXEB8fj/379xsjPiKiWq0y/Y2VOnfuzNZVIqrxKrXYweDBg+Hu7o7Dhw9jxYoVuHLlCpYvX26s2IiIiIiItKpUS+y+ffvQr18/zJ8/H0899RRkMpmx4iIiIiIiKlOlWmIPHTqE7t27IygoCO3bt8dHH32Eq1evGis2AEBRURHu3Lmj9iooKCiz/P37940aDxERERGZXqWS2H79+mHt2rW4du0a3n//fWzbtg3t2rUDAMTGxuLevXsGD/Dtt99G8+bN4ezsrHo988wzGuUWLVqEhg0bon79+ujYsSN27Nhh8FiIiIiIyDxUKolVqlevHl5++WUkJyfj5MmTCA4Oxn/+8x80bdoUI0eONHSMGDVqlFpL7IEDB9T2r1y5Ep999hni4+NRWFiIl19+GQEBAThz5ozBYyEiIiIi09MriX2Uu7s7li9fjsuXLyMyMhKlpaWGiEtDcXFxmfPPfvbZZ3jllVcwcOBAWFtbIzQ0FC1atMDq1auNEgsRERERmVaVk1ilOnXqYPTo0di+fbuhqlTZunUr7O3tUb9+fTz99NP4999/Vftu3ryJc+fOwcfHR+2YQYMG4dChQwaPhYiIiIhMT+/FDqqLp6cndu/eDW9vb2RnZ2PWrFl48sknkZaWhqZNmyI7OxsA0LRpU7XjmjdvXm4SW1xcjOLiYtX7/Px845wAERFVu8zMzHIHAevj1KlTav81pAYNGqBjx44Gr5eoJjP7JHbmzJmqfzs5OWHdunVo1qwZfv75Z/znP/9R7VMoFGrHlZaWljsFWHh4OBYvXmz4gImIyKQyMzPh5uZmtPonTpxolHozMjKYyBJVgtknsY+rV68eWrdujfPnzwMAWrZsCQC4ceOGWrns7GzVPm1CQkIwZ84c1fv8/Hy0adPGCBETEVF1UrbAVmb5XV0UFRXhwoULcHZ2hp2dncHqVS4XbOiWY2OSUks3W7lrLsklsTk5Obh48SLatm0LAGjYsCE8PDywZ88ejBkzBsDDVtm9e/fi1VdfLbMeGxsb2NjYVEvMRERU/Yyx/O6AAQMMWp8USbGlW0qt3FK6QQBMe5Ng1klscXExhg8fjvnz56Nr1664dOkS5s2bh2bNmql9yENDQzFlyhT4+PjgiSeewCeffILi4mK8/vrrJoyeiIio5pFSS7fUWrmleIMAmO4mwayTWBsbG3z44YdYunQpTpw4gUaNGsHHxwexsbFo1KiRqty4ceNQVFSEpUuX4saNG/D09MTevXvh5ORkwuiJiIhqLrZ0G56UbhAA098kmHUSCzz8QP/2228Vlps6dSqmTp1aDRERERERGQ9vEHRjsHliiYiIiIiqi9m3xBKVhx3giYiIaicmsdXA0IkWk6yH2AGeiIio9mISa2TGTLRqe5LFDvDGx5ZuIiIyV0xijcwYiRaTLHXsAG8cbOk2Lt4gEBFVDZPYamLoRItJFhkbW7qNhzcIRERVxySWiMrFlm7D4w0CEVHVMYklIjIR3iAQEemPSSwREdU4LerLYHcnA7hq/tOh293JQIv6MlOHQWaCn13dMYklIqIa57Xe1ui87zVgn6kjqVhnPIyXCOBntzKYxBIRUY2z5lgJxr7/Izq7u5s6lAqdSk/HmojxGGXqQMgs8LOrOyaxRERU41y/K1DU0A1o2cPUoVSo6LoC1+8KU4dBZoKfXd0xiSUiIqJKkUq/TVP32STjYhJLRGQCUkkCACYCpEkq/TZN3WeTjItJLBGViYmW8UglCQCYCJAmqfTbNHWfTTIuJrFEVCYmWsYjlSQAYCJAmqTSb9PUfTYrq7CwEABw/Phxg9ZrzIVQTIlJbDWQSmuW1FqyAOlcW0Ca15eJlvFIJQkApJcIEElVeno6AGD69OkmjqRyGjRoYJKfyyS2GkilNUtqLVmAdK4tIM3ry0SLpIitWSRVAQEBAAB3d3fUrVvXYPUql4429FLXwMMEtmPHjgatU1dMYquBVFqzpNaSBUjn2gLSvL5EUsTWLOOS0k2C1G4QmjZtildeecVo9RtjqWtTYhJbDaTSmiXFliypXFtAmteXjENKSQAgvUSArVnGJcWbBKncIFDlMIklIqpmUkwCAOkkAmzNMi6p3SRI6QaBKodJLBFpxdZC45FaEgAwEaD/4U0CmQsmsUSkFVsLjYdJABFR1TGJJSKt2FpIRETmjEksSRYfdxsXWwuJiMicMYklyeLjbiIiotqLSayRGaO1kC2FD/FxNxERUe3FJNbIpNhaKJWWQj7uJiIyb4WFhaq/gxVRNqRUpkHF0I0YJC1MYo3MGK2FbCkkIiIpSE9PR+/evSt1zMSJE3Uue+zYMTY21GJMYo3MmK2FbCkkIiJz5u7ujmPHjulUVp+ucu4SWHLcWNjKzSSWiAyAX6bGw2trXLy+xlW3bt1KNbYMGDDAiNHULGzlZhJLRAbAL1Pj4bU1Ll5fkiq2cjOJJSID4Jep8fDaGhevL0kVW7mZxBKRAfDL1Hh4bY2L15dIupjEUq3Afm9EREQ1S41KYk+ePIkbN26gS5cuaNmypanDITPCfm9EREQ1S41IYu/evQt/f3/89ddf6NixI1JTU7FgwQIsXLjQ1KGRmWC/NyIiopqlRiSx7733Hs6fP4/Tp0+jSZMm2LVrF55++mkMGjQIgwYNMnV4OuHjbuNivzciIqKaRSaEEKYOoiqEEGjWrBnmzJmD0NBQ1fbevXujZ8+e+Pbbb3WqJz8/Hw4ODsjLy4O9vb2xwi3T8ePHK/24uzL4uJuIiIikQNecTPItsVeuXMGtW7fQvXt3te09evTAX3/9VeZxxcXFKC4uVr3Pz883Woy64ONuIiIiIt1JPom9c+cOAKBx48Zq25s0aaLap014eDgWL15sxMgqh4+7iYiIiHRnYeoAqsra2hrAwz6ljyosLFTt0yYkJAR5eXmqV1ZWllHjJCIiIiLDkXxLbNu2bWFpaYnLly+rbc/KyoKLi0uZx9nY2MDGxsbY4RERERGREUi+JdbW1haDBw9GXFycatudO3ewZ88ePPvssyaMjIiIiIiMRfKzEwDA4cOHMWjQILz88st44oknsHr1auTn5+PIkSM6D3wy9ewERERERKR7Tib5llgA6NevHw4fPgyZTIYtW7bg6aefxv79+3VOYImIiIhIWmpES6whsCWWiIiIyPRqVUssEREREdUuTGKJiIiISHKYxBIRERGR5Eh+nlhDUXYNNvXys0RERES1mTIXq2jYFpPY/yooKAAAtGnTxsSREBEREVFBQQEcHBzK3M/ZCf5LoVDg6tWraNCgAWQymanDKVd+fj7atGmDrKwszqRgBLy+xsXrazy8tsbF62tcvL7GI7VrK4RAQUEBWrZsCQuLsnu+siX2vywsLNC6dWtTh1Ep9vb2kvgwShWvr3Hx+hoPr61x8foaF6+v8Ujp2pbXAqvEgV1EREREJDlMYomIiIhIcpjESpCNjQ0WLVoEGxsbU4dSI/H6Ghevr/Hw2hoXr69x8foaT029thzYRURERESSw5ZYIiIiIpIcJrFEREREJDlMYomIiIhIcjhPrATduHEDmZmZ8PDwQMOGDU0dTo1y9+5dZGZmokWLFnBycjJ1ODXOrVu3cPHiRbRu3RrNmzc3dTg10s2bN5Geno527dpxBUIDUCgUOHDggMZ2Nzc3foYNLDMzEw8ePEDnzp3NftEhc1JUVIRjx45p3efq6ooWLVpUc0TVhwO7JCQ1NRXh4eFITExEdnY2tm/fjmeffdbUYdUIWVlZePfdd7Ft2za4uLjg/Pnz6NWrFyIjI9GyZUtThyd5GRkZmD17No4fP45WrVohPT0dQ4YMwYYNG3gjZkClpaUYPHgwDh48iMWLF2PhwoWmDkny7t69iwYNGqBbt25o0KCBavuCBQswfPhwE0ZWcxw+fBgvvfQS8vLy4OTkBLlcjo0bN8Ld3d3UoUnCpUuXMH78eLVtt2/fxqlTpxAbG4vAwECtx+3fvx99+/aV9IwF7E4gIf/88w+ef/55nDhxwtSh1Djnz5/HqFGjcPv2bZw4cQIXL17E3bt38corr5g6tBrh/PnzCAkJwfXr13Hs2DFcuHAB//zzDz744ANTh1ajvP/++3BxceFTBCNYs2YN9u/fr3oxgTWMS5cu4amnnsKIESNw+fJlHDt2DBs3bsTVq1dNHZpktG3bVu2zqfx8NmnSBH5+flqPKS0thY+PD65du1bN0RoWuxNIyIQJEwA8fFxIhjVo0CC19w4ODpgwYQLee+89E0VUszzzzDNq75s1a4auXbviypUrJoqo5tmzZw82btyIEydOoGvXrqYOp8ZRJlgdOnTg0wMDWr58ORo2bIiPP/4YFhYP29Xc3d3ZClsFDx48wPr16/HSSy/B2tra1OEYFZNYojIcOXIErq6upg6jRklOTkZxcTGSk5Nx5MgRbN261dQh1Qg5OTl46aWXEB0drdN641R5r7/+Olq0aIGMjAwEBARgzZo1TGYNYM+ePRg+fDgUCgWOHz+Oxo0bo127duwTWwW//fYbcnJyasWTRCaxRFr89ttviIqKQmxsrKlDqVEWLFiA/Px8nD59Gi+//DI8PT1NHZLkCSEwefJkTJkyBT4+PqYOp8axsrLCunXrMGnSJADAuXPnMGzYMMycORMbNmwwcXTSd/XqVeTm5qJz585o0KABrly5glatWiE6OhqdO3c2dXiS9N1332HgwIEa1+/EiRO4d+8eAEAulwMAjh49isuXLwN4+Fn38vKq3mCriAO7JOjmzZto1qwZB3YZyb59+zB8+HCEhIRwYIyR3LhxA08++SR69eqF9evXmzocSfv++++xcOFCbNiwAXXq1AEABAYGIiAgAFOnTpXcHyUpWL16Nd566y3cu3cPVlZsC6qK5s2b4969ezh06BA8PDxw//59+Pv74/bt2zhy5Iipw5Ocy5cvo127dvjhhx8wefJktX2TJ0/GuXPnVO+Tk5PRp08f1cCuBg0aYPv27dUab1Xx/z6iR+zfvx8jRozAvHnzmMAakaOjIyZPnoxPPvnE1KHUCO3bt1frv52Xl4edO3ciKytLcn+UpMDR0RElJSW4efNmjZ6+qDo4OzujUaNG8PDwAADY2tpiypQpmDBhAgoLC1G3bl0TRygtP/74Ixo0aIAxY8Zo7Fu3bp3q36WlpahTpw5++eUXODs7V2OEhsUklui/kpOTMXz4cMyZM4ej5g3s3r17qFevntq2M2fOoEmTJiaKqOaYOnUqpk6dqratdevWmD59Om/EDEDbZ/ePP/5As2bNOE+sATzzzDPYvHmz2rbLly+jfv36sLOzM01QEiWEwA8//ICJEyfWmmvHJFZClJOY5+XlAXg45Vb9+vXRpk0btGvXzsTRSVtqaiqGDx+OQYMG4amnnsL+/ftV+7y9vVWjZkk/06ZNQ7t27eDt7Q0rKyvs2rULP/zwg1rLAJE5+u6773DgwAGMGjUKDRs2xLZt2/Dtt9/im2++4feCAcyePRs//fQTpk2bhhdeeAFnz57FRx99hHfeeYeDuypp7969OHfuHKZPn27qUKoN+8RKyJ49e7Bo0SKN7RMnTsSMGTNMEFHNERcXh08//VTrvl27dtWau1pjKSkpwTfffIO9e/fi/v37cHV1xfTp01WPEMmwlH1iH+8TR/rZunUrfvnlF+Tk5KB9+/Z49dVX0b17d1OHVWNcvXoVy5YtQ1paGpo3b46goKAyJ+insi1btgx///03IiMjKywrl8sxePBgxMTESLpLDJNYIiIiIpIcPgshIiIiIslhEktEREREksMkloiIiIgkh0ksEREREUkOk1giIiIikhwmsUREREQkOUxiiYiIiEhymMQSEZHOfvvtN5w/f97UYRARcbEDIqqZtm3bhvz8fACAnZ0dXFxc0K1bNxNHVb7U1FTk5eVh8ODBpg6lTK1bt8aSJUswZcoUU4dCRLWclakDICIyhjlz5sDCwgLdunVDUVER9u/fDzc3N+zcuRP29vamDk+ryMhIpKWlmXUSS0RkLpjEElGNNWrUKHz88ccAgOvXr6Nz585YunQpPvroI7Vy//77LzIyMtCqVSv07NkTVlb/+2o8fvw4CgsL0bNnT6SmpiI3NxeDBg3SmgiXV09GRgaOHz8OALC3t4eHhwfatm2r2n/q1CmcPn0a169fx8aNGwEAPj4+aNWqldrP0CWelJQUWFlZoW/fvqptaWlpuHHjBoYOHapRT0pKCgoLCzFq1CgAQGlpKQ4fPoxbt26hT58+cHJy0jjX7OxspKamom7duvDy8lI717179yI7OxsymQwtWrRAjx494ODgoFFHamoqLl26hA4dOqBr165q+x48eICDBw/izp07cHd3R8eOHSt1PBHVfExiiahWaNGiBXr27Im//vpLta2oqAjjx4/HoUOH0KdPH5w9exZWVlbYunWrKsH8/vvvkZiYiOLiYrRr1w43btzA9evX8ccff6Bnz54613P27Fls3rwZAHDnzh3s378fwcHBWLx4MQAgMzMTmZmZyMvLU5VzdXXVSGJ1iefzzz9H/fr11ZLYmJgY7N69W5XEKuspLS1F27Zt0b59e4waNQonT57E888/D7lcji5duuDff//FwoULMW3aNFVdUVFR+PDDD9GlSxecOHECrVu3xr59+2BtbQ0A2L9/P/79918IIXDhwgWcPXsWP//8M5588kkADxPUkSNHIi0tDX369MGFCxfQtm1bxMXFwcrKCv/88w9GjRqFBg0aoG3btjh48CBGjBiB77//HjKZrMLjiaiWEERENVCnTp3Eu+++q3pfWloqXFxcxMsvv6zaNmfOHDF48GBRVFQkhBBCoVCISZMmCX9/f1WZmTNnCgBi69atqjLjxo0T/fv3r1Q9j/v333+Fra2tOHnypGpbcHCweOaZZ8o9L13iGTt2rJg2bZracYsWLRIDBgxQq0cmk4nk5GTVtgcPHoj27duLcePGiZKSEiGEEMXFxWL79u2qMq1atRI9evQQd+/eFUIIcevWLWFvby+ioqLKjDkiIkK4urqq3u/evVvUr19f5OXlqbb9/vvv4v79+6K0tFS4ubmJ8PBw1b5bt26Jtm3bih9++KHC44mo9uAtKxHVWKdOncLGjRtRVFSEX3/9Ffn5+Zg7dy4AQAiBH3/8EePHj0d8fDyEEBBCwMnJCVu2bFGrx93dHX5+fgAAmUyGefPmoVevXrh48SLatm2rcz13797FiRMncP36dcjlcjRt2hRHjhyBh4dHpc6rvHjatWuncz19+vSBt7e36n1SUhLOnTuHvXv3ok6dOgAAa2trPPvss2rHTZo0CfXq1QMANG7cGN26dcPp06fVyly+fBn//vsvcnNzYWlpiTNnziA3NxeNGjWCnZ0dSkpKcOrUKfTv3x8A8NxzzwEA9u3bh4yMDLRs2RIxMTGq69mhQwckJCRgypQp5R5PRLUHk1giqrEyMzOxefNm5OfnIyEhAe+88w66dOkCACgoKMDt27eRlpaGnJwcteOGDx+OBw8eqBI5Z2dntf0uLi4AgIsXL6JRo0Y61bNt2zZMnDgRbdu2hbOzM2xtbXHv3j1kZ2dX+rzKi6cySezjfV0vXboEGxubCuto3Lix2nsbGxvcv39f9X7u3Ln4+uuv0bdvXzRr1gwlJSUAHvajbdSoEby9vTF//nwMHz4cTZo0wdChQzF9+nT07t0bFy5cgJWVFbZt26b2M5o3b47OnTsDQLnHE1HtwSSWiGqsRwd2HTlyBAMGDIC7uzvGjRsHOzs7WFlZYdKkSZg6dWq59eTm5mp937RpU53rmT17NubNm4eQkBDVNmdnZwg9ZjksLx4AsLCwgEKhUCvzaJKpJJPJ1N43bNgQxcXFKCgoQIMGDSodF/BwcFtERAT++ecf1Q1DamoqfvvtN7VzXbx4Md577z0cP34cGzduRP/+/XH48GHY29ujtLQUq1atQpMmTcr8OWUd36tXL73iJiLp4WIHRFQr9O3bF2+99RaCg4NRWFiIOnXqYOjQoVi7dq1GwnflyhW198ePH0dWVpbqfVxcHBwdHeHq6qpzPdevX0enTp