{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Tutorial\n", "\n", "This tutorial provides a deep dive into the **Online Retail Simulator**, walking through the three-phase data generation process, exploring the generated data, and demonstrating how to inject and measure treatment effects using the enrichment layer." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The simulator operates in three phases. First, the **products** phase generates a product catalog with key attributes such as category and price tier. Second, the **product_details** phase enriches each product with detailed content including brand names, titles, descriptions, and feature lists. Third, the **metrics** phase simulates sales transactions with realistic temporal patterns and conversion metrics. You can choose between **rule-based generation**, which is deterministic and interpretable, and **synthesizer-based generation**, which learns patterns from your own data. An **enrichment layer** then allows you to inject known treatment effects—such as a quantity boost or a gradual rollout—so you can validate whether your causal models recover the true effect size." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Simulate Product Data" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We start by importing the **Online Retail Simulator** alongside a few standard libraries." ] }, { "cell_type": "code", "execution_count": 1, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:49.597012Z", "iopub.status.busy": "2026-03-01T03:45:49.596855Z", "iopub.status.idle": "2026-03-01T03:45:50.305744Z", "shell.execute_reply": "2026-03-01T03:45:50.304871Z" } }, "outputs": [], "source": [ "# Standard Library\n", "import inspect\n", "\n", "# Third-party packages\n", "from IPython.display import Code\n", "import pandas as pd\n", "\n", "from online_retail_simulator import enrich, simulate, load_job_results\n", "\n", "# Local imports\n", "from support import (\n", " plot_revenue_by_category,\n", " plot_daily_metrics_trend,\n", " plot_conversion_funnel,\n", " plot_treatment_effect,\n", " display_product_details,\n", ")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's start by simulating 100 products. The configuration is defined using [YAML](https://yaml.org/) and organizes the three-phase simulation process.\n", "\n", "The **STORAGE** section specifies where to save output files.\n", "\n", "The **RULE** section contains two subsections—**PRODUCTS** and **METRICS**—each defining a `FUNCTION` to call and a `PARAMS` subsection for function-specific parameters like `num_products`, `date_start`, `date_end`, and `seed`.\n", "\n", "The **PRODUCT_DETAILS** section specifies the `FUNCTION` used to enrich products with additional content." ] }, { "cell_type": "code", "execution_count": 2, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:50.307796Z", "iopub.status.busy": "2026-03-01T03:45:50.307550Z", "iopub.status.idle": "2026-03-01T03:45:50.436182Z", "shell.execute_reply": "2026-03-01T03:45:50.435298Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "STORAGE:\r\n", " PATH: output\r\n", "\r\n", "RULE:\r\n", " PRODUCTS:\r\n", " FUNCTION: simulate_products_rule_based\r\n", " PARAMS:\r\n", " num_products: 100\r\n", " seed: 42\r\n", "\r\n", " METRICS:\r\n", " FUNCTION: simulate_metrics_rule_based\r\n", " PARAMS:\r\n", " date_start: \"2024-11-01\"\r\n", " date_end: \"2024-11-30\"\r\n", " sale_prob: 0.7\r\n", " seed: 42\r\n", "\r\n", "PRODUCT_DETAILS:\r\n", " FUNCTION: simulate_product_details_mock\r\n" ] } ], "source": [ "! cat \"config_tutorial.yaml\"" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "To understand how the simulation functions work under the hood, you can explore their source code directly. Python's `inspect` module lets you view the implementation right inside the notebook, or you can browse the source files on GitHub ([products](https://github.com/eisenhauerIO/tools-online-retail-simulator/blob/main/online_retail_simulator/simulate/products_rule_based.py), [product_details](https://github.com/eisenhauerIO/tools-online-retail-simulator/blob/main/online_retail_simulator/simulate/product_details_mock.py), [metrics](https://github.com/eisenhauerIO/tools-online-retail-simulator/blob/main/online_retail_simulator/simulate/metrics_rule_based.py)) or the API documentation ([products](https://eisenhauerio.github.io/tools-online-retail-simulator/_modules/online_retail_simulator/simulate/products_rule_based.html#simulate_products_rule_based), [product_details](https://eisenhauerio.github.io/tools-online-retail-simulator/_modules/online_retail_simulator/simulate/product_details.html#simulate_product_details_mock), [metrics](https://eisenhauerio.github.io/tools-online-retail-simulator/_modules/online_retail_simulator/simulate/metrics_rule_based.html#simulate_metrics_rule_based))." ] }, { "cell_type": "code", "execution_count": 3, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:50.438334Z", "iopub.status.busy": "2026-03-01T03:45:50.438144Z", "iopub.status.idle": "2026-03-01T03:45:50.476898Z", "shell.execute_reply": "2026-03-01T03:45:50.476152Z" } }, "outputs": [ { "data": { "text/html": [ "
def simulate_products_rule_based(config: Dict) -> pd.DataFrame:\n",
       "    """\n",
       "    Generate synthetic products (rule-based).\n",
       "    Args:\n",
       "        config: Complete configuration dictionary\n",
       "    Returns:\n",
       "        DataFrame of products\n",
       "    """\n",
       "    params = config["RULE"]["PRODUCTS"]["PARAMS"]\n",
       "    num_products, seed = params["num_products"], params["seed"]\n",
       "\n",
       "    rng = np.random.default_rng(seed)\n",
       "    products: List[Dict] = []\n",
       "    for i in range(num_products):\n",
       "        category = rng.choice(_CATEGORIES)\n",
       "        price_min, price_max = _PRICE_RANGES[category]\n",
       "        price = round(rng.uniform(price_min, price_max), 2)\n",
       "        products.append(\n",
       "            {\n",
       "                "product_identifier": generate_random_product_identifier(rng),\n",
       "                "category": category,\n",
       "                "price": price,\n",
       "            }\n",
       "        )\n",
       "    return pd.DataFrame(products)\n",
       "
\n" ], "text/latex": [ "\\begin{Verbatim}[commandchars=\\\\\\{\\}]\n", "\\PY{k}{def}\\PY{+w}{ }\\PY{n+nf}{simulate\\PYZus{}products\\PYZus{}rule\\PYZus{}based}\\PY{p}{(}\\PY{n}{config}\\PY{p}{:} \\PY{n}{Dict}\\PY{p}{)} \\PY{o}{\\PYZhy{}}\\PY{o}{\\PYZgt{}} \\PY{n}{pd}\\PY{o}{.}\\PY{n}{DataFrame}\\PY{p}{:}\n", "\\PY{+w}{ }\\PY{l+s+sd}{\\PYZdq{}\\PYZdq{}\\PYZdq{}}\n", "\\PY{l+s+sd}{ Generate synthetic products (rule\\PYZhy{}based).}\n", "\\PY{l+s+sd}{ Args:}\n", "\\PY{l+s+sd}{ config: Complete configuration dictionary}\n", "\\PY{l+s+sd}{ Returns:}\n", "\\PY{l+s+sd}{ DataFrame of products}\n", "\\PY{l+s+sd}{ \\PYZdq{}\\PYZdq{}\\PYZdq{}}\n", " \\PY{n}{params} \\PY{o}{=} \\PY{n}{config}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{RULE}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{PRODUCTS}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{PARAMS}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\n", " \\PY{n}{num\\PYZus{}products}\\PY{p}{,} \\PY{n}{seed} \\PY{o}{=} \\PY{n}{params}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{num\\PYZus{}products}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\\PY{p}{,} \\PY{n}{params}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{seed}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\n", "\n", " \\PY{n}{rng} \\PY{o}{=} \\PY{n}{np}\\PY{o}{.}\\PY{n}{random}\\PY{o}{.}\\PY{n}{default\\PYZus{}rng}\\PY{p}{(}\\PY{n}{seed}\\PY{p}{)}\n", " \\PY{n}{products}\\PY{p}{:} \\PY{n}{List}\\PY{p}{[}\\PY{n}{Dict}\\PY{p}{]} \\PY{o}{=} \\PY{p}{[}\\PY{p}{]}\n", " \\PY{k}{for} \\PY{n}{i} \\PY{o+ow}{in} \\PY{n+nb}{range}\\PY{p}{(}\\PY{n}{num\\PYZus{}products}\\PY{p}{)}\\PY{p}{:}\n", " \\PY{n}{category} \\PY{o}{=} \\PY{n}{rng}\\PY{o}{.}\\PY{n}{choice}\\PY{p}{(}\\PY{n}{\\PYZus{}CATEGORIES}\\PY{p}{)}\n", " \\PY{n}{price\\PYZus{}min}\\PY{p}{,} \\PY{n}{price\\PYZus{}max} \\PY{o}{=} \\PY{n}{\\PYZus{}PRICE\\PYZus{}RANGES}\\PY{p}{[}\\PY{n}{category}\\PY{p}{]}\n", " \\PY{n}{price} \\PY{o}{=} \\PY{n+nb}{round}\\PY{p}{(}\\PY{n}{rng}\\PY{o}{.}\\PY{n}{uniform}\\PY{p}{(}\\PY{n}{price\\PYZus{}min}\\PY{p}{,} \\PY{n}{price\\PYZus{}max}\\PY{p}{)}\\PY{p}{,} \\PY{l+m+mi}{2}\\PY{p}{)}\n", " \\PY{n}{products}\\PY{o}{.}\\PY{n}{append}\\PY{p}{(}\n", " \\PY{p}{\\PYZob{}}\n", " \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{product\\PYZus{}identifier}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{generate\\PYZus{}random\\PYZus{}product\\PYZus{}identifier}\\PY{p}{(}\\PY{n}{rng}\\PY{p}{)}\\PY{p}{,}\n", " \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{category}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{category}\\PY{p}{,}\n", " \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{price}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{price}\\PY{p}{,}\n", " \\PY{p}{\\PYZcb{}}\n", " \\PY{p}{)}\n", " \\PY{k}{return} \\PY{n}{pd}\\PY{o}{.}\\PY{n}{DataFrame}\\PY{p}{(}\\PY{n}{products}\\PY{p}{)}\n", "\\end{Verbatim}\n" ], "text/plain": [ "def simulate_products_rule_based(config: Dict) -> pd.DataFrame:\n", " \"\"\"\n", " Generate synthetic products (rule-based).\n", " Args:\n", " config: Complete configuration dictionary\n", " Returns:\n", " DataFrame of products\n", " \"\"\"\n", " params = config[\"RULE\"][\"PRODUCTS\"][\"PARAMS\"]\n", " num_products, seed = params[\"num_products\"], params[\"seed\"]\n", "\n", " rng = np.random.default_rng(seed)\n", " products: List[Dict] = []\n", " for i in range(num_products):\n", " category = rng.choice(_CATEGORIES)\n", " price_min, price_max = _PRICE_RANGES[category]\n", " price = round(rng.uniform(price_min, price_max), 2)\n", " products.append(\n", " {\n", " \"product_identifier\": generate_random_product_identifier(rng),\n", " \"category\": category,\n", " \"price\": price,\n", " }\n", " )\n", " return pd.DataFrame(products)" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from online_retail_simulator.simulate.products_rule_based import simulate_products_rule_based\n", "\n", "Code(inspect.getsource(simulate_products_rule_based), language=\"python\")" ] }, { "cell_type": "code", "execution_count": 4, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:50.478696Z", "iopub.status.busy": "2026-03-01T03:45:50.478535Z", "iopub.status.idle": "2026-03-01T03:45:50.490963Z", "shell.execute_reply": "2026-03-01T03:45:50.490086Z" } }, "outputs": [ { "data": { "text/html": [ "
def simulate_product_details_mock(\n",
       "    products_df: pd.DataFrame,\n",
       "    seed: int = None,\n",
       "    prompt_path: str = None,\n",
       "    treatment_mode: bool = False,\n",
       ") -> pd.DataFrame:\n",
       "    """Generate mock product details (rule-based).\n",
       "\n",
       "    Args:\n",
       "        products_df: Input products with product_identifier, category, price\n",
       "        seed: Random seed for reproducibility\n",
       "        prompt_path: Ignored for mock backend (accepted for API compatibility)\n",
       "        treatment_mode: If True, generate "improved" product details\n",
       "\n",
       "    Returns:\n",
       "        DataFrame with added title, description, brand, features\n",
       "    """\n",
       "    # prompt_path is ignored for mock backend but accepted for API compatibility\n",
       "    _ = prompt_path\n",
       "\n",
       "    rng = np.random.default_rng(seed)\n",
       "    results = []\n",
       "\n",
       "    for product in products_df.to_dict("records"):\n",
       "        category = product.get("category", "General")\n",
       "        data = _get_mock_data(category, treatment_mode=treatment_mode)\n",
       "\n",
       "        # Preserve existing brand if present, otherwise generate new one\n",
       "        brand = product.get("brand") or rng.choice(data["brands"])\n",
       "        adj = rng.choice(data["adjectives"])\n",
       "        num_features = min(4, len(data["features"]))\n",
       "        features = list(rng.choice(data["features"], size=num_features, replace=False))\n",
       "\n",
       "        title = f"{brand} {adj} {category} Item"\n",
       "        if treatment_mode:\n",
       "            description = f"Premium {category.lower()} product with exceptional quality. {features[0]}. {features[1]}."\n",
       "        else:\n",
       "            description = f"Quality {category.lower()} product for everyday use. {features[0]}. {features[1]}."\n",
       "\n",
       "        result = {\n",
       "            **product,\n",
       "            "title": title,\n",
       "            "description": description,\n",
       "            "brand": brand,\n",
       "            "features": features,\n",