1U748cOYJLly6pla9fv77WZPNx5cUDAK1atcKZM2fUjklMTKyw3oEDB8LOzg7r169X2/54C3N5rl+/jjp16qBDhw6qbTExMWplbty4AblcDisrK/Tr1w+ffvopHB0dcfToUfj4+KBu3bpYs2aN2jEKhQLXr1+v8Hgiqj3YEktEtUZoaCi+/fZbrFixAgsWLMAXX3yBwYMHw8fHB+PGjYNcLkdCQgIaN26M77//XnVc/fr18fTTT2PmzJm4fv06PvnkE6xatUo1Gl+XegICAjBv3jxkZ2fjzp07+Oyzz1T9SpX69OmDTz75BCtXrkSzZs20TrGlSzwTJkzAihUr8Oabb6Jbt2747bffcO7cObi5uZV7fZo0aYIVK1bgjTfewKlTp9C9e3ccOnQItra2WLlypU7XWNmFICgoCM8//zyOHj2KTZs2qZU5cuQIQkJCEBQUBGdnZyQlJaGwsBBPPfUUmjRpgi+//BKvvfYazp49C29vb1y5cgVxcXF47733MHr06HKPJ6Lagy2xRFQjjRgxAt27d1fb1qhRI3zxxRe4dOkSFAoF3Nzc8M8//2D06NE4duwYLl68iGnTpqklsAAwaNAgrFmzBhcuXMD169cRFxeHV155RbVfl3q++eYbvPnmmzh48CBu3LiBLVu24M0331Q9cgcAPz8/rFmzBidPnsTmzZtx7do1redWUTw9evRAcnIyAODvv//GtGnT8NVXX2HYsGGqMr1794aPj49G3a+99hr279+POnXq4ODBg/D29sbnn3+u2u/v76/qg6vk6+urutYNGjRASkoKunbtisTERLRo0QJJSUkYO3asai5bPz8/xMTEQKFQ4M8//0Tbtm2Rmpqqqvfll1/GsWPH4OjoiH379kGhUCAqKgqjR4/W6Xgiqh247CwRUTlmzZqFy5cvq+ZuNTVzi4eIyFTYEktEREREksM+sURE5ejdu7dZPaY2t3iIiEyF3QmIiIiISHLYnYCIiIiIJIdJLBERERFJDpNYIiIiIpIcJrFEREREJDlMYomIiIhIcpjEEhEREZHkMIklIiIiIslhEktEREREksMkloiIiIgk5/8BfoeClawX9tsAAAAASUVORK5CYII=",
            "text/plain": [
              "<Figure size 700x400 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "check = clv.independence_check()\n",
        "print(repr(check), \"-> holds:\", check.holds())\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(7, 4))\n",
        "check.plot(ax=ax)\n",
        "fig.tight_layout()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-38",
      "metadata": {},
      "source": [
        "Spearman rho comes out 0.21 across 1,139 repeat buyers, and the spread inside each\n",
        "box dwarfs the drift between boxes. The assumption holds on CDNOW. It won't hold\n",
        "everywhere, which is why the check runs by default rather than living in a doc."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-39",
      "metadata": {},
      "source": [
        "### Plot and export\n",
        "\n",
        "Every clvkit result draws itself and hands back a plain DataFrame. `to_pandas()`\n",
        "is the escape hatch, so nothing here is a trap."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 21,
      "id": "cell-40",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x600 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "fig, ax = plt.subplots(figsize=(8, 6))\n",
        "result.plot(\n",
        "    ax=ax,\n",
        "    scatter_kwargs={\"s\": 14, \"alpha\": 0.55, \"edgecolors\": \"none\"},\n",
        "    title=\"CLV\",\n",
        ")\n",
        "# ax.set_xlim(0, 20)\n",
        "# ax.set_ylim(0, 60)\n",
        "fig.tight_layout()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 18,
      "id": "cell-41",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "wrote examples/output/cdnow_clv.csv (192,544 bytes)\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>expected_purchases</th>\n",
              "      <th>discounted_expected_transactions</th>\n",
              "      <th>expected_spend</th>\n",
              "      <th>clv</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>customer_id</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1981</th>\n",
              "      <td>6.20</td>\n",
              "      <td>6.16</td>\n",
              "      <td>41.44</td>\n",
              "      <td>255.37</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1203</th>\n",
              "      <td>5.89</td>\n",
              "      <td>5.86</td>\n",
              "      <td>37.71</td>\n",
              "      <td>220.77</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1516</th>\n",
              "      <td>5.87</td>\n",
              "      <td>5.83</td>\n",
              "      <td>35.81</td>\n",
              "      <td>208.88</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1081</th>\n",
              "      <td>2.00</td>\n",
              "      <td>1.99</td>\n",
              "      <td>101.32</td>\n",
              "      <td>201.56</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2149</th>\n",
              "      <td>3.88</td>\n",
              "      <td>3.85</td>\n",
              "      <td>47.43</td>\n",
              "      <td>182.70</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "             expected_purchases  discounted_expected_transactions  \\\n",
              "customer_id                                                         \n",
              "1981                       6.20                              6.16   \n",
              "1203                       5.89                              5.86   \n",
              "1516                       5.87                              5.83   \n",
              "1081                       2.00                              1.99   \n",
              "2149                       3.88                              3.85   \n",
              "\n",
              "             expected_spend     clv  \n",
              "customer_id                          \n",
              "1981                  41.44  255.37  \n",
              "1203                  37.71  220.77  \n",
              "1516                  35.81  208.88  \n",
              "1081                 101.32  201.56  \n",
              "2149                  47.43  182.70  "
            ]
          },