       "        }\n",
       "        result["quality_score"] = calculate_quality_score(result)\n",
       "        results.append(result)\n",
       "\n",
       "    return pd.DataFrame(results)\n",
       "
\n" ], "text/latex": [ "\\begin{Verbatim}[commandchars=\\\\\\{\\}]\n", "\\PY{k}{def}\\PY{+w}{ }\\PY{n+nf}{simulate\\PYZus{}product\\PYZus{}details\\PYZus{}mock}\\PY{p}{(}\n", " \\PY{n}{products\\PYZus{}df}\\PY{p}{:} \\PY{n}{pd}\\PY{o}{.}\\PY{n}{DataFrame}\\PY{p}{,}\n", " \\PY{n}{seed}\\PY{p}{:} \\PY{n+nb}{int} \\PY{o}{=} \\PY{k+kc}{None}\\PY{p}{,}\n", " \\PY{n}{prompt\\PYZus{}path}\\PY{p}{:} \\PY{n+nb}{str} \\PY{o}{=} \\PY{k+kc}{None}\\PY{p}{,}\n", " \\PY{n}{treatment\\PYZus{}mode}\\PY{p}{:} \\PY{n+nb}{bool} \\PY{o}{=} \\PY{k+kc}{False}\\PY{p}{,}\n", "\\PY{p}{)} \\PY{o}{\\PYZhy{}}\\PY{o}{\\PYZgt{}} \\PY{n}{pd}\\PY{o}{.}\\PY{n}{DataFrame}\\PY{p}{:}\n", "\\PY{+w}{ }\\PY{l+s+sd}{\\PYZdq{}\\PYZdq{}\\PYZdq{}Generate mock product details (rule\\PYZhy{}based).}\n", "\n", "\\PY{l+s+sd}{ Args:}\n", "\\PY{l+s+sd}{ products\\PYZus{}df: Input products with product\\PYZus{}identifier, category, price}\n", "\\PY{l+s+sd}{ seed: Random seed for reproducibility}\n", "\\PY{l+s+sd}{ prompt\\PYZus{}path: Ignored for mock backend (accepted for API compatibility)}\n", "\\PY{l+s+sd}{ treatment\\PYZus{}mode: If True, generate \\PYZdq{}improved\\PYZdq{} product details}\n", "\n", "\\PY{l+s+sd}{ Returns:}\n", "\\PY{l+s+sd}{ DataFrame with added title, description, brand, features}\n", "\\PY{l+s+sd}{ \\PYZdq{}\\PYZdq{}\\PYZdq{}}\n", " \\PY{c+c1}{\\PYZsh{} prompt\\PYZus{}path is ignored for mock backend but accepted for API compatibility}\n", " \\PY{n}{\\PYZus{}} \\PY{o}{=} \\PY{n}{prompt\\PYZus{}path}\n", "\n", " \\PY{n}{rng} \\PY{o}{=} \\PY{n}{np}\\PY{o}{.}\\PY{n}{random}\\PY{o}{.}\\PY{n}{default\\PYZus{}rng}\\PY{p}{(}\\PY{n}{seed}\\PY{p}{)}\n", " \\PY{n}{results} \\PY{o}{=} \\PY{p}{[}\\PY{p}{]}\n", "\n", " \\PY{k}{for} \\PY{n}{product} \\PY{o+ow}{in} \\PY{n}{products\\PYZus{}df}\\PY{o}{.}\\PY{n}{to\\PYZus{}dict}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{records}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\\PY{p}{:}\n", " \\PY{n}{category} \\PY{o}{=} \\PY{n}{product}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{category}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{General}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\n", " \\PY{n}{data} \\PY{o}{=} \\PY{n}{\\PYZus{}get\\PYZus{}mock\\PYZus{}data}\\PY{p}{(}\\PY{n}{category}\\PY{p}{,} \\PY{n}{treatment\\PYZus{}mode}\\PY{o}{=}\\PY{n}{treatment\\PYZus{}mode}\\PY{p}{)}\n", "\n", " \\PY{c+c1}{\\PYZsh{} Preserve existing brand if present, otherwise generate new one}\n", " \\PY{n}{brand} \\PY{o}{=} \\PY{n}{product}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{brand}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)} \\PY{o+ow}{or} \\PY{n}{rng}\\PY{o}{.}\\PY{n}{choice}\\PY{p}{(}\\PY{n}{data}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{brands}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\\PY{p}{)}\n", " \\PY{n}{adj} \\PY{o}{=} \\PY{n}{rng}\\PY{o}{.}\\PY{n}{choice}\\PY{p}{(}\\PY{n}{data}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{adjectives}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\\PY{p}{)}\n", " \\PY{n}{num\\PYZus{}features} \\PY{o}{=} \\PY{n+nb}{min}\\PY{p}{(}\\PY{l+m+mi}{4}\\PY{p}{,} \\PY{n+nb}{len}\\PY{p}{(}\\PY{n}{data}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{features}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\\PY{p}{)}\\PY{p}{)}\n", " \\PY{n}{features} \\PY{o}{=} \\PY{n+nb}{list}\\PY{p}{(}\\PY{n}{rng}\\PY{o}{.}\\PY{n}{choice}\\PY{p}{(}\\PY{n}{data}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{features}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\\PY{p}{,} \\PY{n}{size}\\PY{o}{=}\\PY{n}{num\\PYZus{}features}\\PY{p}{,} \\PY{n}{replace}\\PY{o}{=}\\PY{k+kc}{False}\\PY{p}{)}\\PY{p}{)}\n", "\n", " \\PY{n}{title} \\PY{o}{=} \\PY{l+s+sa}{f}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+si}{\\PYZob{}}\\PY{n}{brand}\\PY{l+s+si}{\\PYZcb{}}\\PY{l+s+s2}{ }\\PY{l+s+si}{\\PYZob{}}\\PY{n}{adj}\\PY{l+s+si}{\\PYZcb{}}\\PY{l+s+s2}{ }\\PY{l+s+si}{\\PYZob{}}\\PY{n}{category}\\PY{l+s+si}{\\PYZcb{}}\\PY{l+s+s2}{ Item}\\PY{l+s+s2}{\\PYZdq{}}\n", " \\PY{k}{if} \\PY{n}{treatment\\PYZus{}mode}\\PY{p}{:}\n", " \\PY{n}{description} \\PY{o}{=} \\PY{l+s+sa}{f}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{Premium }\\PY{l+s+si}{\\PYZob{}}\\PY{n}{category}\\PY{o}{.}\\PY{n}{lower}\\PY{p}{(}\\PY{p}{)}\\PY{l+s+si}{\\PYZcb{}}\\PY{l+s+s2}{ product with exceptional quality. }\\PY{l+s+si}{\\PYZob{}}\\PY{n}{features}\\PY{p}{[}\\PY{l+m+mi}{0}\\PY{p}{]}\\PY{l+s+si}{\\PYZcb{}}\\PY{l+s+s2}{. }\\PY{l+s+si}{\\PYZob{}}\\PY{n}{features}\\PY{p}{[}\\PY{l+m+mi}{1}\\PY{p}{]}\\PY{l+s+si}{\\PYZcb{}}\\PY{l+s+s2}{.}\\PY{l+s+s2}{\\PYZdq{}}\n", " \\PY{k}{else}\\PY{p}{:}\n", " \\PY{n}{description} \\PY{o}{=} \\PY{l+s+sa}{f}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{Quality }\\PY{l+s+si}{\\PYZob{}}\\PY{n}{category}\\PY{o}{.}\\PY{n}{lower}\\PY{p}{(}\\PY{p}{)}\\PY{l+s+si}{\\PYZcb{}}\\PY{l+s+s2}{ product for everyday use. }\\PY{l+s+si}{\\PYZob{}}\\PY{n}{features}\\PY{p}{[}\\PY{l+m+mi}{0}\\PY{p}{]}\\PY{l+s+si}{\\PYZcb{}}\\PY{l+s+s2}{. }\\PY{l+s+si}{\\PYZob{}}\\PY{n}{features}\\PY{p}{[}\\PY{l+m+mi}{1}\\PY{p}{]}\\PY{l+s+si}{\\PYZcb{}}\\PY{l+s+s2}{.}\\PY{l+s+s2}{\\PYZdq{}}\n", "\n", " \\PY{n}{result} \\PY{o}{=} \\PY{p}{\\PYZob{}}\n", " \\PY{o}{*}\\PY{o}{*}\\PY{n}{product}\\PY{p}{,}\n", " \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{title}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{title}\\PY{p}{,}\n", " \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{description}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{description}\\PY{p}{,}\n", " \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{brand}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{brand}\\PY{p}{,}\n", " \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{features}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{features}\\PY{p}{,}\n", " \\PY{p}{\\PYZcb{}}\n", " \\PY{n}{result}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{quality\\PYZus{}score}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]} \\PY{o}{=} \\PY{n}{calculate\\PYZus{}quality\\PYZus{}score}\\PY{p}{(}\\PY{n}{result}\\PY{p}{)}\n", " \\PY{n}{results}\\PY{o}{.}\\PY{n}{append}\\PY{p}{(}\\PY{n}{result}\\PY{p}{)}\n", "\n", " \\PY{k}{return} \\PY{n}{pd}\\PY{o}{.}\\PY{n}{DataFrame}\\PY{p}{(}\\PY{n}{results}\\PY{p}{)}\n", "\\end{Verbatim}\n" ], "text/plain": [ "def simulate_product_details_mock(\n", " products_df: pd.DataFrame,\n", " seed: int = None,\n", " prompt_path: str = None,\n", " treatment_mode: bool = False,\n", ") -> pd.DataFrame:\n", " \"\"\"Generate mock product details (rule-based).\n", "\n", " Args:\n", " products_df: Input products with product_identifier, category, price\n", " seed: Random seed for reproducibility\n", " prompt_path: Ignored for mock backend (accepted for API compatibility)\n", " treatment_mode: If True, generate \"improved\" product details\n", "\n", " Returns:\n", " DataFrame with added title, description, brand, features\n", " \"\"\"\n", " # prompt_path is ignored for mock backend but accepted for API compatibility\n", " _ = prompt_path\n", "\n", " rng = np.random.default_rng(seed)\n", " results = []\n", "\n", " for product in products_df.to_dict(\"records\"):\n", " category = product.get(\"category\", \"General\")\n", " data = _get_mock_data(category, treatment_mode=treatment_mode)\n", "\n", " # Preserve existing brand if present, otherwise generate new one\n", " brand = product.get(\"brand\") or rng.choice(data[\"brands\"])\n", " adj = rng.choice(data[\"adjectives\"])\n", " num_features = min(4, len(data[\"features\"]))\n", " features = list(rng.choice(data[\"features\"], size=num_features, replace=False))\n", "\n", " title = f\"{brand} {adj} {category} Item\"\n", " if treatment_mode:\n", " description = f\"Premium {category.lower()} product with exceptional quality. {features[0]}. {features[1]}.\"\n", " else:\n", " description = f\"Quality {category.lower()} product for everyday use. {features[0]}. {features[1]}.\"\n", "\n", " result = {\n", " **product,\n", " \"title\": title,\n", " \"description\": description,\n", " \"brand\": brand,\n", " \"features\": features,\n", " }\n", " result[\"quality_score\"] = calculate_quality_score(result)\n", " results.append(result)\n", "\n", " return pd.DataFrame(results)" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from online_retail_simulator.simulate.product_details_mock import simulate_product_details_mock\n", "\n", "Code(inspect.getsource(simulate_product_details_mock), language=\"python\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Calling `simulate(\"config_tutorial.yaml\")` triggers a three-phase data generation process. First, it reads and validates the YAML configuration, then runs the **products** phase to create a product catalog with the specified number of products, assigning each a unique identifier, category, and price. Second, it runs the **product_details** phase to enrich each product with brand names, titles, descriptions, and feature lists. Third, it runs the **metrics** phase, simulating a shopper conversion funnel (impressions, visits, cart additions, orders, and revenue) for each product across the defined date range. The function writes the generated `DataFrame`s to disk and returns a `JobInfo` object that tracks where the results are stored—which you then pass to `load_job_results()` to retrieve the data.\n", "\n", "The products data represents the catalog of items available for sale. Each product starts with core **attributes**—a unique `product_identifier`, `category` (such as Electronics, Clothing, or Books), `brand`, and `price`—which are then enriched with **product_details** including `title`, `description`, and `features`. These attributes influence the simulated sales behavior—for example, higher-priced items may have different conversion patterns than low-cost products, and categories can exhibit distinct purchasing dynamics." ] }, { "cell_type": "code", "execution_count": 5, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:50.492530Z", "iopub.status.busy": "2026-03-01T03:45:50.492345Z", "iopub.status.idle": "2026-03-01T03:45:50.926533Z", "shell.execute_reply": "2026-03-01T03:45:50.925580Z" } }, "outputs": [], "source": [ "job_info = simulate(\"config_tutorial.yaml\")\n", "\n", "results = load_job_results(job_info)\n", "products = results[\"products\"]\n", "metrics = results[\"metrics\"]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Products" ] }, { "cell_type": "code", "execution_count": 6, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:50.928402Z", "iopub.status.busy": "2026-03-01T03:45:50.928077Z", "iopub.status.idle": "2026-03-01T03:45:50.937283Z", "shell.execute_reply": "2026-03-01T03:45:50.936446Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
product_identifiercategorypricetitledescriptionbrandfeaturesquality_score
0B1P4DZHDS9Electronics686.37NexGen Smart Electronics ItemQuality electronics product for everyday use. ...NexGen[np.str_('Long battery life'), np.str_('Voice ...0.800
1B1SE4QSNG7Toys & Games80.75PrimePick Premium Toys & Games ItemQuality toys & games product for everyday use....PrimePick[np.str_('High quality materials'), np.str_('D...0.882
2BXTPQIDT5CFood & Beverage42.02TrustMark Premium Food & Beverage ItemQuality food & beverage product for everyday u...TrustMark[np.str_('Easy to use'), np.str_('High quality...0.879
3B3F1ZMC8Q6Food & Beverage33.42BestChoice Classic Food & Beverage ItemQuality food & beverage product for everyday u...BestChoice[np.str_('High quality materials'), np.str_('G...0.885
4B2NQRBTF0YToys & Games27.52QualityFirst Classic Toys & Games ItemQuality toys & games product for everyday use....QualityFirst[np.str_('Great value'), np.str_('Easy to use'...0.830
\n", "