          "execution_count": null,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "csv_path = OUTPUT / \"cdnow_clv.csv\"\n",
        "result.to_pandas().to_csv(csv_path)\n",
        "print(f\"wrote {csv_path.relative_to(REPO)} ({csv_path.stat().st_size:,} bytes)\")\n",
        "\n",
        "result.to_pandas().nlargest(5, \"clv\").round(2)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-42",
      "metadata": {},
      "source": [
        "## 8. What `time_unit=\"W\"` does on its own\n",
        "\n",
        "`time_unit=\"W\"` without `collapse` doesn't merely re-scale the ruler. It buckets\n",
        "events at weekly grain, merging a Monday and a Wednesday purchase into one. That's\n",
        "a coarser sufficient statistic than the published fit used, and it takes the most\n",
        "from the most frequent buyers, whose behaviour is the whole reason to fit a model.\n",
        "\n",
        "`CustomerBase` won't do it silently. Naming the grain is consent to it. Inheriting\n",
        "it is the trap."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 19,
      "id": "cell-43",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "UserWarning: time_unit='W' collapsed 558 of 6919 transactions into earlier purchases in the same period. This biases the fit downward, and it takes the most from your most frequent buyers. Pass collapse='D' to keep them and still report time in 'W', or pass collapse='W' to say you meant this.\n",
            "\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>published</th>\n",
              "      <th>collapse=\"D\"</th>\n",
              "      <th>collapsed weekly</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>r</th>\n",
              "      <td>0.243</td>\n",
              "      <td>0.243</td>\n",
              "      <td>0.291</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>alpha</th>\n",
              "      <td>4.414</td>\n",
              "      <td>4.414</td>\n",
              "      <td>6.852</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>a</th>\n",
              "      <td>0.793</td>\n",
              "      <td>0.793</td>\n",
              "      <td>0.665</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>b</th>\n",
              "      <td>2.426</td>\n",
              "      <td>2.426</td>\n",
              "      <td>2.320</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       published  collapse=\"D\"  collapsed weekly\n",
              "r          0.243         0.243             0.291\n",
              "alpha      4.414         4.414             6.852\n",
              "a          0.793         0.793             0.665\n",
              "b          2.426         2.426             2.320"
            ]
          },
          "execution_count": null,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "import warnings\n",
        "\n",
        "with warnings.catch_warnings(record=True) as caught:\n",
        "    warnings.simplefilter(\"always\")\n",
        "    coarse = CustomerBase.from_transactions(log, amount_col=None, time_unit=\"W\")\n",
        "    for warning in caught:\n",
        "        print(f\"{warning.category.__name__}: {warning.message}\\n\")\n",
        "\n",
        "coarse_cal, _ = coarse.split(calibration_period_end=\"1997-09-30\")\n",
        "coarse_params = BGNBD().fit(coarse_cal).params_\n",
        "\n",
        "pd.DataFrame(\n",
        "    {\n",
        "        \"published\": published,\n",
        "        'collapse=\"D\"': model.params_,\n",
        "        \"collapsed weekly\": coarse_params,\n",
        "    }\n",
        ").round(3)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-44",
      "metadata": {},
      "source": [
        "Alpha moves 55%, from 4.41 to 6.85. Nothing raised, and the optimiser converged\n",
        "happily onto the wrong sufficient statistic. A fit that fails loudly is cheap to\n",
        "debug; this one doesn't, so the warning has to carry the whole signal."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-45",
      "metadata": {},
      "source": [
        "## 9. Exercise\n",
        "\n",
        "`CLV` accepts any transaction model with `fit` and `predict(t)`. The coupling is\n",
        "structural, not an inheritance hierarchy. MBG/NBD is the never-returner variant of\n",
        "BG/NBD, and it lets a customer drop out immediately after a purchase, which BG/NBD\n",
        "forbids.\n",
        "\n",
        "Refit lifetime value with `MBGNBD()` as the transaction model and compare its\n",
        "`clv` column against the BG/NBD one. Which direction does the mean move, and would\n",
        "you expect that from a model that lets customers leave sooner? Run the cell below\n",
        "for one answer, then try `horizon=104` to see whether the gap widens."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 20,
      "id": "cell-46",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>clv_bgnbd</th>\n",
              "      <th>clv_mbgnbd</th>\n",
              "      <th>delta</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>2349.00</td>\n",
              "      <td>2349.00</td>\n",
              "      <td>2349.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>7.73</td>\n",
              "      <td>29.48</td>\n",
              "      <td>21.75</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>18.92</td>\n",