" ], "text/plain": [ " product_identifier category price \\\n", "0 B1P4DZHDS9 Electronics 686.37 \n", "1 B1SE4QSNG7 Toys & Games 80.75 \n", "2 BXTPQIDT5C Food & Beverage 42.02 \n", "3 B3F1ZMC8Q6 Food & Beverage 33.42 \n", "4 B2NQRBTF0Y Toys & Games 27.52 \n", "\n", " title \\\n", "0 NexGen Smart Electronics Item \n", "1 PrimePick Premium Toys & Games Item \n", "2 TrustMark Premium Food & Beverage Item \n", "3 BestChoice Classic Food & Beverage Item \n", "4 QualityFirst Classic Toys & Games Item \n", "\n", " description brand \\\n", "0 Quality electronics product for everyday use. ... NexGen \n", "1 Quality toys & games product for everyday use.... PrimePick \n", "2 Quality food & beverage product for everyday u... TrustMark \n", "3 Quality food & beverage product for everyday u... BestChoice \n", "4 Quality toys & games product for everyday use.... QualityFirst \n", "\n", " features quality_score \n", "0 [np.str_('Long battery life'), np.str_('Voice ... 0.800 \n", "1 [np.str_('High quality materials'), np.str_('D... 0.882 \n", "2 [np.str_('Easy to use'), np.str_('High quality... 0.879 \n", "3 [np.str_('High quality materials'), np.str_('G... 0.885 \n", "4 [np.str_('Great value'), np.str_('Easy to use'... 0.830 " ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "products.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's examine a single product to understand the structure better." ] }, { "cell_type": "code", "execution_count": 7, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:50.938975Z", "iopub.status.busy": "2026-03-01T03:45:50.938700Z", "iopub.status.idle": "2026-03-01T03:45:50.942969Z", "shell.execute_reply": "2026-03-01T03:45:50.942256Z" } }, "outputs": [ { "data": { "text/plain": [ "product_identifier B1P4DZHDS9\n", "category Electronics\n", "price 686.37\n", "title NexGen Smart Electronics Item\n", "description Quality electronics product for everyday use. ...\n", "brand NexGen\n", "features [np.str_('Long battery life'), np.str_('Voice ...\n", "quality_score 0.8\n", "Name: 0, dtype: object" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "example_product = products.iloc[0]\n", "example_product" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This product has a unique `product_identifier` for tracking, belongs to the `category` Clothing with a `price` of $19.63, and includes enriched content: a `brand` name (StyleFit), a descriptive `title`, a marketing-focused `description`, and a list of product `features`. These attributes combine catalog characteristics with content details to create a realistic product listing." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Metrics" ] }, { "cell_type": "code", "execution_count": 8, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:50.944665Z", "iopub.status.busy": "2026-03-01T03:45:50.944503Z", "iopub.status.idle": "2026-03-01T03:45:50.955223Z", "shell.execute_reply": "2026-03-01T03:45:50.954407Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
product_identifiercategorypricetitledescriptionbrandfeaturesquality_scoredateimpressionsvisitscart_addsordered_unitsrevenue
73B0D4QF30BGSports & Outdoors194.55QualityFirst Quality Sports & Outdoors ItemQuality sports & outdoors product for everyday...QualityFirst[np.str_('Long lasting'), np.str_('Easy to use...0.8812024-11-011001421194.55
173B0D4QF30BGSports & Outdoors194.55QualityFirst Quality Sports & Outdoors ItemQuality sports & outdoors product for everyday...QualityFirst[np.str_('Long lasting'), np.str_('Easy to use...0.8812024-11-02254000.00
273B0D4QF30BGSports & Outdoors194.55QualityFirst Quality Sports & Outdoors ItemQuality sports & outdoors product for everyday...QualityFirst[np.str_('Long lasting'), np.str_('Easy to use...0.8812024-11-03253000.00
373B0D4QF30BGSports & Outdoors194.55QualityFirst Quality Sports & Outdoors ItemQuality sports & outdoors product for everyday...QualityFirst[np.str_('Long lasting'), np.str_('Easy to use...0.8812024-11-0400000.00
473B0D4QF30BGSports & Outdoors194.55QualityFirst Quality Sports & Outdoors ItemQuality sports & outdoors product for everyday...QualityFirst[np.str_('Long lasting'), np.str_('Easy to use...0.8812024-11-05101000.00
\n", "
" ], "text/plain": [ " product_identifier category price \\\n", "73 B0D4QF30BG Sports & Outdoors 194.55 \n", "173 B0D4QF30BG Sports & Outdoors 194.55 \n", "273 B0D4QF30BG Sports & Outdoors 194.55 \n", "373 B0D4QF30BG Sports & Outdoors 194.55 \n", "473 B0D4QF30BG Sports & Outdoors 194.55 \n", "\n", " title \\\n", "73 QualityFirst Quality Sports & Outdoors Item \n", "173 QualityFirst Quality Sports & Outdoors Item \n", "273 QualityFirst Quality Sports & Outdoors Item \n", "373 QualityFirst Quality Sports & Outdoors Item \n", "473 QualityFirst Quality Sports & Outdoors Item \n", "\n", " description brand \\\n", "73 Quality sports & outdoors product for everyday... QualityFirst \n", "173 Quality sports & outdoors product for everyday... QualityFirst \n", "273 Quality sports & outdoors product for everyday... QualityFirst \n", "373 Quality sports & outdoors product for everyday... QualityFirst \n", "473 Quality sports & outdoors product for everyday... QualityFirst \n", "\n", " features quality_score \\\n", "73 [np.str_('Long lasting'), np.str_('Easy to use... 0.881 \n", "173 [np.str_('Long lasting'), np.str_('Easy to use... 0.881 \n", "273 [np.str_('Long lasting'), np.str_('Easy to use... 0.881 \n", "373 [np.str_('Long lasting'), np.str_('Easy to use... 0.881 \n", "473 [np.str_('Long lasting'), np.str_('Easy to use... 0.881 \n", "\n", " date impressions visits cart_adds ordered_units revenue \n", "73 2024-11-01 100 14 2 1 194.55 \n", "173 2024-11-02 25 4 0 0 0.00 \n", "273 2024-11-03 25 3 0 0 0.00 \n", "373 2024-11-04 0 0 0 0 0.00 \n", "473 2024-11-05 10 1 0 0 0.00 " ] }, "execution_count": 8, "metadata": {}, "output_type": "execute_result" } ], "source": [ "metrics.sort_values([\"product_identifier\", \"date\"]).head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Let's examine a single metric record to understand the conversion funnel structure." ] }, { "cell_type": "code", "execution_count": 9, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:50.956787Z", "iopub.status.busy": "2026-03-01T03:45:50.956625Z", "iopub.status.idle": "2026-03-01T03:45:50.960610Z", "shell.execute_reply": "2026-03-01T03:45:50.959728Z" } }, "outputs": [ { "data": { "text/plain": [ "product_identifier B1P4DZHDS9\n", "category Electronics\n", "price 686.37\n", "title NexGen Smart Electronics Item\n", "description Quality electronics product for everyday use. ...\n", "brand NexGen\n", "features [np.str_('Long battery life'), np.str_('Voice ...\n", "quality_score 0.8\n", "date 2024-11-01\n", "impressions 25\n", "visits 4\n", "cart_adds 1\n", "ordered_units 0\n", "revenue 0.0\n", "Name: 0, dtype: object" ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "example_metric = metrics.iloc[0]\n", "example_metric" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This metric record tracks the shopper journey as a conversion funnel for one product on one date. It includes `product_identifier`, `category`, and `date` for identification, then follows shoppers through each stage: `impressions` (product shown), `visits` (clicked to view), `cart_adds` (added to cart), `ordered_units` (purchased), and `revenue` (total sales). This funnel structure captures how shoppers drop off at each stage of the purchase process." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Exploration\n", "\n", "Now that we understand the structure of both DataFrames, let's explore the generated data through visualizations and summary statistics." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "Things to explore\n", "\n", "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "How is revenue distributed across product categories?" ] }, { "cell_type": "code", "execution_count": 10, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:50.962276Z", "iopub.status.busy": "2026-03-01T03:45:50.962120Z", "iopub.status.idle": "2026-03-01T03:45:51.102214Z", "shell.execute_reply": "2026-03-01T03:45:51.101426Z" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "category_revenue = metrics.groupby(\"category\")[\"revenue\"].sum().sort_values()\n", "plot_revenue_by_category(category_revenue)" ] }, { "cell_type": "code", "execution_count": 11, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:51.103944Z", "iopub.status.busy": "2026-03-01T03:45:51.103766Z", "iopub.status.idle": "2026-03-01T03:45:51.111597Z", "shell.execute_reply": "2026-03-01T03:45:51.110847Z" } }, "outputs": [ { "data": { "text/html": [ "
def plot_revenue_by_category(category_revenue):\n",
       "    """\n",
       "    Plot horizontal bar chart of revenue by category.\n",
       "\n",
       "    Parameters\n",
       "    ----------\n",
       "    category_revenue : pandas.Series\n",
       "        Series with category names as index and revenue values.\n",
       "    """\n",
       "    fig, ax = plt.subplots(figsize=(10, 6))\n",
       "    category_revenue.plot(kind="barh", ax=ax, color=sns.color_palette("viridis", len(category_revenue)))\n",
       "    ax.set_xlabel("Revenue ($)")\n",
       "    ax.set_ylabel("Category")\n",
       "    ax.set_title("Total Revenue by Category")\n",
       "    ax.xaxis.set_major_formatter(plt.FuncFormatter(lambda x, _: f"${x:,.0f}"))\n",
       "    plt.tight_layout()\n",
       "    plt.show()\n",
       "
\n" ], "text/latex": [ "\\begin{Verbatim}[commandchars=\\\\\\{\\}]\n", "\\PY{k}{def}\\PY{+w}{ }\\PY{n+nf}{plot\\PYZus{}revenue\\PYZus{}by\\PYZus{}category}\\PY{p}{(}\\PY{n}{category\\PYZus{}revenue}\\PY{p}{)}\\PY{p}{:}\n", "\\PY{+w}{ }\\PY{l+s+sd}{\\PYZdq{}\\PYZdq{}\\PYZdq{}}\n", "\\PY{l+s+sd}{ Plot horizontal bar chart of revenue by category.}\n", "\n", "\\PY{l+s+sd}{ Parameters}\n", "\\PY{l+s+sd}{ \\PYZhy{}\\PYZhy{}\\PYZhy{}\\PYZhy{}\\PYZhy{}\\PYZhy{}\\PYZhy{}\\PYZhy{}\\PYZhy{}\\PYZhy{}}\n", "\\PY{l+s+sd}{ category\\PYZus{}revenue : pandas.Series}\n", "\\PY{l+s+sd}{ Series with category names as index and revenue values.}\n", "\\PY{l+s+sd}{ \\PYZdq{}\\PYZdq{}\\PYZdq{}}\n", " \\PY{n}{fig}\\PY{p}{,} \\PY{n}{ax} \\PY{o}{=} \\PY{n}{plt}\\PY{o}{.}\\PY{n}{subplots}\\PY{p}{(}\\PY{n}{figsize}\\PY{o}{=}\\PY{p}{(}\\PY{l+m+mi}{10}\\PY{p}{,} \\PY{l+m+mi}{6}\\PY{p}{)}\\PY{p}{)}\n", " \\PY{n}{category\\PYZus{}revenue}\\PY{o}{.}\\PY{n}{plot}\\PY{p}{(}\\PY{n}{kind}\\PY{o}{=}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{barh}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{n}{ax}\\PY{o}{=}\\PY{n}{ax}\\PY{p}{,} \\PY{n}{color}\\PY{o}{=}\\PY{n}{sns}\\PY{o}{.}\\PY{n}{color\\PYZus{}palette}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{viridis}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{n+nb}{len}\\PY{p}{(}\\PY{n}{category\\PYZus{}revenue}\\PY{p}{)}\\PY{p}{)}\\PY{p}{)}\n", " \\PY{n}{ax}\\PY{o}{.}\\PY{n}{set\\PYZus{}xlabel}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{Revenue (\\PYZdl{})}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\n", " \\PY{n}{ax}\\PY{o}{.}\\PY{n}{set\\PYZus{}ylabel}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{Category}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\n", " \\PY{n}{ax}\\PY{o}{.}\\PY{n}{set\\PYZus{}title}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{Total Revenue by Category}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\n", " \\PY{n}{ax}\\PY{o}{.}\\PY{n}{xaxis}\\PY{o}{.}\\PY{n}{set\\PYZus{}major\\PYZus{}formatter}\\PY{p}{(}\\PY{n}{plt}\\PY{o}{.}\\PY{n}{FuncFormatter}\\PY{p}{(}\\PY{k}{lambda} \\PY{n}{x}\\PY{p}{,} \\PY{n}{\\PYZus{}}\\PY{p}{:} \\PY{l+s+sa}{f}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{\\PYZdl{}}\\PY{l+s+si}{\\PYZob{}}\\PY{n}{x}\\PY{l+s+si}{:}\\PY{l+s+s2}{,.0f}\\PY{l+s+si}{\\PYZcb{}}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\\PY{p}{)}\n", " \\PY{n}{plt}\\PY{o}{.}\\PY{n}{tight\\PYZus{}layout}\\PY{p}{(}\\PY{p}{)}\n", " \\PY{n}{plt}\\PY{o}{.}\\PY{n}{show}\\PY{p}{(}\\PY{p}{)}\n", "\\end{Verbatim}\n" ], "text/plain": [ "def plot_revenue_by_category(category_revenue):\n", " \"\"\"\n", " Plot horizontal bar chart of revenue by category.\n", "\n", " Parameters\n", " ----------\n", " category_revenue : pandas.Series\n", " Series with category names as index and revenue values.