              "      <td>72.62</td>\n",
              "      <td>53.72</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "      <td>0.00</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>1.36</td>\n",
              "      <td>3.94</td>\n",
              "      <td>2.57</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>1.45</td>\n",
              "      <td>4.29</td>\n",
              "      <td>2.84</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>5.36</td>\n",
              "      <td>21.93</td>\n",
              "      <td>16.71</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>255.37</td>\n",
              "      <td>949.17</td>\n",
              "      <td>693.80</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       clv_bgnbd  clv_mbgnbd    delta\n",
              "count    2349.00     2349.00  2349.00\n",
              "mean        7.73       29.48    21.75\n",
              "std        18.92       72.62    53.72\n",
              "min         0.00        0.00     0.00\n",
              "25%         1.36        3.94     2.57\n",
              "50%         1.45        4.29     2.84\n",
              "75%         5.36       21.93    16.71\n",
              "max       255.37      949.17   693.80"
            ]
          },
          "execution_count": null,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "from clvkit import MBGNBD\n",
        "\n",
        "mbg_result = (\n",
        "    CLV(transaction_model=MBGNBD()).fit(cb).predict(horizon=52, discount_rate=0.001)\n",
        ")\n",
        "\n",
        "side_by_side = pd.DataFrame(\n",
        "    {\n",
        "        \"clv_bgnbd\": result.to_pandas()[\"clv\"],\n",
        "        \"clv_mbgnbd\": mbg_result.to_pandas()[\"clv\"],\n",
        "    }\n",
        ")\n",
        "side_by_side[\"delta\"] = side_by_side[\"clv_mbgnbd\"] - side_by_side[\"clv_bgnbd\"]\n",
        "side_by_side.describe().round(2)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "cell-47",
      "metadata": {},
      "source": [
        "### Where to go next\n",
        "\n",
        "`opinions.md` in the repo root separates canon, what the papers settle, from\n",
        "opinion, what clvkit chose, for every default you just accepted.\n",
        "`docs/references.md` has the DOIs. None of the papers carries a redistribution\n",
        "licence, so fetch them from there.\n",
        "`examples/online_retail_ii_cohort.ipynb` covers the descriptive half of the\n",
        "library, where there is no likelihood at all."
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "name": "python",
      "version": "3.13"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}