\n", " \"\"\"\n", " fig, ax = plt.subplots(figsize=(10, 6))\n", " category_revenue.plot(kind=\"barh\", ax=ax, color=sns.color_palette(\"viridis\", len(category_revenue)))\n", " ax.set_xlabel(\"Revenue ($)\")\n", " ax.set_ylabel(\"Category\")\n", " ax.set_title(\"Total Revenue by Category\")\n", " ax.xaxis.set_major_formatter(plt.FuncFormatter(lambda x, _: f\"${x:,.0f}\"))\n", " plt.tight_layout()\n", " plt.show()" ] }, "execution_count": 11, "metadata": {}, "output_type": "execute_result" } ], "source": [ "Code(inspect.getsource(plot_revenue_by_category), language=\"python\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "How do metrics vary over time?" ] }, { "cell_type": "code", "execution_count": 12, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:51.113295Z", "iopub.status.busy": "2026-03-01T03:45:51.113125Z", "iopub.status.idle": "2026-03-01T03:45:51.262578Z", "shell.execute_reply": "2026-03-01T03:45:51.261675Z" } }, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAABKQAAAHqCAYAAAA6SZZrAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQAA6zRJREFUeJzs3Xl4VPXZPvD7zJKZrJN9I2Rh30QEFVEUUSq4Fmuttm6oVbHFV6HV1tZaq7WtO2r9ySu+dd+qVetGBBUBZZUthCWEJJA9IWTfZzm/P86ckzMkgSyznDNzf66Ly8lkZvJNTGbmPOd57q8giqIIIiIiIiIiIiIiPzEEegFERERERERERBRaWJAiIiIiIiIiIiK/YkGKiIiIiIiIiIj8igUpIiIiIiIiIiLyKxakiIiIiIiIiIjIr1iQIiIiIiIiIiIiv2JBioiIiIiIiIiI/IoFKSIiIiIiIiIi8isWpIiIiIiIiIiIyK9YkCIiIqKg8uqrr0IQBBw+fFi57vzzz8f5558fsDWRPvT1u0NERES+wYIUERER+Z184C//s1qtSE9Px/z58/Hcc8+hpaUl0Es8IfXaBUFATEwM5syZg88//zzQS9Ok7OzsXj+zvv69+uqrgV4qERER+Ykp0AsgIiKi0PXwww8jJycHdrsd1dXV+Pbbb3HPPffg6aefxieffIKpU6cO+jFvuOEGXHvttbBYLD5YcY8f/ehHuPHGGyGKIo4cOYIXX3wRl19+OVatWoX58+f79GvrzfLly9Ha2qp8/MUXX+Cdd97BM888g8TEROX6s88+OxDLIyIiogBgQYqIiIgC5uKLL8bpp5+ufHz//ffjm2++wWWXXYYrrrgC+/fvR3h4+KAe02g0wmg0enupvYwbNw7XX3+98vFVV12FSZMm4dlnn2VB6jgLFy70+Li6uhrvvPMOFi5ciOzs7H7v19bWhsjISN8ujoiIiAKCI3tERESkKRdccAH+9Kc/4ciRI3jzzTeV6/Py8rBo0SKMGjUKVqsVqampuOWWW3Ds2DGP+58sB6i1tRWRkZG4++67e32uvLwcRqMRf//73we97okTJyIxMRFFRUUe13d1deHPf/4zxowZA4vFgpEjR+K+++5DV1eXcpspU6Zg7ty5vR7T5XJhxIgR+OlPf+px3fLlyzF58mRYrVakpKTgjjvuQENDg8d9s7Ozcdlll+G7777DmWeeCavVilGjRuH111/3uN1DDz0EQRB6fe3+fo6rVq3Cueeei8jISERHR+PSSy/F3r17B/xz6s+iRYsQFRWFoqIiXHLJJYiOjsZ1113nk+8ZAPbu3YsLLrgA4eHhyMjIwF//+le4XK5hfx9EREQ0MCxIERERkebccMMNAIDVq1cr161ZswbFxcW4+eab8fzzz+Paa6/Fu+++i0suuQSiKA74saOionDllVfivffeg9Pp9PjcO++8A1EUlULIYDQ1NaGhoQFxcXHKdS6XC1dccQWefPJJXH755Xj++eexcOFCPPPMM7jmmmuU211zzTVYv349qqurPR7zu+++Q2VlJa699lrlujvuuAP33nsvzjnnHDz77LO4+eab8dZbb2H+/Pmw2+0e9z906BB++tOf4kc/+hGeeuopxMXFYdGiRUMuIL3xxhu49NJLERUVhcceewx/+tOfsG/fPsyePdsrQeAOhwPz589HcnIynnzySVx11VUAvP89V1dXY+7cudi1axd+//vf45577sHrr7+OZ599dtjfAxEREQ2QSERERORnr7zyighA3LZtW7+3sdls4mmnnaZ83N7e3us277zzjghAXL9+fa/HLikpUa6bM2eOOGfOHOXjL7/8UgQgrlq1yuPxpk6d6nG7/gAQb731VvHo0aNibW2t+MMPP4gLFiwQAYhPPPGEcrs33nhDNBgM4oYNGzzuv2LFChGA+P3334uiKIoFBQUiAPH555/3uN2vfvUrMSoqSvneN2zYIAIQ33rrLY/b5ebm9ro+Kyur18+mtrZWtFgs4m9+8xvluj//+c9iX28Jj/85trS0iLGxseJtt93mcbvq6mrRZrP1uv5EnnjiiV7/j2666SYRgPj73//e47a++J7vueceEYC4ZcsWj9vZbLZe6yIiIiLfYIcUERERaVJUVJTHbnvqLKnOzk7U1dXhrLPOAgDs2LFjUI89b948pKen46233lKuy8/PR15enkcu1In83//9H5KSkpCcnIzTTz8dX3/9Ne677z4sW7ZMuc3777+PiRMnYsKECairq1P+XXDBBQCAtWvXApDyqKZNm4b33ntPua/T6cQHH3yAyy+/XPne33//fdhsNvzoRz/yeLwZM2YgKipKeTzZpEmTcO655yofJyUlYfz48SguLh7UzwuQOtQaGxvx85//3ONrG41GzJw5s9fXHqo777zT42NffM9ffPEFzjrrLJx55pketxtKZxwRERENDUPNiYiISJNaW1uRnJysfFxfX4+//OUvePfdd1FbW+tx26ampkE9tsFgwHXXXYcXX3wR7e3tiIiIwFtvvQWr1Yqrr756QI/x4x//GEuWLEF3dze2bduGv/3tb2hvb4fB0HO+r7CwEPv370dSUlKfj6H+Pq655hr84Q9/QEVFBUaMGIFvv/0WtbW1HqN9hYWFaGpq8vi59Pd4AJCZmdnrNnFxcb2ylwaisLAQAJRi2vFiYmIG/ZjHM5lMyMjI6PV1vf09HzlyBDNnzux1u/Hjxw9l2URERDQELEgRERGR5pSXl6OpqQljxoxRrvvZz36GjRs34t5778W0adMQFRUFl8uFBQsWDCmM+sYbb8QTTzyBjz/+GD//+c/x9ttv47LLLoPNZhvQ/TMyMjBv3jwAwCWXXILExEQsWbIEc+fOxU9+8hMAUobUKaecgqeffrrPxxg5cqRy+ZprrsH999+P999/H/fccw/+/e9/w2azYcGCBcptXC4XkpOTPTq71I4vfPW326CoytzqK9AcQK98Lfln/MYbbyA1NbXX7U2m4b+ttFgsHgU9+et6+3smIiKiwGNBioiIiDTnjTfeAADMnz8fANDQ0ICvv/4af/nLX/Dggw8qt5O7doZiypQpOO200/DWW28hIyMDpaWleP7554f8eHfccQeeeeYZPPDAA7jyyishCAJGjx6N3bt348ILL+y38CPLycnBmWeeiffeew9LlizBhx9+iIULF8JisSi3GT16NL766iucc845HiOMwyGHsDc2NiI2Nla5/siRIx63Gz16NAAgOTlZKcT5gy++56ysrD5/dwoKCrzy+ERERHRyzJAiIiIiTfnmm2/wyCOPICcnR8n0kbteju9yWb58+bC+1g033IDVq1dj+fLlSEhIwMUXXzzkxzKZTPjNb36D/fv347///S8AqauroqICK1eu7HX7jo4OtLW1eVx3zTXXYPPmzfjXv/6Furo6j3E9+fGcTiceeeSRXo/ncDjQ2Ng46HXLhab169cr17W1teG1117zuN38+fMRExODv/3tb712tgOAo0ePDvprD4QvvudLLrkEmzdvxtatW5Xrjh492m8XFhEREXkfO6SIiIgoYFatWoUDBw7A4XCgpqYG33zzDdasWYOsrCx88sknsFqtAKR8ovPOOw+PP/447HY7RowYgdWrV6OkpGRYX/8Xv/gF7rvvPnz00Ue48847YTabh/V4ixYtwoMPPojHHnsMCxcuxA033IB///vfWLx4MdauXYtzzjkHTqcTBw4cwL///W98+eWXOP3005X7/+xnP8Nvf/tb/Pa3v0V8fHyvTqQ5c+bgjjvuwN///nfs2rULF110EcxmMwoLC/H+++/j2WefxU9/+tNBrfmiiy5CZmYmbr31Vtx7770wGo3417/+haSkJJSWliq3i4mJwYsvvogbbrgB06dPx7XXXqvc5vPPP8c555yDf/7zn8P6+fXFF9/zfffdhzfeeAMLFizA3XffjcjISLz00kvIyspCXl6e178HIiIi6o0FKSIiIgoYefwuLCwM8fHxOOWUU7B8+XLcfPPNiI6O9rjt22+/jbvuugsvvPACRFHERRddhFWrViE9PX3IXz8lJQUXXXQRvvjiC9xwww3D+l4AaSfAJUuW4KGHHsK3336L888/Hx9//DGeeeYZvP766/joo48QERGBUaNG4e6778a4ceM87p+RkYGzzz4b33//PX75y1/2WSBbsWIFZsyYgf/93//FH/7wB5hMJmRnZ+P666/HOeecM+g1m81mfPTRR/jVr36FP/3pT0hNTcU999yDuLg43HzzzR63/cUvfoH09HT84x//wBNPPIGuri6MGDEC5557bq/bepO3v+e0tDSsXbsWd911F/7xj38gISEBixcvRnp6Om699VYffAdERER0PEFkwiMRERGFsCuvvBJ79uzBoUOHAr0UIiIiopDBDCkiIiIKWVVVVfj888+90h1FRERERAPHkT0iIiIKOSUlJfj+++/x8ssvw2w244477gj0koiIiIhCCjukiIiIKOSsW7cON9xwA0pKSvDaa68hNTU10EsiIiIiCinMkCIiIiIiIiIiIr9ihxQREREREREREfkVC1JERERERERERORXDDUPIJfLhcrKSkRHR0MQhEAvh4iIiIiIiIhCkCiKaGlpQXp6OgwG//QusSAVQJWVlRg5cmSgl0FEREREREREhLKyMmRkZPjla7EgFUDR0dEApP/hMTExAV4NEREREREREYWi5uZmjBw5UqlT+AMLUgEkj+nFxMSwIEVEREREREREAeXPOCGGmhMRERERERERkV+xIEVERERERERERH7FghQREREREREREfkVC1JERERERERERORXLEgREREREREREZFfsSBFRERERERERER+xYIUERERERERERH5FQtSRERERERERETkVyxIERERERERERGRX7EgRUREREREREREfmUK9AKIiLwpN78Ky78qREldG3ISI3HPvLFYMCUt0MsiIiIiIiIiFXZIEVHQyM2vwuI3d6CgugVdDhcKqluw+M0dyM2vCvTSiIiIiIiISIUFKSIKGsu/KoQAQHR/LAIQBODZrwsDuCoiIiIiIiI6HgtSRBQ0SuralGKUTBSB4qNtAVkPERERERER9Y0FKSIKGolRYX1en5MY6eeVEBERERER0YmwIEVEQSG/ogm1LV19fi4nMcLPqyEiIiIiIqITYUGKiHTvaEsXbn/9B9id0sBeRJgRJoOgfH71vlrkVzQFanlERERERER0HBakiEjXuh0u3PnmdlQ2dQIARiVG4qmrT8WK62fgx6emAwCcLhG/fX83uh2uQC6ViIiIiIiI3FiQIiLdEkURD/43Hz8caQAAxEWY8avzR8NslJ7aLp6Sioy4cADAgeoWvPhtUcDWSkRERERERD1YkCIi3Xp90xG8u60MAGA2CvjV+WMQG9ETbG4yGnDz2dmQp/f+ubYQB6qbA7FUIiIiIiIiUmFBioh06ftDdXj4s33Kx4tmZfe5m15WQiQWTEkFANidIu77IA8OJ0f3iIiIiIiIAokFKSLSnSPH2vCrt3bA6ZJCzC+ekoqZoxL6vf3lU9ORZrMCAPLKm/DydyV+WScRERERERH1jQUpItKVlk47fvnaD2jqsAMApo6w4cppI054H7PRgEVnZ0Ped+/pNQdRdLTVxyslIiIiIiKi/rAgRUS64XKJWPreLhTWSsWkNJsVt507CgY5JOoERidFYd6kFADSznz3fZCndFgRERERERGRf7EgRUS68dSaAny1vxYAEBFmxJK5YxAeZhzw/RdOS0dStAUAsP1IA17fdNgXyyQiIiIiIqKTYEGKiHThv7sq8MLaIgCAQQAWnzcaKTHWQT2GxWTEolnZyseP5xag9Fi7N5dJREREREREA6CZgtSiRYsCvQQi0qg95U2474M85eOfnT4Sk9JjhvRY41OjMXd8EgCgw+7E7/6TB1Hk6B4REREREZE/aaYg1ZcPP/wQF110ERISEiAIAnbt2tXrNp2dnfj1r3+NhIQEREVF4aqrrkJNTc0JH1cURTz44INIS0tDeHg45s2bh8LCQo/b1NfX47rrrkNMTAxiY2Nx6623orXVMwQ5Ly8P5557LqxWK0aOHInHH3982N8zEXmqbenE7W/8gC6HCwAwe0wiLpyQPKzHvGp6BuIjwwAAm4qP4Z2tZcNeJxEREREREQ1cQAtSdXV1uOmmm5CZmYl33nkHY8aMwdVXX43u7m4AQFtbG2bPno3HHnus38dYunQpPv30U7z//vtYt24dKisr8ZOf/OSEX/fxxx/Hc889hxUrVmDLli2IjIzE/Pnz0dnZqdzmuuuuw969e7FmzRp89tlnWL9+PW6//Xbl883NzbjooouQlZWF7du344knnsBDDz2El156aZg/FSKSdTmcWPzGdlQ1SX+bo5Micd3MTAjCyUPMT8RqNuKmWVnKx3/7Yj8qGzuG9ZhEREREREQ0cIIYwFmVG264AVu3bsVLL72E5cuX43/+53+Qm5uLv/zlL7Bae7JhDh8+jJycHOzcuRPTpk1Trm9qakJSUhLefvtt/PSnPwUAHDhwABMnTsSmTZtw1lln9fqaoigiPT0dv/nNb/Db3/5WeZyUlBS8+uqruPbaa7F//35MmjQJ27Ztw+mnnw4AyM3NxSWXXILy8nKkp6fjxRdfxB//+EdUV1cjLEzqtPj973+Pjz/+GAcOHBjQ99/c3AybzYampibExAxt/IgoWImiiHs/yMMH28sBAHERZjxw6STYws1e+xqvbjyM7w7VAQDmjEvCqzefMexiFxERERERkd4Eoj4R0A6pnTt34sYbb8ScOXNgs9kwd+5cPPbYYx7FqBPZvn077HY75s2bp1w3YcIEZGZmYtOmTX3ep6SkBNXV1R73sdlsmDlzpnKfTZs2ITY2VilGAcC8efNgMBiwZcsW5TbnnXeeUowCgPnz56OgoAANDQ19fu2uri40Nzd7/COivv3r+8NKMSrMaMCSuWO8WowCgJ+dnqE85rqDR/GfHRVefXzSjtz8KixYvh7jH1iFBcvXIze/KtBLIiIiIiIKaQEtSJ1zzjl45ZVX8Nlnnw3p/nJ3UmxsrMf1KSkpqK6u7vc+8m36u091dTWSkz0zakwmE+Lj4z1u09djqL/G8f7+97/DZrMp/0aOHDmA75Io9Kw/eBSPfr5P+XjR2dnISoj0+teJCDPhhrN6Rvce/nQvaps7T3AP0qPc/CosfnMHCqpb0OVwoaC6BYvf3MGiFBERERFRAAW0IPX000/jmmuuwdKlS/H6669j2rRpWLFiRSCX5FP3338/mpqalH9lZQxSJjpeSV0blry9Ay73MPGlp6ThzJx4n329aSNjMdP9+M2dDvzx43zuuhdkln8lbVoh/18VAQgC8OzXhf3eh4iIiIiIfCugBanIyEg8+uijKCwsxBVXXIE777wTy5YtG3AweGpqKrq7u9HY2OhxfU1NDVJTU/u9j3yb/u6TmpqK2tpaj887HA7U19d73Kavx1B/jeNZLBbExMR4/COiHs2ddvzytW1o7nQAkIpFP56W7vOv+/MzMhFtNQEA1uyrwWd57JwJJiV1bb2uE0Wg+Gjv64mIiIiIyD8CWpBSi42NxR133IGLL74YGzZsGNB9ZsyYAbPZjK+//lq5rqCgAKWlpZg1a1af98nJyUFqaqrHfZqbm7FlyxblPrNmzUJjYyO2b9+u3Oabb76By+XCzJkzldusX78edrtduc2aNWswfvx4xMXFDfwbJyIAgNMl4u53dqLIXSRIj7Xil7NzYPBDyHiU1YTrZmYqH//5k7041trl869L/pGT2HvcUxCAUUneHwMlIiIiIqKBCWhBaunSpVi3bh2amprgdDqxdu1arFu3DjNmzAAA1NfXY9euXdi3T8qSKSgowK5du5SMJpvNhltvvRXLli3D2rVrsX37dtx8882YNWtWnzvsAYAgCLjnnnvw17/+FZ988gn27NmDG2+8Eenp6Vi4cCEAYOLEiViwYAFuu+02bN26Fd9//z2WLFmCa6+9FunpUrfGL37xC4SFheHWW2/F3r178d577+HZZ5/FsmXLfPxTIwpOT3xZgLUFRwEAkWFGLJk7Blaz0W9f//SseMzIlIrJ9W3deOjTfSe5B+nFr+eO6XWdKAJ3XzguAKshIiIiIiIgwAWpzMxMLFu2DCNHjsTbb7+NG2+8EbfccgvuuusuAMAnn3yC0047DZdeeikA4Nprr8Vpp53mkTP1zDPP4LLLLsNVV12F8847D6mpqfjwww89vk52djYeeugh5eP77rsPd911F26//XacccYZaG1tRW5ursfufm+99RYmTJiACy+8EJdccglmz57tMUpos9mwevVqlJSUYMaMGfjNb36DBx98ELfffrsvflREQe3jnRVYsa4IAGAQgMVzRiM5emC7bXrTL2ZmIjJMKoJ9ursSX+7te4MC0peJadEeH0dbTFhx/QwsmNL3eDUREREREfmeIGokvXfRokV49dVXvf647e3tSEhIwKpVq3D++ed7/fGHo7m5GTabDU1NTcyTopC1u6wRV//vJnQ7XACAX5yZiQsmJJ/kXr6zqfgY/u+7EgBAUrQFXy2dA1uEOWDroeH7cm817nijZwR7YloMVt19bgBXRESkbbn5VVj+VSFK6tqQkxiJe+aNxYIpaYFeFhER+VAg6hOayZDylbVr1+KCCy7QXDGKiIDa5k7c/sYPSjHqvLGJmDs+KaBrOisnHlNH2AAAR1u68MjnHN3Tu+NDzY8ca+NOikRE/cjNr8LiN3fgQHULuhwuFFS3YPGbO5Cbzw0/iIjIuzRTkPJFdxQAXHrppfj888998thENHSddiduf2M7apql8PCxyVH4xZmZEPwQYn4igiDg+rOyEO7Or/pgezm+Lag9yb1Iy4qPtnp83N7txFGG1hMR9Wn5V4VQvxKLkDaCePbrwkAtiYiIgpRmClJEFDpEUcQfPtyDXWWNAID4yDDcOWc0TEZtPCXFR4bh6tMzlI//8OEetHTaT3AP0rLio229ris91h6AlRARaV9JXRuO7yEVxb6fS4mIiIZDG0d/RBRSXt5Qgg93VgAAwkwGLJk7BjHh2sppOndMIiamSmHYlU2d+PuqAwFeEQ1VcV3vg6jDLEgREfUpMyGi13WCAIxKigzAaoiIKJixIEVEfvVtQS3+vmq/8vGt5+QgM773m99AEwQBN87KhsUkPU2+vaUUGw/VBXhVNFiN7d2ob+sGABhV46BHjvFMPxFRX2ZkxvW6ThSBuy8cF4DVEBFRMGNBioj8puhoK+56Zydc7lmAy6emYUZW7ze+WpEUbcFV03tG9373YR7aux0BXBENVpFqxGRMcpRymR1SRES9iaKI3eVNva5/+menYsGU1ACsiIiIghkLUkTkF00ddtz22g9o6ZQKOqdlxuLyU9MDvKqTO398Esa6Cxll9R144suCAK+IBkMdaD5lRIwS1FvKDikiol72VDRhf1Vzr+vHpUQHYDVERBTsWJAiIp9zukTc9c5OJctnRGw4bj0nB4YA76g3EAZBwKKzs2E2Smt9deNh/HC4PsCrooFS50eNiA1HfGQYAHZIERH15d1tZcrllGiLcrm0ns+ZRETkfSxIEZHP/WPVfqw/eBQAEGUx4a4LxsBqNgZ4VQOXEmPFwmkjAEg5Gvd9kIdOuzPAq6KBUHdIpdqsSHYfYDV12NHY3h2oZRERaU5blwOf7KoEAFhMBsyf3DOix4IUERH5AgtSRORT/9lejpUbSgBIodJ3zhmNxCjLSe6lPT+amIKcRGmHoeK6Niz/qjDAK6KBkLcpNxkEJEZakBxjVT53hF1SRESKz/Oq0NoljdWfmR2PjPhw5XMsSBERkS+wIEVEPrOjtAH3f7hH+fjnZ47E+FR95lAYDNLonskgje69tL4Iu8saA7soOiGnS1SKTsnRFhgMApJUxdDDzJEiIlK8u61UuXzu2ESP58syFqSIiMgHWJAiIp+oburEHW9sR7fTBQA4f1wSzh+fHOBVDc+I2HBcNjUNAOByj+51O1wBXhX1p7yhXfn9S7FJnVHJMT0HWOyQIiKSHKxpwY7SRgDSa11OYiSiLCZYTNKhAgtSRETkCyxIEZHXddqduP2NH3C0pQsAMD4lGteeOTLAq/KOBVNSMTJOGmMoqGnBC2sPBXhF1B95XA8AUt2jesnRLEgRER3v3a09Yebnjk2EIAgQBAFJ7ufM8oYOOF1ioJZHRERBigUpIvKa3PwqLFi+HpMezEVeeRMAIDEqDIvnjILJEBxPNyaDATefkwOje4fAF9Ye6nOLbAq8InWgubsgpR5BOcKRPSIidNqd+HBnOQApb++sUQnK5+TnTIdLRFVTR0DWR0REwSs4jhCJKOBy86uw+M0dOFDdAvVJ1AsmJCPaag7cwnwgMz4CF0+Rdh9yuETc+8Fu2J0c3dOa4jpVh5R7ZM9iNsIWLv0+HmaHFBERVu+rQWO7HQAwIysOURaT8rlEVVcpg82JiMjbWJAiIq9Y/lUhhD6u31h0zO9r8YdLp6Yh3V3kyK9oxkvriwO8IjpesapDKkW1u548tlfX2qXsKEVEFKre3eoZZq7GYHMiIvIlFqSIyCtK6trQV7pEdVOn39fiD2ajAYvOzoZ7cg/PflWIQ7UtgV0UeZAzpKIsJo8z/knqM/7skiKiEHbkWJty4ig52oLxKZ474aqfL8vqObJHRETexYIUEXlFTmJkr+sE9IxKBaNRSVG4aGIKAKDb6cK9H+Qx9FUjWjrtqHWH6qfGeP4OegabM0eKiELXe9t6h5mrJUaFKZc5skdERN7GghQRecWv547x+FgAIAK44tT0gKzHX348bQRS3AWOnaWNeOX7kgCviACpY092fFE0ObrnY+ZIEVGocjhdeH+7FGZuFAScPTqx120So5ghRUREvsOCFBF5RZrqoF8QgBFx4fjV+aMxPTMugKvyvTCTATedna18/OTqAhyuY9dNoMnjegCQEmPx+FxyjPoAi/+viCg0fXOgFkfdnaRTR9qUDR/UzEYD4iKk65khRURE3saCFBF5xSZVePlNZ2XjocsnB30xSjYuJRoXjE8GAHTaXfjdf/Lg4uheQKkDzU80sne4jgdYRBSa1ON6541N6vd2cpfUsbZubgRBRERexYIUEXnFpuKegtT41OgT3DI4/WT6CCVrY0tJPd5S7VpE/ld0gpG9iLCekHNmSBFRKKpq6sDagloAQHxEGCanxfR7W89gcxbxiYjIe1iQIqJh67Q7sf1IAwAgITLM481rqLCajbhpVrby8T++2I/yBr5xDxR5ZM8geG5bLpO7pKqaO9Fpd/p1bUREgfb+D+WQG3nPGZMAg0Ho97bq51AWpIiIyJtYkCKiYdtZ2oguhwsAMCEEu6NkE9NicN5YKRS2rduJi55Zj/EPrMKC5euRm18V4NWFDpdLREmdNLKXFGWBydj7pU4umooiWDgkopDiconKuJ4AYPaY3mHmagw2JyIiX2FBioiGTT2uNyG1/7b/UPDTGRmIDDMCANq7nehyuFBQ3YLFb+5gUcpPpK4nqUCacty4now5UkQUqr47VIeKxg4AwOT0GCT00UWqlhgdplxmhxQREXkTC1JENGybi0I7P0otIswEq9nocZ0IaefBZ78uDMyiQsyJAs1lydE91x9mjhQRhRB1mPm5JwgzlyWxQ4qIiHyEBSkiGpaObid2lkn5UcnRFsRHhp3kHsGvqcPe6zpR7Mk1It9S/5z7K0ipc854gEVEoeJYaxdW76sGAERbTTg1w3bS+9jCzTAbpYwpPl8SEZE3sSBFRMPyw5F62J1SMmoo50epHb+rGyB1SI1KigzAakKPR4fUQEb2jvEAi4hCw4c7KpTX7HNGJ/aZsXc8QRCULqmyhg645DR0IiKiYWJBioiGZVMR86OOd/nU9F7XiSJw94XjArCa0FNc19MhldJPh1S01QSrWXoJPMKRPSIKAaIo4p1tpcrHs8eeOMxcLdFdxO92uHC0tcvrayMiotDEghQRDYs60DzU86NkM7LicMd5oyBvom02Clhx/QwsmJIa0HWFCnlkL9xsRIzV1OdtBEFQcqQqGjpgd7r8tj4iokD44UiD8vw4LiWq35HmvnCnPSIi8gUWpIhoyFq7HMgrbwIApNmssIWbA7wi7TgjOx45idKInt0pYtbohACvKDR02p2obJJ2j0qJsUAQhH5vK+dIOVwiKt07ThERBat3tvZ0Rw0kzFzNI9icY85EROQlLEgR0ZBtK6mH08X8qP5kJUQol/dWNgVwJaGjpK4NojvepL/8KBlzpIgoVDR12PHFnioAQESYETMy4wZ1f24EQUREvsCCFBENmXpcj/lRvWXF94SY51ewIOUPA9lhT6YuSDFHioiC2Se7KtBpl0aTz8pJQJhpcIcA6g6pMhakiIjIS1iQIqIhUweaj0uJCuBKtEndIbWnojmAKwkdHjvsnaQgleRRkOIBFhEFr3e3lSmXzx038DBzWWJUmHKZHVJEROQtLEgR0ZA0tduVMbSMuHBEW5kfdby0WCtMBinDaC87pPzCY4e9k47s9XyeHVJEFKz2lDdhb6V0UiQ7IQIj4yJOco/eLKpNIliQIiIib2FBioiGZEvJMbjjo5gf1Q+TwYCMuHAAUqGkpdMe4BUFP7lDSgCQEn3iglRshBlmo1QwZIYUEQWrd7b1hJmfN8gwczV5p73ali502p3DXhcRERELUkQ0JMyPGpishJ4cKfkMNfmGKIpKhlR8ZNhJM1IMgqDkopTWt8MlV1iJiIJEe7cDn+yqBABYTAacmRM/5MdSjzmXN7CIT0REw8eCFBENiZwfJQjMjzoRdY4Ug81962hrF1q6HABOnh8lk8f2uh0uVDd3+mxtRESB8FleFVrdz4tnZMfDajYO+bHUweYc2yMiIm9gQYqIBq2+rRsHqlsAAJnxEYgIMwV4RdqVFc+ClL+od9g7WX6ULCmm5wDrMHOkiCjIvKcOMx87+DBztURVh1Qpx5yJiMgLWJAiokHb7DGux/yoExkRG64Em+9hQcqn1AWpgXdI8QCLiILTwZoWbD/SAEB6LRqVGHmSe5yYZ4dUx7Aei4iICGBBioiGQB7XA5gfdTImowEjVMHmbe7RCfI+OdAcGFpBisHmRBRMju+OEgRhWI+nzpDiyB4REXkDC1JENGhyoLlBAMYmMz/qZOSxPVEE9lUx2NxXiutUHVIDHNlLVu3Ed4Qje0QUJLocTny4oxwAYDIIOGtUwrAfMzbcrHT8MtSciIi8gQUpIhqU2pZOHKqVOlGyEyKHFZAaKtQ77e0p59ier8gdUmEmA2IjzAO6T3xkGIzurgF2SBFRsFi9twYN7XYAwIysOERZhp/1aDAISIgMAyB1SIkidyYlIqLhYUGKiAbFY1wvjflRA+ERbF7JgpQvdDtcKGuQMk1Soi0wDHA0xWgQkBDlPsA61sYDLCIKCu9uK1UuDzfMXE0ONm/vduJYW7fXHpeIiEITC1JENCgegeYpzI8aiBFx4UoXDnfa843S+jY4XVIxaaDjejI5R6qt24m6Vh5gEZG+lR5rx/eHpNfq5GgLxqd47+SRZ7A5u0qJiGh4WJAiokGRO6SMBgGjk4e3Y0+oMBsNSI+ViiSHalvR3s1gc28rGsIOezLmSBFRMHnvh57uqNljhh9mrqYONi9jQYqIiIaJBSkiGrDKxg4lZ2dUYiQsJuZHDZScI+USgf1VLQFeTfApHkZBSn2AdYQ5UkSkYw6nC+//IIWZGwTgnDHeG9cDgER1hxSfL4mIaJhYkCKiAfPIj0plftRgeORIcWzP6+RAcwBIGezIXoy6IMUOKSLSr7UFR1Hb0gUAODUjFrbwgW3wMFAeHVLcaY+IiIZJNwWpRYsWBXoJRCFvkzo/KpX5UYORldBTkNrDgpTXFdcNZ2Sv5wCLO+0RkZ69u9U3YeayRPcmEAAzpIiIaPh0U5DqS2trK5YsWYKMjAyEh4dj0qRJWLFixUnv9/7772PChAmwWq045ZRT8MUXX3h8XhRFPPjgg0hLS0N4eDjmzZuHwsJCj9vU19fjuuuuQ0xMDGJjY3HrrbeitbUVRMFM7pAyGwWMSmJ+1GBkxEXA4I7xYIeU98kdUrHhZljNgxslTYyyQE5YYYcUEelVdVMn1hbUAgDiIsyYkm7z+teICDMhMkx6ji2r7/D64xMRUWjRdEGqrq4ON910EzIzM/HOO+9gzJgxuPrqq9HdLe2CtGzZMuTm5uLNN9/E/v37cc8992DJkiX45JNP+n3MjRs34uc//zluvfVW7Ny5EwsXLsTChQuRn5+v3Obxxx/Hc889hxUrVmDLli2IjIzE/Pnz0dnZqdzmuuuuw969e7FmzRp89tlnWL9+PW6//Xbf/TCIAqysvh0VjdKbz9FJUTAbNf30oTlhJgPSbOEAgMLaVnTanQFeUfBoaOtGQ7sdwOB32AOk0Pm4SOms/xGe8ScinXr/hzK4NxvF7DGJMBi8F2auJo/tVTZ1oNvh8snXICKi0KDpI8qlS5di8+bNeOONN3DJJZdg5cqVGDVqFFwu6cVv48aNuOmmm3D++ecjOzsbt99+O0499VRs3bq138d89tlnsWDBAtx7772YOHEiHnnkEUyfPh3//Oc/AUjdUcuXL8cDDzyAH//4x5g6dSpef/11VFZW4uOPPwYA7N+/H7m5uXj55Zcxc+ZMzJ49G88//zzeffddVFZW+vznQhQIG4vqlMvMjxoaeWzP6RKxv6o5wKsJHsV1qvyoQY7ryeSxvcZ2Oxrbu72yLiIif3G5RLz3QxkAQIBUkPIVOdhcFKGcqCIiIhoKTRekdu7ciRtvvBFz5syBzWbD3Llz8dhjj8FqlQ44zj77bHzyySeoqKiAKIpYu3YtDh48iIsuuqjfx9y0aRPmzZvncd38+fOxadMmAEBJSQmqq6s9bmOz2TBz5kzlNps2bUJsbCxOP/105Tbz5s2DwWDAli1bvPb9E2mJZ6A586OGIjuhZ8yRY3veUzSMHfZkydxpj4h07PuiOpQ3SMWhSekxSFDthudt6mBz5kgREdFwmAK9gBM555xz8Morr+DUU0/t8/PPP/88br/9dmRkZMBkMsFgMGDlypU477zz+n3M6upqpKSkeFyXkpKC6upq5fPydSe6TXJyssfnTSYT4uPjldv0paurC11dXcrHzc3skCB9EEVRCTS3mAzITow4yT2oL+pg8/wK/v17S7G6IDWEkT3A8wDrSH07Th0ZO9xlERH5zbvbypTL541N8unXSlIVu8pYkCIiomHQdIfU008/jWuuuQZLly7F66+/jmnTpnmElj///PPYvHkzPvnkE2zfvh1PPfUUfv3rX+Orr74K4Kr79/e//x02m035N3LkyEAviWhASuraUNMsFVPHJEfBZND0U4dmZcSFQ3BHenCnPe+RA82B4XRI9dzvSB2DzYlIP461dmH1XumEaLTVhFMzvB9mrpbIghQREXmJpo8qIyMj8eijj6KwsBBXXHEF7rzzTixbtgwvvfQSOjo68Ic//AFPP/00Lr/8ckydOhVLlizBNddcgyeffLLfx0xNTUVNTY3HdTU1NUhNTVU+L193otvU1tZ6fN7hcKC+vl65TV/uv/9+NDU1Kf/Kysr6vS2Rlmz0GNdjftRQWUxGpLk7eA7WtDDY3EtK3AUkk0FAQmTYSW7dt+SYngOswxzZIyId+WhnBexOKc387NEJMPl40xGO7BERkbdouiClFhsbizvuuAMXX3wxNmzYALvdDrvdDsNxnRpGo1EJPe/LrFmz8PXXX3tct2bNGsyaNQsAkJOTg9TUVI/bNDc3Y8uWLcptZs2ahcbGRmzfvl25zTfffAOXy4WZM2f2+7UtFgtiYmI8/hHpgTyuBzA/ariy4qUcKYdLxMGalgCvRv+cLlHJfEqOsQx5V6nkKPUBFjukiEgfRFHEO1tLlY/PHePbcT0AiIs0Q36qZUGKiIiGQ9MFqaVLl2LdunVoamqC0+nE2rVrsW7dOsyYMQMxMTGYM2cO7r33Xnz77bcoKSnBq6++itdffx1XXnllv4959913Izc3F0899RQOHDiAhx56CD/88AOWLFkCABAEAffccw/++te/4pNPPsGePXtw4403Ij09HQsXLgQATJw4EQsWLMBtt92GrVu34vvvv8eSJUtw7bXXIj093R8/GiK/EUURW9wFqXCzEZnxzI8aDnWOFMf2hq+8oR3dTukkxFDH9QDAYjbCFm4GwA4pItKP7UcalI0dxqVEDTlHbzBMBgPi3d2opcfaIYqiz78mEREFJ02HmmdmZmLZsmUoLCxEW1sbvv32W9xyyy246667AADvvvsu7r//flx33XWor69HVlYWHn30USxevFh5jEWLFuHw4cP49ttvAUg787399tt44IEH8Ic//AFjx47Fxx9/jClTpij3ue+++9DW1obbb78djY2NmD17NnJzc5Xd/QDgrbfewpIlS3DhhRfCYDDgqquuwnPPPeefHwyRHxXWtqKutRsAMDYlCsYhdqCQJCteHWzOgtRwFXthhz1ZcrQFTR12HG3pQluXA5EWTb9EEhHhna098Q/+6I6SJUVZUNfajZYuB5o67IiNGNq4NBERhTZNv9teunQpli5dCkAqLL366qsen09NTcUrr7xywscoKSnB3LlzPa67+uqrcfXVV/d7H0EQ8PDDD+Phhx/u9zbx8fF4++23T/IdEOnfxkN1ymXmRw3fyPgICABEcKc9byhSBZqnDLMglRRtQWGt9HhHjrVjUjrHU4lIu5o77fh8TyUAqYN5Rlac3752UrQF+6ulsfOy+g4WpIiIaEg0PbI3XE1NTSgqKsJvf/vbQC+FSLeYH+VdVrMRKe6RioLqFnQ7+s+8o5MrVu2IN9xRleRo5kgRkX78d1clOu3Sa8hZo+IRZvLf2/rEKAabExHR8OmmIHV8d9RA2Gw2lJeXIyoqyvsLIgoBLpeILSX1AIDIMCMy4sIDvKLgII/tdTtdDDYfpmIvdkglR/fcnzlSRKR176rCzM8b679xPYAFKSIi8g7dFKSIyP/2Vzejsd0OABiXGg2DwPwob1AHmzNHanjkDKkoiwlRw8x8UndIHTnGDiki0q78iibsrZTGvrMTIjDSzxuOJEWzIEVERMPHghQR9WtTUc+43kSO63lNVnykcpk77Q1dS6cdtS1dAIYfaA54HmAdYYcUEWnYO6ruqHP93B0FSKHmsjIWpIiIaIhYkCKifqkLUuMZaO41mdxpzytKvJgfBQCRFhMiw4wAWJAiIu1q73bgk11SmLnFZMDMnHi/ryHSYkS4WXq+ZIcUERENFQtSRNQnh9OFre78qGirCeleOOAnSXiYESkx0tnl/dUtsDsZbD4U8rgeAOXnOVzJ7k6ryqYOdDmcXnlMIiJv+jyvCi1dDgDAGdnxsLoLQ/4kCAISo6Sd9SoaO+Dg6xgREQ0BC1JE1Ke9lc3KG97xKdEQmB/lVfLYXrfDhcKa1pPcmvqiDjT3xsge0JMjJYrSVuZERFrz3rYy5fK5YxMDto5E9/Ol0yWiqqkzYOsgIiL9YkGKiPq0UZ0flcb8KG9jsPnwFXl5ZA84PkeKweZEpC2FNS344UgDACA91opRiZEnuYfvMEeKiIiGiwUpIurTpmLmR/mSuiDFYPOhkUf2DILngdFwqHfaO8wcKSLSGHV31HljkwLavax+3mWOFBERDQULUkTUi93pwg+Hpfyo2HAzUqK9c7BPPTyCzStZkBosl0tESZ00spcUZYHJ6J2Xs+Tonk4rdkgRkZZ0OZz4z45yAIDJIOCsnISArkfdUcqCFBERDYUp0AsgGo7c/Cos/6oQJXVtyEmMxD3zxmLBlLRAL0v38sob0d4tBTqPT2V+lC9EhJmQFG3B0ZYu7K9qhsPp8lpRJRRUNXei0y6F6KZ4MXDfc2SPB1hEpB1r9tWgod0OAJieGYcoa2DfxieyIEVERMPEox/Srdz8Kix+cwcOVLegy+FCQXULFr+5A7n5VYFemu5tPKTKj0plfpSvZLm7pDrtLhw6ymDzwfBFoDkAxFhNsJikl0Z2SBGRlry7VRth5rKEyDDIp6uYIUVEREPBghTp1vKvCqHu2xEBCALw7NeFgVpS0GB+lH94Bps3B3Al+iPnRwHeLUgJgqDkSJU3cCtzItKG0mPt+O5QHQCpk1MLr81mowFxEWEAgLIG7kpKRESDx4IU6VZJXRvE464TRc8DVRq8LocT2907+CREhnmMMJF3ZcX37I7EnfYGx6NDyosje0BPjpTDJaKykVuZE1Hg/fsHVXfUmEQYNDJKnxgtFaTq27rR0mkP8GqIiEhvWJAi3crpY6tjQQBGJQVuC+RgsLO0EV0OqStEC2dgg1kmd9obsuK6nsJzihc7pADPHKnDHNsjogBzOF14f7tUkDIIwNmjAxtmrpao2mmvrJ5dUkRENDgsSJFu3TNvbK/rRBG4+8JxAVhN8NhYxPwof4mymJAYJZ1d3lfZDKfr+J4/6o/cCRluNiLGy8G+yTHqYHMWpIgosL4tOIqa5i4AwNSMWMS6x+S0gDvtERHRcLAgRbo1Z1xyr+vuumAMFkxJDcBqgsfmIuZH+ZM8ttdhd3qMoVH/OrqdqGiUzsSn2qxe3wUy2aNDigdYRBRY724rVS6fp4Ewc7Ukjw4pPl8SEdHgsCBFulVY29LrOnl3LBqajm4ndpZJ+VHJ0RbER2rnLGywyuLY3qCV1Pkm0FwmZ0gBwBEWpIgogKqbOvHNgVoAQFyEGVPSbQFekSd2SBER0XDw6J1060BV74JUXjkP6Idj+5EG2J3S2NgEdkf5BXfaG7ziup5OspQY74fux0aYYTJIXVcc2SOiQPpgexnkae5zxiTCYNBGmLnMI0OqgQUpIiIaHBakSLcOVPcuSHGnsuHZWFSnXJ7A/Ci/yIxXF6T4+zsQ6p00vb3DHgAYBEEZ2ztS3w4Xs72IKABcLhHvuXfXEwDMHqOtcT0AiLGaEObuTmeHFBERDRYLUqRbB6p7uknS3AellU2dqGvtCtSSdG9TMfOj/C3aalZGI/dWNrH4MQDqrC1fjOwBPWMo3Q4Xalo6ffI1iIhOZGPRMWXnuklpMR7dSFohCIKyOUd5fQdfw4iIaFBYkCJdEkVR6ZCKsZowZURPpgJzeIamtcuhjDym2aywhZsDvKLQIY/ttXU7UVzHEbGTkX9GAjzznrxJ/biH63jWn4j8Tx1mfu447XVHyeRg824nC/hERDQ4LEiRLh1t7UJ9WzcAYERcOLLVY0/MkRqSbYfr4XQxPyoQslS/v3sr+ft7IqIoosQ9spcQFaaMinibeqc95kgRkb/Vt3Vj9d4aAEC01YRpGbGBXdAJeASbcyMIIiIaBBakSJcKVPlRGXERyEqMVD5mh9TQbCrqGddjfpR/ZSWofn9ZUD2ho61daOlyAABSfDSuB3geYB1hLgoR+dmHO8rR7XQBAM4elQCTUbtv2dWjhMyRIiKiwdDuqxvRCXgUpGLDkRxtgdUs/TqzIDU06oLUuJSoAK4k9KiDzfn7e2IegeY+LEglx7BDiogCY9WeKjyeW6B8HOfOGdQqdQG/rKEjgCshIiK9YUGKdGl/VU9BakRcOAyCoBzUVzHYfNCaOuzKqFhGXDiircyP8idbuBlxEdLPfF9lM0NhT8BfBamESAuMgrS9OjOkiMhfcvOrcOdbO5TuKAB4d1sZth9pCOCqTkzdIVXGDikiIhoEFqRIlwpqpB32BAFIt4UDALITOLY3VFtL6iHXQJgfFRhZ8dLvb0uXgyNiJ+Cxw57NdwUpo0FAgnvnqCPH2iCKLBISke8t/6qw13UCgE/zKv2/mAGSd9kDOLJHRESDw4IU6Y7D6cLBGumgNCXaqoQayzuVAczhGayNRXXKZeZHBYbH7y8Lqv1S70LoywwpoGcMpa3biWPuTRSIiHyppI+dVkUA1U3a3b3OYjIqO/OyIEVERIPBghTpzuFj7eh2SK3sGXHhyvVZ7JAaMjk/ShCYHxUomaqC1F7+/vZL7pAKMxmUMUdf4U57RORv2apNWmQCfNsR6g1yl9TRli50dDsDvBoiItILFqRIdw5UNyuXR6gKUsnRFoSbjQCAfB7QD1h9WzcOuEPiM+MjEBFmCvCKQlMWg81PqtvhUgJzU6ItENwZT76SHN1zAMgcKSLyh6tOG+HxsQCpQ+qKU9MDsp6B8gw25/MlERENDAtSpDvH77AnOz7Y/GgLg80HYktxz+56zI8KnNiIMGXkIb+iiZlFfSitb4PTHXbmj24B9QEWc72IyB+iVJuKGATpxNuvzh+N6ZlxAVzVySUx2JyIiIaABSnSnQPqglRchMfn1Dk87JIamI1F6oIU86MCSf79be50oKyeW2cfr8hPO+zJOLJHRP6WV96oXL73ovF46PLJmi9GAZ477TFHioiIBooFKdIdeWTPYjIou2DJuNPe4G1yd0gZBGBsMvOjAoljeydW7OeCVFK0BfJQ4OFjPMAiIt/b7d6URRCgdH3rgbqjlAUpIiIaKBakSFdau3o6R0bEhsNwXIaMukMqjzvtnVRtSycO1Uoh0dkJkbC6M7goMBjMf2JyoDkApPhhZM9sNCAuQip6s0OKiHyto9uJgzVSF3i6LRwWHb0mJ3Jkj4iIhmBIBSm73Y6ysjIUFBSgvr7e22si6pdHfpQq0FyWxGDzQdlc3PP3OyGN+VGBpi6o7q3k7+/xiuv82yEFAMkx0kFWY7sdTe12v3xNIgpN+6qalJy8nD5229Oy2AgzTAbpJCE7pIiIaKAGXJBqaWnBiy++iDlz5iAmJgbZ2dmYOHEikpKSkJWVhdtuuw3btm3z5VqJjitI9W5lNwiCclBf3dyJ2pZOv61NjzYV1SmXJ6QwPyrQYsPNiLZKuxzuYbB5L3KHVGy42W/dfB45UvXskiIi39ld1nMiIjtBP+N6gPT+S+6SKq1v5+sXERENyIAKUk8//TSys7PxyiuvYN68efj444+xa9cuHDx4EJs2bcKf//xnOBwOXHTRRViwYAEKCwt9vW4KUXJ+FCCN7PVFncPDLqkT2+QONDcaBIxO1tfZ2GAkqAqqje12lDcw2FzW0NaNBneHkj922JOpc1GYI0VEvqQONM/WWYcUACRGSyPOnXYXjrZyp2MiIjo500ButG3bNqxfvx6TJ0/u8/NnnnkmbrnlFqxYsQKvvPIKNmzYgLFjx3p1oUSA5w57I/oY2QOOy+Epb8YFE1J8vi49qmrqUA6wRyVGwmLST1ZFMMuOj0R+hVR4za9owkgdhdr6UnGdKj/KT+N6AJAc3fO1SpkjRUQ+JGdfmgwCMvo56aZlSR45Uh0ez59ERER9GVBB6p133hnQg1ksFixevHhYCyLqjyiKOFAlHajHRZgRZen711edw8Ng6P7J3VEAMCGV+VFakan6/c2vbMLFp6QFcDXaUeTnHfZkyeyQIiI/aOqwKzl5I+MjYDLqb9+h44PNZ2TFBXA1RESkBwMqSBFpQXVzJ5o7HQD6744CpAPIcLMRHXYn9lQ0+ml1+rPRoyDF/CitUI+c7qloPsEtQ0uxuiAVoJG9YNhpLze/Csu/KkRJXRtyEiNxz7yxWDCFRU+iQFNHDOgtP0qmfr5ksDkREQ3EoE6/1NbWory8XPnY4XDggQcewJw5c/Cb3/wG7e188SHfUY/rZcT2/2ZNncNT09zFYPN+yB1SZqOAUUn6y6oIVvGRYUr3Xz6DzRVyoDng3w4pq9mIGHfQ/BGdd0jl5ldh8Zs7UFDdgi6HCwXVLVj85g7k5lcFemlEIW+3zvOjAM+RPRakiIhoIAZVkLrtttvw2muvKR8/8cQTWLlyJc444wx88sknWLp0qdcXSCQ7UHXy/CiZemyPwea9ldW3o6JRCswenRQFsw5HA4KVIAhKl1R9WzeqmlhQBaCMspgMAhIiw/z6teUclNqWLrR3O/z6tb1p+VeFEADIJU4RgCAAz37NjUiIAi1PtcNeToI+C1KJLEgREdEgDeooNC8vD3PnzlU+fuONN/Dcc8/hySefxLvvvotPP/3U6wskkhWodtjLOFlBKr7nzZwcEko9mB+lbcxB8+RwupRxueQYCwwGwa9fPzlGPban34Oskro2HN9vJ4qe45BEFBjyDnsWk8GvXaDeFB5mVDp8y1iQIiKiARhQhtTNN98MAKisrMTTTz+NlStXoru7GwUFBfjoo4/w5ZdfwuVyoba2FrfccgsA4F//+pfvVk0hSR7ZMwoC0k7yZo0dUie2sahOucz8KO1R7xSZX9GE+ZNTA7iawCtv6IDdKZVSAnGgdnyO1MQ0ff7N5CRGeow+yziySxRYR1u6UOnuhs1KiPB70d2bkqItaO1yoLq5E10OJ3fwJSKiExpQQeqVV14BAKxfvx633norLr74Yrz33nvYs2cP3n33XQDAsWPH8Mknn7AQRT7R7XChyJ0hk2qznnT3Gc9gcxak1ERRxKZiqUPKYjIgO1Gf4anBLDOeBVW14rrA5EfJkqODo0PqnnljsfjNHb2uv/vCcQFYDQEMmSdJnio/Sq/jerLEqDCpG1MEKho6MCopKtBLIiIiDRvUyN6ll16KW265BXfccQf+53/+BzfddJPyua1bt2LSpEleXyARIB2Qyh0SI2JPPK4H9BFs3swcHllJXRtqmrsAAGOSo2AyMD9KaxKjwhARJp1V3lPRHPLB5uqRshQ/7rAnkzOkAOCwjgtS8yenIjbC7HGdAOCMbG7NHggMmSfZblW0gF4DzWUMNiciosEY1JHo448/jl/+8peorKzE0qVLPULMt2zZgsWLF3t9gUQAUKDeYe8k+VEy5vD0Te6OApgfpVXqgmpda5dSQAxVRaqCVOA7pPSbt1Ra347GdrvHdSKAr/bXBGZBIY4h8yRTd0hl671DSvV8yRwpIiI6mQGN7MmsViseeeSRPj/30EMPeWM9RH06MISClPpN3Z6KJlw4McXr69KjjR6B5vrMwgkFWfGR2O/eWTK/ogmpAegM0orio4Ed2Yu0mBARZkR7t1PXI3vqzQymjYzFrrJGAEBufjWuOSMzQKsKXQyZJ0Aao5c3X4mymJAY5d9dRL2NHVJERDQYnNUhXThQ1bPD3kBG9oDjOqS40x4A6Y3vFneHVLjZ6JFVRNrCDr8exXXSAXq01YRIy6DOo3iN3CVV2dSBLoczIGsYLnV35PzJKYhzj+99f+gYWjrt/d2NfCSnj9EsAQyZDzXlDR2ob+sGAGQnREAQ9BtoDnhuAsGCFBERncyAClILFizA5s2bT3q7lpYWPPbYY3jhhReGvbDjLVq0yOuPSfohj+yFm42IjxzY2cOkKIsqhye0D+hlhbWtqGuV3viOTYmCUcc7+QQ77hQpaem042iLNLIYyK3Q5RwpUQTK6jsCto6hEkURm1WbGeQkRuK0TCk7qtvpwtqCo4FcXki6Z97YXteJYMh8qMkLovwoAIiLCIPRXVTT43MlERH514AKUldffTWuuuoqTJo0Cb/73e/w/vvv4/vvv8f27dvx1Vdf4bnnnsPPfvYzpKWlYceOHbj88st9vW7F/v37ccUVV8BmsyEyMhJnnHEGSktLT3if999/HxMmTIDVasUpp5yCL774wuPzoijiwQcfRFpaGsLDwzFv3jwUFnpmOtTX1+O6665DTEwMYmNjceutt6K1tRXkfU3tdmU75Iy48AGfPRQEAVnuDqDali7UMNjcY2SH+VHalhQl7RQJAPmVoVuQKqlTBZoHtCClPuuvv5GqvjYzmJ4Zq3z+y/zqAK0sdI1J7r372GmZsVgwJTUAq6FACab8KAAwGgTlxGFZfXvIb8pBREQnNqCC1K233ori4mL84Q9/wL59+3D77bfj3HPPxRlnnIH58+dj5cqVyMzMxLZt2/Dee+8hM9M7WRR1dXW46aabkJmZiXfeeQdjxozB1Vdfje5uqcOjqKgIs2fPxoQJE/Dtt98iLy8Pf/rTn2C19n/QsnHjRvz85z/Hrbfeip07d2LhwoVYuHAh8vPzlds8/vjjeO6557BixQps2bIFkZGRmD9/Pjo7ewoa1113Hfbu3Ys1a9bgs88+w/r163H77bd75fsmTwU1g8+PkmWpc6Q4toeNRXXKZeZHaVuvnSJbQrOgWhzgQHNZUkxPQepwnf7GUDYX1yuXx6dIxeixydGIco9Ari2oRaddn6OIevXFnt5FwOqm0Pw7D2W7VQWpvsY49SgxWipItXQ5em2kQEREpDbgDCmLxYLrr78en376KRoaGtDQ0IDKykp0dnZiz549ePLJJzFx4kSvLm7p0qXYvHkz3njjDVxyySVYuXIlRo0aBZfLBQD44x//iEsuuQSPP/44TjvtNIwePRpXXHEFkpOT+33MZ599FgsWLMC9996LiRMn4pFHHsH06dPxz3/+E4DUHbV8+XI88MAD+PGPf4ypU6fi9ddfR2VlJT7++GMAUldWbm4uXn75ZcycOROzZ8/G888/j3fffReVlZVe/RkQcKB68PlRMubw9HC5RGwpkQ5KI8OMgy7ukf9lxXNszyPQPIDB7nrfaa+v3TWNBgHTRsYCANq7nfiusK6vu5KPfLGnSrks/25XNXWiqoljTqHC5RKRXyG9x4mLMMMWbg7wiryDweZERDRQQw41t9lsSE1NhdnsuxfPnTt34sYbb8ScOXNgs9kwd+5cPPbYY7BarXC5XPj8888xbtw4zJ8/H8nJyZg5c6ZSNOrPpk2bMG/ePI/r5s+fj02bNgEASkpKUF1d7XEbm82GmTNnKrfZtGkTYmNjcfrppyu3mTdvHgwGA7Zs2dLv1+7q6kJzc7PHPzo5zx32BhfCrW5/D9UDetn+6mblTOW41GgYdB6cGgo8O/xC8/miqE4bHVJyhhQAHNbZTnuiKCrjuhaTAZmqQv1pqrG93L0c2/OX4qOtymvbqMRIj/HJXaWNgVkU+V1xXStauxwAgiM/SsZgcyIiGihN77J3zjnn4JVXXsFnn33W63O1tbVobW3FP/7xDyxYsACrV6/GlVdeiZ/85CdYt25dv49ZXV2NlJQUj+tSUlJQXV2tfF6+7kS3Ob4Ly2QyIT4+XrlNX/7+97/DZrMp/0aOHHmC755kBaqC1GA7pBKjwpRg87wQL0ip86MmclxPF9SFg1DNkZJH9oyCoIyBBEKM1QSLSXrJ1NsBVtHRNtS1SvlRY1Ok/CjZpLQY5fv6an8NHE5XQNYYalapMrtmZMVhVGJPntTOssYArIgCYXdZz/N6ThDkR8nYIUVERAOl6YLU008/jWuuuQZLly7F66+/jmnTpmHFihUAoIzt/fjHP8bSpUsxbdo0/P73v8dll12m3EZr7r//fjQ1NSn/ysrKAr0kzRNFUSlIJUSGIdxdXBoodbD50RAPNlcXpMYz0FwXkqMtsJqlp+lQ7PBzuUSU1Ekje4nRYR6FFH8TBEEZ2yurb9dV4cZjXC/FsxhtNhowNcMGAGhst2NrST3I91bl94zrSQWpnmLEztKGQCyJAiDYAs1liaoOqfIGFqSIiKh/mi5IRUZG4tFHH0VhYSGuuOIK3HnnnVi2bBleeuklJCYmwmQyYdKkSR73mThx4gl32UtNTUVNTY3HdTU1NUhNTVU+L193otvU1tZ6fN7hcKC+vl65TV8sFgtiYmI8/tGJlTd0KO3sQ808YrA54HC6lAPNaKsJ6QHM4qGBMwgCMt0F1aqmTqXLJVRUNXei0y4VfgI5rieTx/YcLhGVjfopbm8uPnEx+rSRccplju35XumxdiU3KDshAolRFsSEm5EYJXUA7qlogl1HBU8aut2q9yTZiYOLJNCyRHZIERHRAGm6IKUWGxuLO+64AxdffDE2bNiAsLAwnHHGGSgoKPC43cGDB5GVldXv48yaNQtff/21x3Vr1qzBrFmzAAA5OTlITU31uE1zczO2bNmi3GbWrFlobGzE9u3bldt88803cLlcmDlz5rC/V+qhzo8aMcSCVDaDzbG3shkt7sLe+JRoCMyP0o2s+NDNQfMINNdAQUqdi3KkXh/B5qIoYou7IBVuNioFTrWpGTaYDNJzwpd7q+FycZt2X/riuO4omTy212l3eYyqU3Dqdriwr0oqTKZEWxARZgrwirwnMsyIcLPU0c6CFBERnciQClKNjY14+eWXcf/996O+Xuq62LFjByoqKry6uKVLl2LdunVoamqC0+nE2rVrsW7dOsyYMQMAcO+99+K9997DypUrcejQIfzzn//Ep59+il/96lf9Pubdd9+N3NxcPPXUUzhw4AAeeugh/PDDD1iyZAkAaSzjnnvuwV//+ld88skn2LNnD2688Uakp6dj4cKFAKQurAULFuC2227D1q1b8f3332PJkiW49tprkZ6e7tWfQagrUO2wlxE7tLOHHh1SIXZAL1OP7ExMY2eenqh3igy9glRP0SdFA1196p329BJsfqi2FXWt3QCAsclRMBp6F6OtZiMmpUvPCzXNXR7b0JP3rdrTT0EqiWN7oeRgTQu6HVInXDAFmgPSe2m5gF/Z2MmOPyIi6tegC1J5eXkYN24cHnvsMTz55JNobGwEAHz44Ye4//77vbq4zMxMLFu2DCNHjsTbb7+NG2+8EbfccgvuuusuAMCVV16JFStW4PHHH8cpp5yCl19+Gf/5z38we/Zs5TEWLVqE888/X/n47LPPxttvv42XXnoJp556Kj744AN8/PHHmDJlinKb++67D3fddRduv/12nHHGGWhtbUVubi6s1p4DorfeegsTJkzAhRdeiEsuuQSzZ8/GSy+95NXvn4D9HjvsDa1DSh1sHqoFqY3Mj9KtrBDu8NNah1RyjKpDqk4fHVKbTjKuJ5vOsT2/KG9oV8a0MuMjPHZv9CxINfp7aeRnu4M0P0omB5s7XSKqdDTiTERE/jXo/uBly5Zh0aJFePzxxxEd3fPm9pJLLsEvfvELry5u6dKlWLp0KQCpsPTqq6/2us0tt9yCW265pd/HKCkpwdy5cz2uu/rqq3H11Vf3ex9BEPDwww/j4Ycf7vc28fHxePvtt0/yHdBwyWMLJoOAlCEekAqCgKyECOyvalGCzYf6WHpkd7rww2GpkzE23IwUVZcHaV9KtBUWkwFdDpeSOxMqilVFHy38zap3jtJLh9RANzM4daQNwmZAFIEv86vx+wUTONrrA7nH7a6nNjIuAiaDAIdLxC7utBf08sqCMz9Kpt4VtbS+3WPXWCIiItmgO6S2bduGO+64o9f1I0aMQHW1ts6qNjU1oaioCL/97W8DvRQagk67EyXuA9I0m7XPUZOBUufw5IVYsHleeSPau50ApANSHmTqi8HQE2xe0diBhrbuAK/If+SRvXCzETHWwOerxEWGKVlLpTrIkHK5RGxxb2YQEWZEZlz/B4TRVjPGp0gFq8PH2nGwprXf29LQfdHPuB4g7Xgo/60X17WF1N96KJI7pAwC+sx20zt1Ab+MO+0REVE/Bl2QslgsaG7ufZb+4MGDSEpK8sqi+tJXd9TJ2Gw2lJeXIyoqyvsLIp87VNsKpztcN+MEB1IDEcrB5uoOiYmpzI/So1Ac2+vodqKisQMAkGqzaqKQalDlohw51q758O+DtS2odxc1xiVHw3CSov5pI2OVy+pOHvKOysYO7HCP4o2IDe9zDFU9treLWV5Bq6PbicJaqeibHhsOi8kY4BV5H3faIyKigRh0QeqKK67Aww8/DLvdDkAahyotLcXvfvc7XHXVVV5fIIWuAi/kR8nUweahFgzN/Cj9U589z68Mjd/fEtW4nhbyo2RysHmXw4WaFm3nomxW/e2PSz35iZnTMpkj5UvqIt/px3VHyeSd9gDmSAWzvZVNygm3nCDMjwI8dyVlQYqIiPoz6ILUU089hdbWViQnJ6OjowNz5szBmDFjEB0djUcffdQXa6QQdUC1w96I2OEVpNTB5nnlTRBFbXc2eEuXw4ntR6TdmhIiwzzeIJJ+hGJBtbiuZ2QsJUY7v7fqv6EjGs+RUgeaTxhAd2R8ZBhy3Lt97a9qRqnGvz+9WZXf/7iebDR32gsJu8vV+VHBWZBKiAyD3JNZxoIUERH1Y9ChHDabDWvWrMF3332HvLw8tLa2Yvr06Zg3b54v1kch7IAXO6TUweZ1rV2oae5Cqga2kfe1naWN6HJvK83uKP1Ki7EizGRAt8MVMiN7cn4UAE39rap3RTtyrA1njUoI4Gr6d3x+1ECfQ08bGat0p325txq3nTfKZ2sMJTXNnfjBfXIg3WZFej8nWeIjwxBjNaG504HdZY1wucSTjlqS/uSpxjGDtUPKZDQgLjIM9W3d7JAiIqJ+DbpDSjZ79mz86le/wn333cdiFPmEXJCKsphgCzcP+/HU2yqHykE986OCg8EgYKS7oFBW34GmdnuAV+R7xUd7OqS0OLIHaHunvQPVLWh0/56MT4mGYYAZXNOzOLbnC1/urYbcmNtfdxQgnTwZlSSN7TV3Ojx2mqTgIW+uYjYKSI/TzvObt8nB5o3tdjR3Bv/rFhERDd6gO6QefvjhE37+wQcfHPJiiGTHWrtwtKULgNQd5Y1AY49g6PJG/GhSyrAfU+sGuuU7aV9WfCSK3F1D+ZVNOGdMYoBX5FvygbgAz66kQPPIRdFwQUo9rjeYv/3UGCvSbVZUNnViR2kDaps7kayhgqBenWh3veONSozErrJGANLY3phkbswSTJra7UoX4si4CJgMQz43rHlJ0RYU1EgnF8vq2zE53RbgFRERkdYMuiD10UcfeXxst9tRUlICk8mE0aNHsyBFXqEONB9ufpQsKz60OqQ6up3YWSaNiCRHWxAfGRbgFdFwHL/TXjAXpERRVEb2EqLCEGbSzgFbQlQYDALgEoHDx7TbvbJ5iAUpAJieGYfKPVUQRWD1vhpcf1aWt5cXUo62dGGre3wyJcZy0tc09U57O8sacfXpI326PvKvvIpG5XKw5kfJEqN63newIEVERH0ZdEFq586dva5rbm7GokWLcOWVV3plUUTezI+SJUaFITLMiLZuJ/ZUNEMURU1sJe8r2480wO6UZkQmsDtK944vSAWzoy1daO1yAABSNNadYzIYkBBlwdGWLhw51q7J5xGnS8QWd0EqymIadFF/emYcPnN39Hy5t5oFqWH6cm81XKpxvZP9vmQnREIQAFEEdnGnvaCTpwo0D9b8KJk8sgdwpz0iIuqbV047x8TE4C9/+Qv+9Kc/eePhiDw7pLxUkJKCzaU3f3WtXahu1vaW7cO1qbhOuTyQHbZI29Js4TAbpQPZvUFekCpSB5prrCAFAMnug6zWLgeOtXUHeDW97a9qRnOnVNAblxI14Pwo2cj4cCS4Oyo3FR0LicwyX1Lvrnd6ZvxJb281G5Ui4oHqZrR3O3y2NvK/3e5xTADITozo/4ZBIDGaBSkiIjoxr81BNDU1oakpuA+SyH8OVDcDkPJjRti8U5ACjs+RCu7f143MjwoqRoOAkXHS7+/hY+1BHRBbXKfNQHNZckzPQdYRDeZIqcf1hlKMFgQB0zOlnCOHS8TXB2q8trZQc6y1C5uLpXG9pGgLRsYP7PVslHuUyyV6dtSQ/sn/P8PNRs11gHqbZ4dURwBXQkREWjXokb3nnnvO42NRFFFVVYU33ngDF198sdcWRqHL6RJxsEY6IE2KtsBiNnrtsdU77eVXNOGiyalee2wtae1yKG9602xWr+xSSIGXlRChhH3nVzTh7NHBmSNVrO6QsmnvgE0dbH7kWNtJQ6r9zRubGUzPjMWa/VIh6su91fjJ9AyvrC3UrNlXA6d7Xm9G5snH9WSjkqKwvlDqct1Z2oizRiX4bI3kP7XNnUp3dlZCxKC7F/Um2mqCxWRAl8OFMnZIERFRHwZdkHrmmWc8PjYYDEhKSsJNN92E+++/32sLo9BVWt+ODrsTgPfG9WTqDqm8IB572na4XjkIYn5U8MiM7/n93VvRHMQFqZ4OKS12EKh3/TussQ4pp0tUArSjrSakD7GgNzopCtFWE1o6HVh38Cjaux2ICBv0W4aQ90V+tXL59EEULkepwq53uTenIP3brep2yw7y/ChA6rZMjLKgorED5Q3tcLpEGA3BXYQjIqLBGfS7y5KSEl+sg0hR4B7XA4AML+2wJ0uI7Ak2z69o0mQgsTdsLhreyA5pU1ZCaOwUKXeBWUwGxEVor7svWZ2LorGd9vZWNqHFHQg/PiV6yM9vBoOA00bGYn1hHTrtLqw/eBQLpqR5c6lBr7G9GxsPSV1OCZFhHidETibVZkW42YgOuxM7ShuD9rUq1OSVNyqXgz0/SpbkLkjZnSJqmjuR7uX3dUREpG/a2UubyG1/lXqHPe++YRMEQTkrWdfaHbTB5ur8qHEpUQFcCXlTeqwVJvfZ5fwgLUh1OZzKaEdKjFWTB+FJ0RbIq9Jah5Q6P2p8yvC6I+UcKQD4ci9zpAZr9b4aOORxvQHsrqdmEATkuLukjrZ0obIpOF+rQs3uENphT5YYHaZcZrA5EREdb9AFqba2NvzpT3/C2WefjTFjxmDUqFEe/4iGS73DXoaXR/aA48b2gjAstqnDjr2V0veVEReOaKv2OkxoaEwGg/I3UVzXhpYgDDYvq2+H+xhek4HmAGA2GhAXIR1kHdFYh5Q38qNkE1KjEe7O8Ptqfw26Ha5hPV6oWbWnZ3e9oeSMjUrqKVjsLOXYnt6Joqh0SEVbTYiPDDvxHYKEZ7A5C1JERORp0CN7v/zlL7Fu3TrccMMNSEtL0+TZa9K3ghqpIBVmNHi8kfGWrOOCzecHWbD51pJ65YCe+VHBJyshUunK2VfZjJlBFnZcpAo0T4nx/t+/tyRFW1Df3o2GdjuaOuya2DjA4XRh22GpcBFjNSFtmIHwJqMBUzNs2FJSj5ZOBzYVH8OccUneWGrQa+qw4zv3uF58RJhHJtRAqe+zs7QRl01N99r6yP/K6jvQ2C6dRMhOiAyZ98/qTSAYbE5ERMcbdEFq1apV+Pzzz3HOOef4Yj0U4tq7HTjs7jhIj7XC4IPwy2xVh1Qw5vBsYn5UUMuK9/z9DbaClNZ32JMlR1uU4nnpsXackmEL8IqA/MpmtMr5UalDz49Sm54Zhy3ukPQv91azIDVAX++vgd0pnRmYnhU7pP8XOR7B5o3eWhoFyG51ftQg8sT0LpEdUkREdAKDHtmLi4tDfHy8L9ZChMKaVoju7h5v50fJ4iPDEGWRarF7yqVg82CysUg6Ky8IzI8KRpmqA5m9lc0nuKU+qXfY0+rIHgAkq7q3DmtkbM8Xxegp6TEwG6Viyuq9NcrunXRiX+zp2V1vKON6ABBtNSsB+nsqmjgyqXPqQPOcIXTM6RULUkREdCKDLkg98sgjePDBB9HezhcV8r4Dqh32RvhoJxZBEJQuk2Nt3agKorDY+rZuHHBncGXGR3Cb9iA0IjZc2TY7GDv85B32ACnUXKvUYyhayZHyZqC5zGI2Ykq61P1V19qFHcwyOqmWTjvWFx4FANjCzRidNPQTA3KOVLfDhf1VwVeADiXqQPPsEAk0B4AwkwGx7pHmsvqOAK+GKLjk5ldhwfL1GP/AKixYvh65+VUnvxORxgy6IPXUU0/hyy+/REpKCk455RRMnz7d4x/RcBzwcaC5LCsxOMf2tqgOSCd46YCUtMVsNCjF2qKjrWhzj2gFC7lDKi7CDKs7UFuLkqN7imVHNLDTnt3pwrbD0midLdzs1fwtj9328qtPcEsCgG8O1CrdTDMy42AYxujkqMSeYhaDzfXL6RKVnVHjI8MQo4HMOX+Su6TqWrvQ3h1cr1lEgZKbX4XFb+5AQXULuhwuFFS3YPGbO1iUIt0ZdPvEwoULfbAMIsmBKj8VpOJ7zk7uKQ+eYPNN6oJUGvOjglVWfARK69shisD+qmacnh0cY9QNbVJIOKDt7igAyigVoI2CVF55E9q7nQCk7ihvBiZPzbDBKAhwiiJy91bjj5dODJlA5qH4Ypi766mpd9pjjpR+FR1tVf4+c0KoO0qWFG3BIffJhrL6jmHvAEpEwPKvCgEA8iC9CCmu49mvC7FgSlrA1kU0WIMuSP35z3/2xTqIIIqiEhJsCzcj2uq7M4jBGmy+0Z0hYxCAscnMjwpWWQkR2HBIurynoiloClLFdfrIjwIAq9mIGKsJzZ0OTWRIqcf1vL27ZqTFhPGp0dhX1Yzyhg7srWzGlBGBD3HXorYuB74tkMb1YqymYT8PZ8SFw2wUYHeK2MmClG7tVv2/y04MnUBzWWJUmHK5tL6dBSkiL1BHHMhE0XNzGCI9GPTIHgA0Njbi5Zdfxv3334/6emlEYMeOHaioqPDq4ii0HG3tQn1bNwDf5UfJ1MHm+RXBEWz+3rZSHKqVDuhNBiEoA69JkqU6wx5MBdUineywJ5NzpGpbAj+G4pEf5YODvemZscrl1Xs5tteftQW16HKP603PjBv2TrEmg0Hp6D1yrB3HWruGvUbyv7wQzY+SqTP3GGxO5B0p0b1H8wXBs7OWSA8GXZDKy8vDuHHj8Nhjj+HJJ59EY2MjAODDDz/E/fff7+31UQhRj+uN8OG4HuAONk/oCTav1HmweW5+FX73nz3Kx91OES+uK8L2I8wcCUYZceEwukem8oOoIKU+q6f1DinAM0cqkAdZ3Q4Xfjgs/a3HRZg9xgm9ZdrIWMillVwWpPq1ygu76x2PY3v6p95hLysh9DqkklQ77ZWxIEXkFX1Fc4gicPeF4wKwGqKhG3RBatmyZVi0aBEKCwthtfa8Gb/kkkuwfv16ry6OQkuBnwLNZeo3hXvK9X1QL8+RqwkAPs2r9P9iyOfMRgPSY6Xn30O1rehwZ5PonRxoDmg/QwoAklXB4YfrAneQlVfeiA67Oz8q1bv5UbLYiDClMHKwptXj/xVJOrqd+OZALQAgymLCOC9tLKEuSO0sbfTKY5L/SDskSu9vUmOsIbn7rbpDigUpIu+o7uNk+vTMWCyYEhy5uBQ6Bl2Q2rZtG+64445e148YMQLV1TxrSkO3v7pnxCzDxyN7gGfbvN67TPqcI0ffL1YUHOSxPZcI7AuS7eDl32OTQUBCZNhJbh14yVHqYPPAZTZsKlKN6/lwd02P3fb21vjs6+jVtwW1SmFwemYsjMMc15Opd9pjh5T+HKhuRrdTGuMMxfwoAIgJN8Pk/nvgyB7R8DW125FfKR27pNqsiLFKhe6dZY0o4gkj0plBF6QsFguam3sf/Bw8eBBJSUleWRSFJrlDShCAdD8UpLLie94Y5um8IKUODJUJ0EcODw2N+vdX7wVVAHA4XUpRJyXGOuzsHX9IUnVIHQngQdbmEnWgue9211QXpDi219sX+d4f1wOkMczYcGmTj11ljXC69J95GEp2h3h+FAAYBAGJ7i4paYdY/g4TDcem4mOQ/4xOSbfhR5NSAEgjey+tKw7gyogGb9AFqSuuuAIPP/ww7HZpa25BEFBaWorf/e53uOqqq7y+QAoNDqcLhe5A7pQYK8zGIeXtD0owBZsnHtdNIkDqkLri1PSArId8LzMhuApS5Q0dsDulv0E95EcBnhlSgeqQ6nI4lfyo+MiwPovT3pIUbcFI9zj17rJGVDV1+Oxr6U2n3Ylv9ktdY5FhRq8GywuCoIzttXY5ePZbZ/JUXW05iaFZkAJ6cqS6HC4cbWE4P9FwbCyqUy5PTIvG+eOSEW42AgA+3FnO12fSlUEf9T/11FNobW1FcnIyOjo6MGfOHIwZMwbR0dF49NFHfbFGCgGHj7Wh270zkT/G9QDpTX62+6C+XsfB5sdau7DPnU9hFKRxpxFx4fjV+aM9OhoouIyMi4DcRBQMO+0V16nzo7wfyu0LURYTIsKkN4CBypDaXdak7Oo2PsU3+VFq6ueU1RzbU6w/eBRt7iy30zLjYDJ496SKemxvZyk3q9ATeYc9oyBgZFxojuwBnsHmHNsjGp7vD0kFKYMAjE2ORniYEXPHS5NKdqeI/9tQEsjlEQ3KoN8x2Ww2rFmzBp9++imee+45LFmyBF988QXWrVuHyMjQPfNDw7O/yr+B5rIsVfv8HtUuOHryWV4VHO4Rjh9NSsWK62fgocsnsxgV5MJMBqTZpL+VwtpWdNr1HWyu3mEvRUejpvKOdlVNHehy+P//gTo/aoIXu3L6c1pmrHI5N59je7JVPhrXk3GnPX1q73agsFZ6f5Mea0WYyffd31qVGN3TvcmCFNHQVTd1osj9niknMRLh7hNjF05MgdkonZR6e2spGtq6A7ZGosEY9CtjWVkZAGD27Nn41a9+hfvuuw/z5s3z+sIotKh32Bvhpw4p4Lid9nTaZfLhzgrl8qxRCQFcCfmb/PvrdIk4oPob0qMiVUFKLyN7QM/YnkuUxg79bXOxKtDcDwWpEbHhShFuS8kx1PMNL7ocTny1T+oWCzcbMdEH/x+yEno6IrnTnn7kVzRDjvwK5XE9wLNDqqye40REQ7WpWDWup8qNtIWbMXtMIgCgvduJ1zcd8fvaiIZi0AWp7OxszJkzBytXrkRDA9vGyTvUB9MZfmxpVweM7qnQ305lxUdbsdt9tnxkXDhG+LG7jAJPHWyu14KqrFiVi6OvglTgdtrrtDux3T2+lRgVhsQo3486CoKgdF+6ROCr/Rzb+66wDi1dDgDAtJGxMPkgA9FiMiqvjQU1LWh1fz3StjxV53WoBprLEqM5skfkDd8fUnVGp3meALloUqpy8uLVjSVo7+ZrBWnfoN81/fDDDzjzzDPx8MMPIy0tDQsXLsQHH3yAri4GFNLQHaiWikEWkwEJPgzlPV5chBnR7q1S95Q36i7Y/GNVd9RZ7I4KOeqR0/xynRek6qRiTrTVhEj3ZgN6kORRkPLvQdbO0kYle298iu+7o2Tqsb0vObaHL/b4dlxPNsrdYSOKnkHZpF0eO+yxQ0q5XMaCFNGQiKKIje78KLNRwOikKI/PJ0VbcEZ2PACgod2Od7eW+X2NRIM16ILUaaedhieeeAKlpaVYtWoVkpKScPvttyMlJQW33HKLL9ZIQa6l066MumTEhcPg41BeNUEQlC6ThnY7Khr100YuiiI+2iUVpAQBmJkTH+AVkb+NjAuH/OeSX6nfglRLp13ZdUlP3VHA8R1S/j3I8ve4niwnMRKx4WYAwIZDdSHdrdPtcGHNPqkgZTUbMDk95iT3GDp1jtROFqR0Qe6QMhsFpMfq67nN26xmo3ICkB1SRENz+Fi7sgnTmOSoPnclv3hKqnJ55YZi5cQVkVYNua9cEATMnTsXK1euxFdffYWcnBy89tpr3lwbhYiDNYHJj5J5dJnoaOxp+5EGJYdhUmoMYiP811lG2mAxG5HmLuAcrGkJSKi2N3gEmuutIKVa72E/j+xtUhek/NghZRAEpUuq2+HCtwW1fvvaWrOxqA7NnVJB7tSM2D4PDrzFc6e9Rp99HfKOxvZupUidGR/h9Z0X9Ujukqpu7tT9RhxEgbCxqO/8KLWMuAhMzbABAKqaOvHfXRV93o5IK4b86lheXo7HH38c06ZNw5lnnomoqCi88MIL3lwbhYhA5UfJsnUabP4hx/UIPQVVu1P02BxAT4rr9JkfBQAxVhMs7p2z/Nkh1Wl3Ype7KJEUZUGCH/Kj1NS7eIbybnurVON6p/twXA8AkmMsiHDvprSrrEF3I+ahJk89rhfi+VEydc5dIDaBINK7jar8qIlp/XfkXjIlTbm8Yl0RXC6+XpB2Dbog9b//+7+YM2cOsrOz8frrr+Oaa65BUVERNmzYgMWLF/tijRTkDlSpC1KB7ZDK00kOT5fDic/zqgAAYSaDR6YLhZZMVbB5vg6D+QHPDqlUm74KUoIgKDlS5Q3tcDj90xq/40gDut1fy5/jerKxKVGIdBdH1h6oDcluB7vThS/d43oWkwGT020+/XoGQVBypOpau3lAr3EegeYhnh8lU2fulTVwbI9oMFwuUemQiggzemxsc7wxyVEYmyx11RYdbcPqfdyAhLRr0AWpv/71r5g5cya2b9+O/Px83H///cjKyvLF2ihEqLs6AjGypw42z69o0sVZ57UHjqKpww4AmJ4ZC6vZGOAVUaBk6bTDT82jIKWzDimgJ0fK7hRR5c528LVNAcqPkpkMBpw6MhYA0Nbt9BgjCBWbi4+hsV16Hp6aYUOYyfcjWaNUAbY7SrnTsZapA81z2CEFQNoNVMZgc6LB2V/djAb3a874lGgYDCfO3L3klJ4uqRe/PaSL4xsKTYN+91RaWorHH38cp556qi/WQyFGFEVlh724CHNAdtcSBEE5qNdLsPlHO8uVy7M4rhfSMuMjIL8l0VMGmlrRUWlkzygISIzWXxaa+qy/v3Kk1IHmEwJQkAI4tuev3fXURqk6bXYx2FzT5A6pcLMRyTH+HanVKvVzZamfN4Eg0jv1uN5AXvenpMdgpHvyZHd5EzYVHTvJPYgCY9AFKUEQsGHDBlx//fWYNWsWKiqkHJs33ngD3333ndcXSMGtqqlTCYQdEYBxPVl2fM+b/D0aH9trbO/GNwekEGFbuBkT+gk1pNBgNRuVIPCC6hbd7abicolKEScxOkyXwb/J0T1dXf7IkerodirFiJRoC+ICtKHBpLQYJT/rq/21fhtX1AKH04XVe6WCVJjJgFNG+HZcT5ajKkgx2Fy7apo7UdMs7RyalRDh192DtSxJlSHFnfaIBscj0PwE+VEyQRBwsSpL6v99W+STdREN16Df+f/nP//B/PnzER4ejp07d6KrS3rBbWpqwt/+9jevL5CCm3pcLyPW/4HmMj2NPX2WVwW7U2q7PTMnHsaTtOxS8JN/f7udLo9dK/WgsqkDnXapkKHHcT2gZ2QPAI74oUNq+5EG5TkgEON6MnUhpr6tG9sOh84I2dbD9TjW1g0AOGWEDRaTf8amIy0m5e9kX2WzbnfWDHa7Vd1rDDTvERcRBqO7OMeCFNHA2Z0ubC2pByCdjE4bYN7mjKw4pRD83aE6j2w7Iq0YUobUihUrsHLlSpjNZuX6c845Bzt27PDq4ij47a/uCWEORKC5TB1srvWC1Meq3fU4rkeAZ0FVb2N7JXX6zo8CPAtSh/3QIbWpuOcsaSALUgA8NlT4cm/ojO2pd9ebkemfcT3ZqCTp9arb6cLeSn1uZBDs1Buk5DDQXGEwCEhw50iV1bcz04ZogHaXNaKtWzoBMSE1GsIAuy6NBgELpqQqH7/ILimf+mRXBS56Zh3GP7AKC5avR25+VaCXpAuDLkgVFBTgvPPO63W9zWZDY2OjN9ZEIcSjQyqABSl1sPkeDQeblx5rxw9HpC6E9FirMhtOoS0rXj8F1eOpA81TdLbDniwuIgwmd6eiPzqkNhfXK5fHpwS2IDV1RKzyvX+5t1qzz53e5HSJWOXOzDIbBUzN8M+4nmwUx/Y0b7d6h72EwHV/a5HcrdHW7US9u8uQiE7se1V+1MRBRnWcPToBtnCpiSR3bzUO1bZ6dW0kyc2vwv+8uwsHa1rR5XChoLoFi9/cwaLUAAy6IJWamopDhw71uv67777DqFGjvLIoCh0HqqSClFEQAtodoQ42b2y3a3Y77Y+O644a6BkSCm4j43sKk/k665goPtrzxkivHVIGg4BEd5dUaX07XC7fFWXauhzKOFBqjBWxAcqPkoWHGZUsi6qmTo/OkGD1w+F61LVKcQVT0m1+3+V0VGLPTnsMNtceURSVEwPRVhPiI/W3UYMvJao6Sss0+l6LSGu+98iPGtyJKLPRgHkTkwEAogi8tJ5dUr7wzJpCj49FAIIAPPt1Yd93IMWgC1K33XYb7r77bmzZsgWCIKCyshJvvfUWfvvb3+LOO+/0xRopSHU7XMruWqk2K0zGwIYZq3MetDj2JIoiPt4lFaQEADNzOK5HkogwE1Lcb/L3VzXDrqNw6eIgGNkDesb2Ou0u1LZ0+ezr/HCkAQ5X4POj1NRje7khMLa3Kt//u+upjYgLR5j79XJnaejkdulFaX07Gt1bs+ckRPLE0XESo3oKdMyRIjq5jm6n8lyfFG1BQtTgd+08f1wywt0nTz7aWYGqJhaDve3Q0d6dZ6LoOQlAfRt0BeD3v/89fvGLX+DCCy9Ea2srzjvvPPzyl7/EHXfcgbvuussXa6QgVVzXqhxYBXJcT5YV39NWn6fBgtSuskYlb2d8ajTPupIHOQet2+FCYY1+2rHlF+qIMKMyNqtHnjlSvnvzsbm4p20/0ON6smkZsZCPub/MD+6xPZdLxCp3+73JIODUjFi/r8FoEJCdKL1elTd0oLal0+9roP7tZn7UCSWpO6RYkCI6qW2H65WNTCYO8URUeJgRcyckAQDsThEvbyjx2voI6LT3vcGIIPTkPlL/Bl2QEgQBf/zjH1FfX4/8/Hxs3rwZR48exSOPPIKODlZbaeDkcT0AGBGrgYKUxjuk1ON6ZzHMnI6TqSqo5ldq7/e3Lx3dTlQ0Sq8bqTFWXXcSJEf3dHeV+jDYfFORqiClkQ6pmHAzxiZLY2TFdW1BnU+xs6wBNc1SB9zk9BiEh/l3XE/mMbbHHClNyVPvsMeCVC9Jqu4OXz5XEgULz3G9weVHqc2bkAKzUXqf9c7WUjQww81r3tlaCudxcQ2CIHVI3X3huACtSj+GPCMVFhaGSZMm4cwzz4TZbMbTTz+NnJwcb66tl0WLFvn08cm/Dmgk0FwWF2FGjEaDzbsdLny6uxKAFKLr712dSPv0uNOeeoe9FB2P6wH+6ZBq7XIo2TRpNqsSUqoF01XPSbn5wTu298WewI7rydRnXJkjpS3qHDUGmvem7pDiyB7RyW1UBZpPGMaJqJhwM2aPSQQAtHc78dqmw8NdGkHqjlqxrieXa0RsOCwmAyakRmPF9TM8djmkvg24INXV1YX7778fp59+Os4++2x8/PHHAIBXXnkFOTk5eOaZZ7B06VJfrXNAFi9eDEEQsHz58pPe9oUXXkB2djasVitmzpyJrVu3eny+s7MTv/71r5GQkICoqChcddVVqKmp8bhNaWkpLr30UkRERCA5ORn33nsvHA6HN7+loHaguid8OSMu8G/apGBz6U2+1oLN1x88igZ3JsW0kbEBOytP2qUuSOllp73iOlWguU532JOpD7KO+Ois/7bD9coZuOG8KfWF00bGKpe/3BecBSmXS8SqPdK4njFA43oy7rSnTU6XqHSoJkaFIdqqnaKxVkSEmRDhfg/DghTRiTW2dyvPKRlx4cN+Tpk/ORXujXHx6sbDaOvicetwvf9DmdI5PW1kLF69+QwU/PVirLr7PBajBmjABakHH3wQL774IrKzs3H48GFcffXVuP322/HMM8/g6aefxuHDh/G73/3O6wusq6vDTTfdhMzMTLzzzjsYM2YMrr76anR3e7YZfvTRR9i8eTPS09NP+pjvvfceli1bhj//+c/YsWMHTj31VMyfPx+1tbXKbZYuXYpPP/0U77//PtatW4fKykr85Cc/UT7vdDpx6aWXoru7Gxs3bsRrr72GV199FQ8++KD3vvkgV+DukIoIMyIuQhtv2rR6UH/87npEx4sIMymjEPurmuHQQbC5OuhRz4HmAJAQGaa8yTtS75sOqc0aHNeTJURZlOfP/IrmoMyG2V3eiMomKa9pYlo0Ii2ByzyLjQhDvHuHxd3ljb1GBSgwDtW2or1byhJRb5RCnuQCflVTh6424SDyt83F9ZAHNiamDn1cT5YYZcGZOfEApJPv724rG/ZjhrIuhxP/79ue7qjLp6YFcDX6NeCC1Pvvv4/XX38dH3zwAVavXg2n0wmHw4Hdu3fj2muvhdHom46NpUuXYvPmzXjjjTdwySWXYOXKlRg1ahRcrp4XsIqKCtx111146623YDafvLDx9NNP47bbbsPNN9+MSZMmYcWKFYiIiMC//vUvAEBTUxP+7//+D08//TQuuOACzJgxA6+88go2btyIzZs3AwBWr16Nffv24c0338S0adNw8cUX45FHHsELL7zQq1hGvTW121HlfmM/IjZcM9kx6jeQWilINXXYsWa/1J0XbTVhUvrwX5AoOMkFgU67C0U62NWjWLUjid4LUiajAQmR0kHWkbp2n4z8ajHQXE09tvdlEO62p95d7/TM+ACuRCKP7bV3O3GwpuUktyZ/2F3eqFxmQap/ie6TJy4RqGzUTjc6kdZs9MiP8s7r/sWTe4omL28oRreDReGh+mB7uXI8OzXD5pFHTAM34IJUeXk5ZsyYAQCYMmUKLBYLli5d6vNCws6dO3HjjTdizpw5sNlsmDt3Lh577DFYrdLBi8vlwg033IB7770XkydPPunjdXd3Y/v27Zg3b55yncFgwLx587Bp0yYAwPbt22G32z1uM2HCBGRmZiq32bRpE0455RSkpKQot5k/fz6am5uxd+/ePr92V1cXmpubPf6FKs9xvcDnR8k8OqTKtVGQys2vUl4szsyOh8kw5Og3CnJa7fDrT7E7Q0oAkBwz+G2MtUbOkWrpcqDey2GhzZ125f/piNjht+37wvTMWOXy6r01/d9Qh0RRxBfyuJ4gYJrqew0UdY4Ux/a0IU9dkEoMfBSBVnkEmwdhNyWRt3x/SCpIGQRgnJdORI2IC8epGTYAQFVTJz7eVXGSe1Bfuh0u/L+16u6ok09pUd8GfGTrdDoRFtazzbzJZEJUVNQJ7uEd55xzDl555RV89tlnfX7+scceg8lkwv/8z/8M6PHq6urgdDo9CkkAkJKSgupq6exndXU1wsLCEBsbe8Lb9PUY8uf68ve//x02m035N3LkyAGtORh5Bppr501bbLj2gs0/3MHd9WhgPHba03hBShRFZWQvISoMZqP+C61JHsHm3j3I+uFwPeSpLC12RwFAmi1cyQLbdqQeR1u6Arwi78mvaFZyBSekRiMqgON6spxEdbB5QwBXQjI50FwAO6ROhMHmRCdX3dSpdLvnJEbCavbeNNIlp/R0Sa1YV8Sx7yH4cEe5slP0lBExHq/JNDgDfkcliiIWLVoEi0V6Eens7MTixYsRGen5w//www+9usCnn34af/vb37B06VIUFRVh165dWLx4MRYvXozt27fj2WefxY4dOzQz8nUi999/P5YtW6Z83NzcHLJFKXVBakSsdjqkBEFAdkIk8iqa0NQhBZuPjA9cway8oR1bSuoBSCNN3LGHTiQrvuf5WOsFqaMtXWh1h2nqfVxPpu7yKq1v8+oubJs0nB+lNj0zFl/sqYYoAmv21eAXMzMDvSSv+NzdHQUEdnc9taz4SBgFAU5RZIeUBnQ5nNhfJXV/p9qsXj14DDaJUT0nuFmQIurbpmLVuJ4X8qPURidFYVxKFA7WtKL4aBvW7KvGginMPxoou9OFF749pHzM7qjhGfAp6ZtuugnJyclKd8/111+P9PR0j44fm83m9QVGRkbi0UcfRWFhIa644grceeedWLZsGV566SVs2LABtbW1yMzMhMlkgslkwpEjR/Cb3/wG2dnZfT5eYmIijEZjrx3zampqkJoqJeGnpqaiu7sbjY2NJ7xNX48hf64vFosFMTExHv9CVYFqZE9LBSnAc+wpL8Bje//dValcPmtUvC4KrxQ4UVaT8kZ/b2Wzps94qTOu9L7Dniw5uuf7OFzn3YOsTRrPj5IFY46UKIpYlS8VpAQBOE0D43oAEGYyYGS89PpZWNuKpg57gFcU2g5UtcDulJ5z2R11YuoOqWDcAIHIG74/1PO6P8FL+VFql6gKUP/v2yJNTIXoxUc7K1BWL3VHTU6Lwegk30+NBbMBd0i98sorvlzHgMTGxuKOO+7A6tWrsWHDBixfvtwj5wmQcpxuuOEG3HzzzX0+RlhYGGbMmIGvv/4aCxcuBCDlUH399ddYsmQJAGDGjBkwm834+uuvcdVVVwEACgoKUFpailmzZgEAZs2ahUcffRS1tbVITk4GAKxZswYxMTGYNGmSL779oOFyicoOe4lRYQgP09ZZxKzjgs0vDdCOCaIo4sMd5crHHNejgciMj0Bdazc67E6U1LViTLI2ixfFdT2B5ilB0iGlPsg6csx7ofJNHXbsrZSK+Blx4YiyBn5crD9Z8RGIjwxDfVs3NhbVobnTjhgN5l0Nxr6qZhxxj2COT4nWVH7XqMQoZTw0r7wR545NCvCKQpc6P4qjGycWHxkGQQBEEcpBHRH1EEURG935UWFGg08KHpPTYzAyLhxlDR3IK2/CxqJjOGdMote/TrBxOF14YW1Pd9Rlp7KzbLg0H9qxdOlSrFu3Dk1NTXA6nVi7di3WrVuHGTNmICEhAVOmTPH4ZzabkZqaivHjx/f7mMuWLcPKlSvx2muvYf/+/bjzzjvR1tamFLFsNhtuvfVWLFu2DGvXrsX27dtx8803Y9asWTjrrLMAABdddBEmTZqEG264Abt378aXX36JBx54AL/+9a+VsUbqW0VjB9rc2yJnxGpvBE09FhfIsaf8imali2RscpSyKw3RiRxfUNWqYnWHVLAUpFR/o0e8eNZ/a0nPts8TNDyuB0hjz3K4ud0pYu2B2sAuyAtW7enp9NLKuJ4sRxVsvotjewG1q6zn+Zbj9SdmMhgQHyF183Jkj6i3w8faUenevW1McpRPcjYFQcDFHl1Sh05wa5J9srtSOUk1MTUaYzV64ldPNF+QyszMxLJlyzBy5Ei8/fbbuPHGG3HLLbfgrrvuGvBjnH/++Vi0aJHy8TXXXIMnn3wSDz74IKZNm4Zdu3YhNzfXI6T8mWeewWWXXYarrroK5513HlJTUz3ysYxGIz777DMYjUbMmjUL119/PW688UY8/PDDXvm+g5mcsQBIOz1oTWxEGGzh0hnwQAabf7izpztqFrujaICy4tU7RWp3J8/ioz0dUsEyshdmMiAuQnruOOLFUPPNOhnXk6nH9nLz9T22p95dT4Dn96YFo1SdODvLGgO3EFI6pIyCENDsSb2QO0qbOuxoaue4KZGavLse4NsTUTOy4pS/xe8PHcNuvo6ckNMl4p/fqLKjTmV2lDdot+/fbenSpVi6dCkAYNGiRXj11VdPePvDhw/3uq6kpMSjIAUAS5YsUUb0+mK1WvHCCy/ghRde6Pc2WVlZ+OKLL064HuqtwGOHPe0VpADpoF4ONi+r70Cmn892OpwufLpbyo8yGQTNnZUn7crSSIffyRTXSR1SFpMBseHaGYEaruRoKxra7ahv6/bauJocaC4AGKuDgtSYpChEW01o6XTg24Kj6LQ7dRvwfLCmVfldHZsSpZys0IrkaAuiLCa0djmws7QBoigyazAAWrscOOQuso+ICw+KXUN9Ter6lt4PljW0wxbh/RxaIr1Sb2QyMc13mcNGg4AFk1PxxuYjAIAXvy3Cihtm+Ozr6d1neZXKe4JxKVEYp4P3ZHoQ9K+Ye/fuhc1mw4033hjopZCbVnfYU1Mf1Adi7GnDoTrUtXYDAE7NiEWkBrYYJ32ItpqVUYi9lU1waTDYvMvhVIJsU2KsQXUA7bGduRe6pBrbu7G/WpUfpYPnAoNBwLSMWABAh92J9QePBnZBw6DeXe/0rPgArqRvgiAoXVIN7XavdubRwOVXNCljtcyPGhiP50qO7REpXC4RG4ukDqmIMKNH57svnD06QTnZ8uW+ahyqbT3JPUKT0yXiua8LlY+vYHeU1+iqIHWy7qi+TJ48GXl5eTAYdPWtBrUD7oMrk0HQbJixOocnr6LR71//ox0VyuWzRmnvIIi0TS6otnU7UeLFcG1vKT3WDrlOFiz5UbJk1UHWYS/87Ld45EfpZ2fW6aquzlwd77a3SlWQmq6R3fWONypJPbbXEMCVhC51oDnzowZGnbnHghRRj/3VzWhwj7GOT4mGweDbk3ZmowE/mijF1ogi8L/rinz69fTqiz1VHtm+eohQ0AtWacivOu1OlLhbHdNjw2H08ZPsUAUy2Ly1y4HV+6QDuMgwI04ZwTZ2Ghytj+0VqQLNU2KCK6w/OUa9097wD7LUbfvjNR5orjYhNRpWs/QW46t9NbA7XQFe0eAV1rSg0H2meExSFGLdnYdao+7I0XOweW5+FRYsX4/xD6zCguXrkZtfdfI7acTuclWgOTukBiQxuufvqYwFKSLFxkM9r/v+2shkzrgkRLh3Pf94VwUqG7n7pZrLJeL5b3q6oy6bmhZU3f2BxoIU+dWh2lalM0Kr43rAccHm5f4NNs/Nr0anXTp4OyM7HiZmUdAgZcZruyBVXBd8geay5Kie7+dw3fA7pORAc0GQ8gr0wmw0YOqIWABAc6cDW4rrA7ugIViVr93d9dRyEiMhvy3Wa7B5bn4VFr+5AwXVLehyuFBQ3YLFb+7QTVFK7pAKMxqQbtPuexstYYcUUd++L+oJNPdlfpRaeJgRc8cnA5B2yH15Q4lfvq5e5O6txsEa6b3r6KRITPLT/5dQwSNd8qsDOgg0l8ldJs2dDr++WfpIvbveaO6uR4OnHjkNRAbayRSrOqSCbWRPnYtyZJjPG/Vt3cpzZmZ8BCLCtJ8fpTY9K1a5nLtXH4UFtS9U43paLkhFhJmQ5i7s7qtsRqfdGeAVDd7yrwohAJBP/YiQirDPqvI6tKq+rRtl9VI3QWZ8hGY7v7UmymKCxSQdhrBDikhid7qwtUQ6gWMLNyvP7f5w4YRkmI3S89c7W0vR0Nbtt6+tZa7jsqMun5rO7igvY0GK/OpAVc829JovSMX7P9i8qqkDG90jOknRFo8tvYkGyhZuRlyE1OG3t6JZc8HmxUd7OqS0miM3VOFhRkRbpcLRkWFmSG0tUY3r6TCrYEq6DSb3wfnqvTWa+z08keKjrUoxcFRiJOIjtTmuJxuVJHXPOVyiJrsiT6akrg3H/3aIomfxWqs88qMSmR81UIIgKAX88oYOOHX0/EDkK7vLGtHeLZ1UmJAa7dfCR0y4GeeOSQIgbUjy6sbDfvvaWrZmf43yfiAnMRKT09kd5W0sSJFfFdSoO6S0/cZNnQPhr4LUJ7sqlQDjs3LiWYGnIZPH9lq6/NvhNxDylrlxEWZYzcYAr8b75GDzmuYudHQPvVtFnR/lrxwJb7KajZiSLmXg1bZ06WqcTC/jejKPHCkd/ZxlfeUuCfAMbNeqPFV+VE6C9terJfLYnsMloqqJmTVE36vyo/w1rqd20eQUyE2er206jLYuh9/XoCWieHx3FLOjfIEFKfKr/VVSQSrKYkKMVdvjJx4dUuX+KUh9tFO9ux7H9WjotDq2V9/WjUb37jHB1h0lS47u+b6GUwzcpMqPGpusv4IUAJymGtv7Uke77ellXE/msdOeDoPNz8zuvZusCODOOaP9v5hB8uyQYkFqMBKjmSNFpOaRHxWAE1GJURbMzJGOPxrb7Xhna6nf16AlX++vxd5KabonKyGCG035CAtS5DfHWrtQ19oFQBrX03qFWR1snl/h+2DzfZXNSkvo6KTIoD1YJ//Q6k576nG9YMuPkiWrDrIOD3Fsr661SwnQzIqPQHiYPjvJTs2IVc625uZX+3WDiKE6cqxNeQOanRCBxCjt7wQ5whau5PHsLG0I8GoGRxRFJbz/eN7YqdKXRFFUdtiLCDN6/O3TyamDzcvr2SFFoa2926E8fydFW5AQoNeeBVNSlcsvbyhBt0N/u+R6gyiKHjmGzI7yHRakyG8KdBRoLsv2Y7D5x7tU3VE57I6i4VF3+OVXaqggpdp5Lth22JOpD0qHmiOl3pVuQqp+8wqiLCYl/6q0vt1jYwut0tu4HgAYDIIytlfZ1Ima5s4Ar2jgvj14FIW1UvF1bHIU/vrjKUoRc8W6IuVElhZVN3fiaIu0vqyECB6sDFJiVE82GzukKNT9cLgBdqd00iYQ3VGyEbHhmJYRC0B6jvtYNb0RSr4tOKpMGIyMC8epGeyO8hUWpMhv9qsOREbE6qMgpR57yvPh2J7TJeK/7oKU0SDgjD7GF4gGIzYiDBHurprvDx3DguXrNbGFejDvsCdL8uiQGtpBlrpjZLwO86PUpmf2FHVy87U/trdKZ+N6MvUmGHoa21u5vli5fNGkFKTarDh3rBSs29btxPMa3mlvdxnzo4YjiSN7RAqPcb0A5EepXXxKT5fUivVFIbfpQK/uqFPZHeVLLEiR3xRUq3fY03agucxfY08bi+pQ0yydZT1lhA1RGs/XIu3bfqRB2akFkDoUF7+5I+BFqWDeYU/mkSE1xIKUnB9lEKSuET2blhmrXNZ6jlR5Q7sygpUZH+Hx/1Lr1MHmO8v0MbaXX9Gk7CybEm3BqSNjAQBXnJqujCC+taUUh+u0udse86OGRz0Oy4IUhbqNh7SzkcnopCiMS5HeexQfbcNqjb92e9v6wjplg5ARseGY5n5tIt9gQYr8Rh7VEACkx+rjTb5HsLkPC1If7ehph53FMHPygk/zKj0+FiGFYz8b4G4DeWTPZBCQEBl2klvrU6TFqHSnDSVDqralE4fcI0zZCZG634kwLiJM6d45UN2i2eICAKzao79xPdmopJ7CpV46pFZu6OmO+tGkFBjcZ6Bt4WZcNCkFgLQD2xOrCwKyvpNRd05ns0Nq0MxGA+IipKzOMhakKIQ1tncr8QoZceGItpoDvCLgkilpyuUX1xXpIgPSG0RRxLNfHVQ+vnxqmvLaRL7BghT5hdMl4mCNVJBKjrbAYtLHAVZsRBhi3cHme3wUbN7e7UCu+8xDuNmIqZxRJi+obuqdISOKUAodgeBwupRMpZQYKwyG4HyBFwRBGUWpbOwYdCCoOj9K7+N6MvXYnpa7pL7I1+e4HiAVceRMnrzyRjic2g6irWjswGd50s87ymLCrNGeJ2PmT05FtLtb+PO8KuVstVaIoqh0SNnCzUphhQZH7pI61tYd8lvMU+jaXFwP+RBjokZyIyenx2CkO/M3r7wJ3x/qe/OJYPP9oWPY4T6pk26zYrrO3gvoEQtS5BdHjrWh0y69OR6hk0BzmTy219Lp8MmOP6v31iijVWdkx8Fs5J8lDV+qzYq+yj12p4i/frYPzZ12v6+pvKFDCewM1vwomRxs7hKlMbDB2FSsnbZ9bzlNNbaXq9GCVGVjh9JZNCI2XJe/o6MSpS6pTrtL8wHyr35fouSSzB2f1OtEldVsxOVT05WP//7Ffk2doT98rB3NnVIBJZuB5kOmzpEqG+RzJVGw2OiRH6WN131BEHCxR5fUoQCuxj+k7Kie7qjLpqazO8oPeORLfuG5w54+8qNk6mBzX4ztfajaveIsjuuRl1w+NV0a0+vjcy9/V4ILnvwW//6hDC4/BlUW16nyo2zBvT26OntosIVsOdDcKAgYnaTv/ChZSoxV2cxiZ2mjJneBUweun67TM6LqHCmtdRSpNXfa8c7WMgCA2Shg7vjkPm933rhEpLgLFltK6rG2oNZvazwZ5kd5h0eOlA9O+hHpwfeHpIKUQQDGpWijIAVIr4XyCbbvDx3Dbg2/rnjD5uJ6bDssZTCm2qy6fS+gNyxIkV/ocYc9WXaC73Kkals68V3hUQBAQmQYxug8vJi0Y0ZWHO6cMxoj4sJhMggYESu9sJqNUomqrrUb932Qhytf3Iidpf4JQA6FHfZknjvtDTwzqaa5U/k5ZSdG6D4/Sm26qktKiwGpq3Q8ricblaSPnfbe3VqKVvd41tmjExET3ve4m8lgwJXTRygfP7aqQDO7PXGHPe9IYrA5hbjqpk4UuV/3cxK1lRtpMAhYMLlnx70Xvy0K4Gp8z6M76pS0oI2W0BoWpMgvPHfY01dByqNDqty7BalPdlVCfm991qgEtoWSV83IisNDl0/Giutn4C9XTMHiOaPxyI+nYIYqz2d3WSOu/H8b8dv3d6O2xbddK0UhVJBKVhWkBtMhtdljXE8bORLeos6R0trYXk1zJ344IhVm021WpOvsxIksMz4CJvcbaK3utNftcOFf3x0GIHVw/mhiyglvPyMzTgnFL6hpwX92lPt4hQOj7pBS78hLg5MY3bO5BYPNKRR5jOtp8HV/1ugE2NwnDb7cVx3QLFJf2lJ8DJvdGZ4p0RackR0f4BWFDhakyC/kkb0wk8HjbJge2MLNSrB5fqV3g80/8hjX4xMf+V5ilAV3nj8av/nROI/dLj/YXo4LnlyHleuLBx3CPVDFR3vexKTaQqkgNfAOKXVBaryG2va9ISMuXHn+31xcj8b27gCvqEdufrUSKKvX7ihA2rUs0707bPHRNjS1+z8r7mQ+31OJavfI5qkZsSd9LhAEAT+dkaF8/Myag+i0O326xpNxOF3KjliJUWGa2BFLr9ghRaFuY5HqRJRG8qPUzEaDsuupKAL/uy44u6Se/6YnI+vSqWkwsjvKb1iQIp9r73bgiPtNRrpNnztryds5ezPY/GBNC/ZWNrsfPwJpNn2ekSd9mpgWgwcvm4RrzxiJcHd7eGuXA49+sR8Lnl2PdQePev1rFtdJhZloqwkRYSavP76W2MLNCDNJL7GDec7Y5H5jajQIGJ0cXGNAgiAoY3tOl4iv9msnD+iLPfof15Opx/Z2qbp4tEAURby0vkT5+KLJJ+6Oko1LiVZ2oK1q6sQr3x/2xfIGrLC2VdmoJZvjesNiCzcro+QsSFGoEUURG935UWFGg2ZzI+eMS0JEmPRe8aOdFahs7Ajwirxr+5F6fOf+/5AUZcHMHGb6+hMLUuRzB2talTPPegs0l6nb8fO8lCP1EcPMKcBMBgPmTUzB366cgvPGJioB6MVH23DTv7bil6/9MKjunhNp6bTjaEsXgOAf1wOk4ot85r+soX1AuTdVTR047C5ejUqM7LXrWDA4TT22l6+Nsb2jLV3Yetjdph9j0V3O4fHUweb+yocbqI1Fx7C/SjoRk5MYibGDyE28anoG5Kn2//ftITS0Ba7DTj2ul8NA82HxfK7s8OtGG0SBdvhYOyqbpI7RMclRmt1p22o24gL35hMOl4iXN5Sc5B768uzXqu6oU9gd5W/a/K2noHKgSr/5UTJ1QSrfCwUpl0vEf90FKYMAnMk5ZQqgaKsZN87Kxh8vnYjRqu6Kr/bX4EdPr8cTXx5AmzuAeKhCKdBclhwjHWTZneKAziZ6jOulaq9t3xtGJUUqWRQbCo8O+/fKG77c6zmuJ+g8y29UYk+RR2vB5i+tL1Yuz5+UMqif9YjYcJwzOhGA1K38z7WB24J8typPkh1Sw5foHnHudrhwtLUrwKsh8h95dz0AmKDx1/0LJyYjzF0we2drKeoDeFLAm3aWNmC9eyogMSoMZ43mMZm/sSBFPndAtcOefgtS3g0231xyTDkjMiXd1u8OQ0T+lJ0Qid8vmIBbZ+coRYNupwsvrC3ChU+tw393VQw5Q624LnTyo2SDDTbfXFSvXA62/CiZQRBw2shYAECXw4VT/7IaC5avR65qhzt/U++ud3qW/t+ISplG0kjsrrJGzXScFFS3KKPAiVFhHt1yA/XjaenKAdEbm44ELARb7pASwEBzb0hkjhSFKI9A8zTtBZqrRVvNmD1WOinQYXfitY2HA7sgL3nu60Ll8iVT0mAysDzib/yJk88dUO2wp9dRCFu4GXER7mDziqZhv8H/mON6pFGCIGDWqAQ8unAKFkxOVdqWq5s7cfe7u/Cz/92EvZWDL8qqO6RSQqVDKrrn+zw8gNHHTe4OKZNB0GyOhDdEWXvywxwuEQXVLVj85o6AFKWOtXYpu+okRVswUqcnTdQEQcBod5dUU4cdJV4aux2ulRt6uqPmTUwZ0khEXEQY5k2Uxka6nS48tbrAa+sbqE67EweqpBNtaTarprZo1yuPYHMv5XQSaZ3LJSq5kRFhRmTFa7+4PX9SCozuztZXNx7WRJfzcOSVN2JtgXSiJD4yDGeP5jFZIAR3qiwFnCiKyg57tnCzrneiyYqPREN7I1q6pJD2oeZGdNqdWLVHyk6xmg2Y5u4WINISq9mIn87IwLljE/HetjIlO23b4QZc/vx3+PmZmfjtReMRFxl2kkeSeIzshWCH1MnO+lc0dii3GZUUqQSiB6Pjc43k8v7S93bjna1liLaaEG01I8ZqUi5HWXouR1tNiLGaEeX+/FAzN3Lzq/DQJ/uUfK8RseG6H9eT5SRFKoHmu0obA17grGnuxH93SSdiIsKMmD0mcciPtWBKKtYX1qG1y4GPd1Xil+eOwpQRNm8t9aT2VzXD4f6dyWZ+lFckDeK5kihY7K9uRoN7J9TxKdG62PQpIcqCM3Pisan4GJo67Hhnayl+ee6oQC9ryDy7o1Jh0miGV7BjQYp86mhLl/Jkm6HT7ihZVmKE8gZ/T0XTkAtSa/bVoMV9RmFGZlxQH3iS/qXEWPE/F45FXnkj3ttWhpqWLrhE4K0tpfgsrwq/uWgcfnFm5klfxIuOSiN7RkFAYtTAilh6py5IHa47cZfKZtW2z8E6rierae47I6bD7hzS7o5Ws0EpVEVbeopW6gJWlEUqYsnX5Vc04R+5BzweZ1dZI7YfadD9LnuAFIov21nWgKtmZARwNdKZdLtTKuKcPy5pWF1FEWEmXHpKGt77oQwA8I9VB/DmL2d6ZZ0Dkcf8KK9Td0gFagyTyN82Hup53df6uJ7agimpSkf3yxtKcMOsLF1uwpJf0aTs9hsXYcY5wzhRQsPDghT51H5VftQInY9CqFtp95Q34opT04f0OBzXIz2amhGLiWkx+Gp/DT7Lq0KXw4WmDjse/O9evL2lFH++fDJm9dPq7HKJyshaUrQlZObz4yLCYDQIcLrEk2ZIbVIFmk9I1c8b06FItVlR3uC9LaM77S502ruUXRyHSgDwaV5lUBSkchIjIUDqPgt0sHlblwNvbT4CADAaBFwwIXnYj3n++CR8faAGda3d+O5QHdYfPIrzxiUN+3EHYrdqh73sRO2P2OiB+iQFO6QoVHxfpJ9Ac7URseGYNjIWu8oaUd3cif/urMTPzhgZ6GUNmro76uIpaZrd4TAUsCBFPlVQrf8d9mQeweZD3GnvWGuX0gEQF2EO2p20KDiZjQZcPCUNs0Yl4D87KpQiyoHqFvx85WZcOjUNf7hkYq+suMqmDnTaXQBCZ4c9ADAYpO3Mq5s7caS+DaIo9jsStlmVHzUqKbi7Li6fmo4X1xUpBRP5v4vPG4UJaTHo6Haiw+7s+a/68kn+22l3YqgJfyKAavdmE3pnNRuRHhuOisYOHKhuQUe3E+FhgTmD/e8fytDcKXUFn5UTj9iI4XdImo0GXHnaCKx0bz3+j1UHMHtMol9GXuQOKaNBwMg4FqS8wWI2IsZqQnOnA2UNLEhR8LM7XdhaIuUX2sLNSNNZlMHFU1Kxq6wRALBiXRGumpExpFzAQNlX2YzV+2oASD//c8eyOyqQWJAin5KDPwEgI1bfb9zkYPOGdjv2VjTD5RIH/eb3092VSvbEzJwEGIIkr4RCS2xEGG6dnYM545LwzrZSpfvn87wqfL2/Br86fwxuP2+UMpbjEWhus/T5mMEqOVoqSHXaXaht6eoz0L2svl3pGBqTHBX0Z+lmZMXhzjmj8WleJaqbOpFqs+KKU9Mx3b3rWpRl6G9NXKKILrtLKVK12x2qYpUL7d0OaTSw4Cjaup0e9xUQXPlmo5MiUdHYAadLxJ6KJpyZ4/8dBB1OF/7vuxLl44smp3rtsc/IjsfqfTU4cqwd+6qa8d/dFbjyNN+OJrZ2OZTx44y48KD/W/WnxCgLmjsdqGnuQqfdybB4Cmq7yxrR7n4NmpAarbv8wtFJURifEo2CmhYU17Vh9d5qXHxKWqCXNWD/XKvujkrlc3mAsSBFPnXAPbJnEIC0WP2/0c9K6Ak2P3ysDaMGGRT7kWpcbxbH9UjnxiRH4Y8XT8R3RXX4cEcFWrsc6LS78PSag/j3D2V44NJJmD85BcXuAzggdHbYkyXHWAD3n/3hurY+v//NqnG9UOmanJEV55PROIMgIDzMKHUDnaDRLCs+ss8uraGOYmtRTmIk1hdKIyE7SxsCUpDK3VutFFunpMd4daddgyDgp9Mz8NSagwCAJ788iIunpPm0kLGnvAmiuwWP+VHelRRtQbE7a6+8oR1jkkPjuZBC0/c6zY9Su3hKKgpqpOO8//dtERZMSdVFYa2gugVfuDeXirGacN5Y/4x7U/9YDiSfsTtdOFQrHYimxFiDovqclaDKkRrk2F7R0Vbsdrf6j4wL132mFhEgjaWdNzYJf7tyCuZNTIbcNFje0IHFb27Hxc9uwJOqrdmbO+wBWmlgqMN6+8uRUudHBXuguVbIXVoj4sJhMggYEReOX50/WunSCgbqEyaByJESRREr1xcrH8/3YneUbGJaDCanSwdzFY0deNOdVeUrear8qKFubEJ9Uz9XMkeKTiQ3vwoLlq/H+AdWYcHy9cjNrwr0kgZNnR81UacnoianxyDTna+7p6IJ43Ty/+P5b3q6oxZMSeXmUhrA/wPkM4fr2tDtlHJjvHlWNJDUZ0TzB1mQUoeZ9xf+TKRXEWEmXHtGJv58+WSPcM4D1S1o7eoZjfp4VyW2H2kIxBIDIlnVEXWkvvdOe6IoKjvshRkNPMj1oxlZcXjo8slYcf0MPHT55KAqRgFAms2KcHe30I7SBojiUNO1hmZrSb3HSRhfhfb+dHoG5HPyz39zCE3tvit6q3fYy2GHlFclqnYlLT3JJhAUulbtqcLiN3egoLoFXQ4XCqpbsPjNHZovgqi1dzuws1R6H5QUbUFClD6jDARB8OjqtjtFzf//OFTbgs/3SGuLtpowx0+bYdCJsSBFPnNAtcOe3gPNZeqd9tRvTE/G5RKVcT1BAM7M9v/oBJE/jIgNx29+NA53zhkNYx+d2/JOZqEiWXWQdbiPg6yy+g5UuoO0RydFBkUnKWmDQRCUXeBqW7pQ5efAdjlwHJC6o3w1yjEyPkLZsbapw47/t+6QT74O0LPDnsVk0F0IsdZ5dkh5bxdO0qcuhxMHa1qwak8V/vlNIZa+twtX/PM7/OqtHQCgbF4hQnpf/axqxzSt++FwA+xO6TvQa3eUbH9ls8fH8gi8Vv9/PP/NIWXsev6kVFhMzKrTAmZIkc8c8NhhT9+B5rIYdbB55cCDzbeXNig5GpP+f3v3HR5Vlb8B/L0zk0x6IZV0AoSEDrEAKl2aBrAg4CqCritgWcB1LWtvWAFRlAUV7AV1EVHwp1SRHhBCLyEQSE9IJ8mU8/tjkpuZmwGSkMxMZt7P8/gsmcxMTu6+mbnzved8T7hfi+wyROSoJElCcmyg6QOoYlaGM+1k1hhB3u6oOwynCxvOkNqWXj9t31X6R5HtxAf74HDt5iJ/ZRYjwkazlU/ml+P3w6YdjAK93HBVXOvOPhvfOwK7MoqgNwos+zMD9/SPa/HftbC8Wn4fj2nnZZMd/VxJiFnxnjvtuY7zFTU4mV9e+18FTuaZ/n2mqBLGRk7qFMJy8xRHZ7Fcr432j6qTU9rwfE7A1KepsLzaoWZ/ncwvx0/7TBdEfbQaDO7C2VGOggUpajVHnXCGFGBatne+shjlTWhs/sOe+uV6/bhcj1xEuL8Hzp2/APNzSmfbyexyNGoVgrzdUVBeg9OFlRBCWMwU2Z5eJP87Mbxtn5iS44kPqV9WtvfMeYyx0S5IH5rNjhqeFAaNqnVn/gX5aDEsMRS/HspFjd60scJbE3q16M/Yb7ZMP45La1tcgKcb1CoJBqNAJntIORWDUeDs+UpT0SmvwqIAVVRR0+jnkWDqW2lQVKokWL7WObqtZg3NW2sps61YO88DAKMABr+1Ef8c1hlT+sc5RJ+mRetPyEXOEV3DuJOnA2FBilpN3VVZDzfTBzJnERvkhb2ZxQBMTfwuV5Cq0hnwc+0SJa1Ghb7RAa08QiLHkNIzwul3MmuMUF8PFJTXoKxKj/OVOrSrfT0UQmBbXf8ojQpxQc4xk5QcR3yweUGq2CY/s6C8Gt/vOQvA9P5/Q+dgm/zc0T3a448TBaisMeD7PWfx9xs6tGiRd38m+0e1JpVKQrC3O3LLqnGmqGHxnhzD2gPZWPD7cZwqqECHYG/MGt4Zo7qbCt0V1Xqk55sXnEwFqFOFFajRGxv9M9w1KoT7eaC9vwfC/T3Q3s/0v2F+Hth/tgQfbDppcX8B4MEhnVry12w1xZU1OJBlei2JCvSEr4ebnUd0ZZTneebKqvR4+efD+HLnGTxzc1cM6RJqjyECMPU1XvmXaXKAt7saQxPtNxZqiAUpahWlVTqcKzZNbY8M8HSqk4pYsxPRtLMlGNc78pL333AkD6VVegBAn5gAaFmRJxdRt5PZT/uzkFNShXB/D4ztFeF0zaMvJ8RXC9T298worJALUhmFlfJ0984hPtCwfxS1MF8PN4T4apFfVo20cyXQGYyt3qfs022n5Q+fAzuHwMvdNqeaPloNburRHitSz0II4LU1R7B82jUt9vzmO+zV9eailhXsq0VuWTUqawworKhBsAMt9yFTMWr653vk4sOR2gbWieG+KLmga3KfugBPN1PByd8D4bVFp/b+ngjwcoPqIp8bzM8rzGfm1C2ndXTb0wvlTgZJTjAr2tp53vCkMBzPLcPWk4UQMC2nnLZsFwZ3CcHTN3VFp9DLryxpaYs21M+OupGzoxwOC1LUKo5ZLNdzrhM388bmaY3Yae9/5rvrxXO5HrmW5NhAJMe6VgFKybyx+enCCrkgtz29fto++0dRa4kP9kZ+WTWq9UYcyS5Djyj/VvtZF2oM+GxbBgBALUkYnhTWaj/LmqGJoVh3JA9FFTXYeDQfW08WYEDHK5+hJYSQdwz0dldbNOCmlmPZ2LySBSkH89b/HQPQcCaM+SZGSmqVhDBfLcLl2U6epn/7ecDTvXlFgbrzisyiSrz48yEIAby77jhu6ROJMD/Hbgmw9WT9+35Se+d437d2nnd9p2AM6RKKr3adwcna/l4bj+Zjy/HNmNI/Dv8c3hn+nraZHXamsBI/1H4W8+LsKIfEghS1Cosd9mzURNVW/Dzd0M7LHUWVNZdtbH6+ogYbjuYBAPw93ZziaggRNY1lQaq+N8q2kyxIUeuLD/bGjlOmXmV7M8+3akHq+z1ncb5SBwC4Ki5Qng1oK25qFcb3jsDHf2YAMM2SWjnzuituQJ5dUoWC8moApj6SzjTr25FYNDYvqnS52bSOSgiBVfuycCKv/KL38XJXo33tDKf62U4eCPbRQt1KGwBEt/PC4IQQbDiaj4oaA15bcwTzJ/ZulZ/VUv48YWporpYkJIQ59/t+XLA3nhiViF0Z57EiNRPnK3XQGwU+/vMUVv51Do+OSMCkq2NaLR913t94Qu47NjwpzGazdqnxuD6AWoV5Q/NIJ2poXie2ttdLebUep6zsnFVndVq2vLXrNR3acVceIhcUanbFtq4gJYSQZ0hpNSr5NYWopZn3OWzNPlIGo8BHW+qbmY/sGt5qP+tS+nUIkjdS2X+2BD+nZV/xc1ou12P/qNZiPiOKjc0dQ35ZNaZ/nop/fv3XRe8T4e+Bdyb2xpOjkzB1QBxGdQ9H7+gAhPl5tHqxYVyvSHjXzrT6395z2J1RdJlH2E9OSZU8Wygu2Msllo1JkoRrOrTDy+O7I6Vne7ipTXkoqqjBf/53ADe/u8Xi4lxLO3u+Et+lmnoaerqpMTyJs6McEQtS1CqO5JTK/450shlSACw+PB64xLK9lVyuR+TyzJehZNQWsNMLKpBXZppx0TnUp9V3ISPXFR3oCU3th8K9Z8632s/5/XAuThWY8p0U7osYOxVZVSoJt/eNkr9+89ejTWqobE3dcj0A3HygFSmX7JF9/bQvCyPmb8KvB3MtbpcU/zu+T6TdZg36eGhwS5/6Xq7P/3SwwS58jmLryQL53662YkKrUWNc70i8PK47rolrJ99+OLsUk5dux4zPU1ulCP3+xpPQ1+ZhWFIoZ0c5KJ4BU4sTQshL9tp5ucNb63x//HFmjc33n7VekDpdWIHU06aT/8gAT0Q74UwxIro8d40KgV6mXglnamdIcbke2YpGXT8DL6OwsknbrDfF0s3p8r9HdLPP7Kg63SL85O3UzxRV4ssdp6/o+cxnSHXgDKlWE+xbv8STBSn7KSyvxswvUvHwV3vlJbi+HhrMGNQRMwZ1RGRtkTsy0BMzB3e0+9LKgZ1D5HPsA+dK8c2uTLuO52L+PFH/vp/oJP2jmirIR4t/DIzHv0d2QYxZT941B3IwbN4mvPXrUVRU61vkZ2UVX8CK3aYseLipbN7TkBqPBSlqcVklVSir3VXOGZfrAZYzpC7W2Ny8mXm/+HbsOUHkwup6oxRW1KC0SseG5mRT8cH1y/b2ZRa3+POnnj6P3bUXYCL8PdA9wr5X/yVJwu3J9bOkFq4/gbIqXbOey2gU8oWnAE83BHjZti+WK/Fy18jLrzKL2sauac7ml7RsjJi/Gb+k5ci3JccG4sWx3eTm1c+ndMPiu5LxfEo3uxejANOsyMnXxMhfv/nrEZRUNu/vvbUIIbCtdoaUu1qFjiG232nOkSSE+eLpMUmY2j8Ovh6miQs1eiPe23ACQ9/eiB/2nIXxCme6fbDxpNw2ZWhiKHyccIKEs2BBilrcUbPlelFOWpDy9XCTm7UePFfS4EVTCCEv15MAXNuBy/WIXFmor1kfqYJKbE839bnwcFMhth1nXFDrig+pz1hrLNv78A/L2VGOcAEmLshbXhpSVFGD/25Kv8wjrMsorJAvsrF/VOurK95nlVy44qWW1HhFFTV46Ms9mPnFHhTWzqL00WrwwMB4zBjUEb4ettkRrbkSwnzlv/fzlTrM//2YnUdkKaOwElklVQCATqE+cFPzI7hKJeH6zsF4dXwPjOwWJvcbyy2txpxv9+HWD7Y2+/0qp6RKnimn1ahwI2dHOTT+NVCLO5ztvDvsmaubJVVRY0B6gWVj872ZxcioXZrTJdzX5jsNEZFjMd9pb/2RPHnHrs6hvq3e9JUo3qyQsreFZ0idLqzArwdNsyn8Pd1wbYd2l3mE7dzSJ1L++/pwSzpyS6ua/Bz72T/KpuoamwsBnCvmLClbWHsgByPmb8Lq/fUbAPSJCcALY7vh6jjH+Xu+nAlXRUGrMX20/Wz7aYt+tvZWt7seACS56HK9i/F0V2NCcjReHNsNvaMC5Nv/yizGLe9vxZxv/mrya/fiTSdRYzAVtId0CXX4gqqrY0GKWpz5DntRgc578hbb7uKNzf+3h83MiaieeUHq2931/S0SuVyPbKCdtzv8PU0n5H+dKb7ipRDmPt5yCnVPNywx1KGu/If4ajGkSwgAoEpnxIJmzJrYx/5RNhXiy532bKW4sgazvt6L6Z+noqDcNCvKy12N+2/ogJmDOsqvGW1FoJc7burRHoBp18/nVx2EEI7R4Ny8oXmiizU0b6wwPw88NLQTZg/vjAj/+lnlP+w9hyFvbcSiDSdQpTNc9nnySqvw1c4zAEw9PEd05ewoR+c4Zw3kNOquSKhVEsL8tZe5d9tl3tjcvI9Ujd6In/ZnAQDc1JJDrK8nIvsy/5BlftWf/aPIFiRJkpftlVXrkV5Q3iLPe76iBt/uNm2prdWoMDAhpEWetyXd1KM9PGu3V/9mVyZO5JVd5hGWzGdIxQaxINXagrnTnk38digXN87fjJV/Zcm39Y4KwItju+HaDkEOsey2OW7sGia/325PL7LohWUvRqOQNzLxcldbXNCmhrpF+OO5lG6485oYeNX2lKusMeDNX49i+LxNWJOWfclC4383p6O6drnv4IQQ+LWxwqorYkGKWlSN3oj0fNPytXA/D6feytyisbnZCeumY/korm2m2Cc6EJ61L6ZE5LrMC1J1PN3UiHHiWaTkWMyX7e05U9wiz/nFjtO4UHvF+rpOwQ7ZNNbXww2ju5t2/TMK4LU1Rxv9WJ3BKM+ADvHVOuTv52xCfDhDqjWVVOow55u/cP+nu5FfZlo67uWuxn3Xd8CDQzq2+ab9bmoVJl0dLX/9ys+HUFnTMru2NdfhnFJ5t8IuYb5QcZn+ZalVEoYmhuLV8T0wtEso6g7Z2fMXMOOLPZi8dDsOZTVckplfVo0vandVdVNLGGnnHV+pcdpUtWDq1Kn2HgJdxsn8cuhr5+47a0PzOhaNzbPqG5v/b+9Z+T794tvO2nsiaj1e7hp5J5k6CWE+PDElmzHfaW9vCxSkqnQGLN9qOvGXJDh009hhSaEI9DJdJf/9cC52ZRQ16nHHcsvkK+0dODvKJsyL95wh1bLWH8nFiAWb8IPZLtA9I/3x4thu6B/fdmdFKfWM9Ef3SNOyuKySKizeeNKu49l6on5X3aT2XK7XFD4eGtx5bQyeu7kbksxmlG9PL8LN7/6Bp/6XhsLanpwAsPSPdFTpTK/ZgxJC2tyyU1fVpgpSSjqdDo8//jh69OgBb29vREREYMqUKcjKyrrsYxctWoS4uDh4eHjg2muvxc6dOy2+X1VVhQcffBBBQUHw8fHBbbfdhtzcXIv7nDlzBjfddBO8vLwQGhqKxx57DHq9favw9nbEBXbYM6dsbF5yQYffD+cBAHw9NOhq562vichxhCpmSXG5HtlSXJAX6j5vtsROez/+dU5uzp8cE2h1FqCj0GrUGNsrQv761V8ON6q3jEVD82DOZrSFQG83eTYEC1Ito+SCDo+t2Id7l+9Gbqnpb9bTTY1pA+Lw8NBObX5WlJIkSZh0VYy8ocHizel2nW33p0X/KL7vN0dkoCfm3JiABwd3lN9rjAL4cscZDH5rI/717V8YPm8Tlmw27aaqkoBRnB3VZjh8QaqgoAD33HMPYmJi8NVXX6FTp06YMGECampqUFlZiT179uCZZ57Bnj178MMPP+Do0aMYO3bsJZ/zm2++wZw5c/Dcc89hz5496NWrF0aOHIm8vDz5PrNnz8ZPP/2EFStWYNOmTcjKysKtt94qf99gMOCmm25CTU0Ntm7dik8++QTLly/Hs88+22rHoi044iINzetY9pEqxpq0bHmb4mvi2jn1kkUiahrlbnqCO5qTDWnd1PLOt8dyy1BR3fwLaEajwNI/Tslft4WmsQM6BsuNcveeKZZ3BryU/WYNzeM4Q8omNCqVPPv8TGGlwzSlbqs2HcvHqAWbsSK1fvZ+9wg/vDC2G67rFOw0s6KUwv09MDwpFICpnchLqw/ZZRw1eiN2njLNyPT3dEN7s2bd1DSSJKFPTCBeHNsNt/WNlHdULKvS47s953Air743olEAJ/MrLvZU5GAc/tPy7NmzsX37dnz22WcYM2YMli5divj4eBiNRvj7++O3337DHXfcgS5duqBfv3547733kJqaijNnzlz0OefNm4f7778f06ZNQ9euXbF48WJ4eXnh448/BgCUlJTgo48+wrx58zB06FAkJydj2bJl2Lp1K7Zv3w4A+L//+z8cOnQIn3/+OXr37o3Ro0fjpZdewqJFi1BTU2OTY+OIjmTXF6QiA1xghlQ78z5SpRbToLm7HhHVST19HsdyLRtJr9hzFqmnr3ymClFjxYeYlu0ZheXsn6badCxfPvnvHOojP68jU6sk3No3Sv76jbVHoTNcuiq8L9N0jCQJbERsQ3V9pMqq9Si5oLPzaNqmsiodnvh+P+75eCeyS6oAAB5uKtzTPxb/HNZZLvo5s5SeEfKSrf87lIvNx/JtPob9Z4tRWWPqs5cY7uu0BUBbclOrMLp7e7x6Sw9c3ynY6n0kQN5gihyfwxek9u7diylTpmDQoEHw9/fHkCFD8Prrr8PDw3qFuaSkBJIkISAgwOr3a2pqkJqaiuHDh8u3qVQqDB8+HNu2bQMApKamQqfTWdwnMTERMTEx8n22bduGHj16ICys/qrgyJEjUVpaioMHD17pr91mHa2dIeXlrpb7NTgz88bm647kyldBwv08LL5HRK7N2okRT5jI1swbm+/NbH4xtG5ZBNA2ZkfV6RXlj4QwU/EsvaAC3+zKvOh9q3QGHM01ndNE+HtC68YNSmzFfPlnZtGFS9yTrPnjeD5Gzt+Mr83y3bW9H15I6YYbOoe4TFHEw02N282K0C/8dPCyReiW9if7R7Uaf083TB0Q12D2OQAIADm1hVhyfA5fkLruuuuwbNkyrF69+rL3raqqwuOPP47JkyfDz8/6H31BQQEMBoNFIQkAwsLCkJNjmr6dk5MDd3f3BkUt5X2sPUfd96yprq5GaWmpxX/OpLiyBjmlpj/+qEBPl3jD8/VwQ1DtVabThfXr0/vFt3OJ35+IGsfaiRFPmMjW4kPMClLNbGyedrYE29JNH7LCfLXoFR3QAiOzDUmSLD6gLvj9+EWXLh7MKoWhdrOSOF5gsqlgHzY2b47yaj2e+l8a7v5oJ7Jq31u0GhXu7heL2cM7I8jHcfu8tZZr49uhY+3r3sn8CnyyNcOmP9+8f1QS+0e1ivb+HlB+4pJgWrZJbYPDF6TmzZuHiRMnYvbs2fj000/Ru3dvLF68uMH9dDod7rjjDggh8MEHH9hhpJc3d+5c+Pv7y/9FR0df/kFtiHn/KFdYrlfH2kyoflyuR0RmwnnCRA4gzM8DXu6mmT57zxQ3qz/P0j/MZkd1C4eqjV18iQ/xQXJsIACgoLza4vcxZ9E/Kpj9o2yJBamm23qiACPnb8aXO+pbliSG++KFsd0wKMF1ZkUpqSQJk6+Jkd9/3/n9OPLLqi/5mJZSWaOXN5AI8dW6ZEHQFlJ6RkAA8v/HEkwX/Mw3siDH5vAFKW9vb7zyyis4fvw4xo4dixkzZmDOnDlYsmSJfJ+6YtTp06fx22+/XXR2FAAEBwdDrVY32DEvNzcX4eGmbvzh4eGoqalBcXHxJe9j7TnqvmfNk08+iZKSEvm/zMyLTxVvi466WEPzOhq15Z9Re38Pi5MpIiKeMJEjUEkSOtQ25y4or8a54qYthzpXfAE/p2UDAHy0mjbbK/HWPpFQ135AX7I53eoHVPMeWx3Y0NymzJfssSB1aRXVejz74wHc+eEO+e9Zq1Hhb9fGYM6NCTwfhWlDghs6m3oNlVXr8cbaIzb5ubsyzkNnMBX9OTuq9STHBmLGoI6IDPSERiUhMtATMwd3RN+YQHsPjRrJ4QtS5gICAvDAAw9g9OjR+OOPPwDUF6OOHz+O33//HUFBlz45cnd3R3JyMtatWyffZjQasW7dOvTv3x8AkJycDDc3N4v7HD16FGfOnJHv079/f6SlpVnszFdXDOvatavVn63VauHn52fxnzM5klO/BDEq0DVmSKWePi/3jaqTXVLFRsVEZIEnTOQormTZ3rItp+RlbEMTQ+GuaVOnkbIwPw8MTDB9QK2sMWDhuuMN7rOvdoaURiW5zDmNowjxMe8hxYLUxWxPL8Sodzbj022n5dsSwnzwfEo3DOkS2uZmL7amW/pEwrO2D9yK1LP4K7O41X/mVvPleuwf1aqSYwPxfEo3LL4rGc+ndOO5VRvj8GcSs2fPxqZNm1BSUgKDwYANGzZg06ZNSE5Ohk6nw+23347du3fjiy++gMFgQE5ODnJyci65092cOXOwdOlSfPLJJzh8+DBmzJiBiooKTJs2DQDg7++P++67D3PmzMGGDRuQmpqKadOmoX///ujXrx8AYMSIEejatSvuvvtu7Nu3D7/++iuefvppPPjgg9BqXfNqhCsu2WOjYiJqLJ4wkSMw3xGvKQWpkgs6fLXTtBzITS1hSJeQlh6aTd3cM0LeNvyrnWeQnl+/C2ZplQ7ptVuGRwV6NpgJTa3LW6uWiwecIdVQZY0ez686iElLtstN3901Ktx5TQz+NaKLxQwzMvH1cMO43vUzkp/78QCMxqYvWW6KrWYNzRM5Q4roojT2HsDlxMTEYM6cOTh+/DgqKiqwceNG3HvvvXj44YeRmZmJVatWAQB69+5t8bgNGzZg8ODBAIDBgwcjLi4Oy5cvBwBMnDgR+fn5ePbZZ5GTk4PevXtj7dq1Fk3K58+fD5VKhdtuuw3V1dUYOXIk3n//ffn7arUaq1evxowZM9C/f394e3vjnnvuwYsvvtiqx8NRGY1CXrIX7OMODxfZjYaNiomIqC0xX37WlJ32vt55BhW125cP6BgMX4+2vZOuv6cbRnULx4/7sqA3Crz561F8cFcyAOCA+XI99o+yOUmSEOzjjszzF5BVfAF6g9Hli4JrD2Rjwe/HcbK2cFq3FAwAOof6YOqAOIT5sSfhpQzpEorNx/ORVVyFfWdL8N2es7jjqtbp51tcWYMDWabXkehAzzb/eknUmhy+IDV79mzMnj0bADB16lS5qAQAcXFxjWrIeerUKUydOtXitoceeggPPfTQRR/j4eGBRYsWYdGiRRe9T2xsLH755ZfL/nxXcPb8BVTWnqi6Uv+ocH8PnDt/AeYpZKNiIiJyVD4eGoT5aZFbWo2DWaWo1hug1Vz6IlKN3ohlf2YAML3H3ZgUdsn7txU3dg3DhqN5KK3SY82BHOw5cx59YwKxz6wgFcf+UXYR7KtF5vkL0BsFskuqEN3Odc4tldYeyMb0z/c0uF2jknBb3ygMS+LyvMZQqyRMvjoGb/92DADwxtojGNU9HH6tUCzanl6Iuo+oieFcrkd0KU5/ueHgwYPw9/fHlClT7D0Up3bYvH+UiyzXA9iomIiI2p74YNOyvRq9EYezyy5zb2D1/izklJpm/vaKDnCaiy4ebmqL9+vXfjkCIQR32HMA7CNVb/7vDXucAabdCG/sGsZiVBMktfcz22WzBu9c5Nheqa0n65frJbXncj2iS2lTBSnz2VGN1a1bN+zfvx8qVZv6Vdsc8x32Il2o+ScbFRMRUVsTH2ze2PzSy/aEEFiyOV3+emRX55gdVef6zsEI8zMVP3ZmFGHd4Tx5hz2tRoX2XAZlF+YFKVfuI1VRrcexXOtF44LyhrtD0uXdkRwF99oloJ9szcDxixzfK/HnCVNDc7UkISGMBSmiS2GVhlqEeUHK1XajYaNiIiJqS5qy096fJwrlTUvig73RKdTnkvdvazQqFW7tEyV//cLqgzhXbGoUHRvkBZWKs0/swbwxt6sWpIora3DXRztgrTsJ20M0X5CPFqO7hwMA9EaBF3461KgWMI2VU1KFk7WbIsQFe7lMX12i5mJBilpE3ZI9jUpCqC/fIImIiBxVZKCnPEPgco3Nl/xRPztqRLcwSE64PKhvTAA61hbp6nYtA9g/yp6CXbwglVtahYn/3W61YMz2EFduZLdwBHm7AwC2nCjA/x3KbbHn3nqyQP53Unv2jyK6HBak6IpV6QzIKDBdCYgI8ISaVxOJiIgclkalQmyQqUl0ZtGFiy79OZxdis3H8gGYdtDtG+2cM4AlScLtyVENbm/BSRPUREHe7nJ/zszzFy55X2eTUVCB2z7YiqO1S8n8PDS446ooRLE9RItx16gsdth7afUhVOkMLfLcf56o7x+VGM7lekSX4/C77JHjO55bDmPtSZurLdcjIiJqi+KDvXE8z7SF/F9nijHcSm+oD/84Jf/7xqQwp16+VnpB3+C23w7nolOoj9wEmWzHTa1CgJcbzlfqXKqp+eHsUtz90U65SBzs4445NyYg1NcDI7qG23l0zqVvTACS2vvicHYZzp6/gCWb0/HIsM5X9JxCCGyrnSHlrlahY4hzLXEmag2cIUVX7Ij5DnssSBERETm8eLMPStaW7eWUVGHVvnMAAC93Na7rFGyzsdnDT/uzGtwmXeR2so26PlJFFTUoq9LZeTStb3dGEe747za5GBUR4IEnRiWyFUYrkSQJk6+OQV2d/f2NJ+T+cc2VUViJrBLTjqSdQn3gpuZHbaLL4V8JXbEj5jvsBbAgRURE5Ogu19h8+dYM6Aym6c+Du4Q4fWPenNoPkebERW4n2wg222nPvLeXM9p4NA93fbQDZVWmmXrxwd7498hEBHi523lkzi0iwBNDE0MBAFU6I179+fAVPV/d7noAkNSey/WIGoMFKbpiljvsedlxJERERNQYgV7uCPRyAwDsP1sCg7G+YVJ5tR5f7jgNwLRZydAuoXYZoy2F+3tAuSCRO5nZl6vstLdqXxb+/sluVOmMAIBu7f3w6I0J8NGys4otjO0VAV8P07H+OS3boil5U5k/NjGcDc2JGoMFKbpidUv2fD008Pd0s/NoiIiIqDHig03L9sqr9ThR208KAL7dlYnS2pka/eKDXGKWRkrPCAhALkpxJzP7C7GYIeWcBanPtp/GP7/eC31tQfiq2EA8NLQTtE4+I9GReLlrcFuf+k0NXlh1CHqDscnPYzQKbDtZWPucasS240V6osZgQYquSEF5NQrKawBwuR4REVFbYrlsz9RHSm8w4qMtZs3MrTQ7d0bJsYGYMagjIrmTmcMwnyE1d81hjFqwGWsPZNtxRC1HCIH31h/HMysPyLs5DuwcjH/cEM++Q3YwoFMQ4mp3Hj2aW4bPt59u8nMczinF+UpTr7MuYb5OvQkEUUviKx5dEcvleixIERERtRXxwQ37SK05kCM39u0e6edSF5uSYwPxfEo3LL4rGc+ndGMxys6yzBpMG4XpnHP653vafFHKaBR4+efDeOv/jsm3je4ejrv7xbKIYScqScLka2Lkr+f9dgyFtc3lG2vriUL530ntuVyPqLFYkKIrcjjbbIe9AE5NJSIiaitigryglkwfgPdmnocQAks2p8vfH8lt5smO1h3Os/i6bknlO+uO22U8LUFvMOKx7/ZbzEKckByF2/pGQZJYjLKnjiE+GNAxCABQWqW3KBg2xp8W/aPY0JyosViQoivCGVJERERtk1ajRmTte/fxvHKsO5yHtHMlAICYdl78UEV2lVNqfefD47nl0DWjx4+9VekMmPHFHny/5ywAQJKAe/rHYmQ3Fn4dxW19o+DhZvp4/PWuMzhQ+3p4OTV6I3aeKgIA+Hu6oT03QyBqNBak6IocqS1ISQDaB/DFl4iIqC3pWNtHSgjgPyvT5NtHdg3jjA2yK2s7HwKA3igwcsFm/H4oF0IIK/dwPGVVOkxbtgu/HcoFYNq9cvrAjrihc4idR0bm/D3dkNLTtJGBEMBzqw42KmP7zhajssYAwDQ7iq+dRI3HghQ1m8EocCzXVJAK9dNCq+GOIERERG1J3U57AJBbauqZEujlhuQ49k8i+1LufGguPb8Cf/90N+5cuqPRs1jspbC8Gncu3YFt6aYeQ1qNCo8M7YzkWP6NOaJhiaEI9zNdZE89fR4r/zp32cewfxRR87EgRc322fYMVOtNU6ZLL+iQevq8nUdERERETWG+016dxHA/aFQ8RST7Uu58GBXoidv6RMqz+gBgW3ohUt7bgke/3YeckoZL/Owtq/gC7vjvNnkprLe7Go+OSEDXCBYtHJVGrcLka6Llr+f+cgTl1fpLPsa8f1QSlzoTNYnG3gOgtmntgWw8v+qQ/PUFnREfbDqJGYM68ooPERFRG3H2fGWD27alF6J3dADfz8nukmMDG+RwVPdwpJ45j+9TzyG/vBpCAN/vOYuf07Lwjxvi8cCgjvDW2v8jzsn8ctz94Q5k1RbKAr3cMHt4AiJcaOfKtqpbhD96Rwfgr8xi5JVV4931x/Hk6CSr962s0WPvGdNF+VBfLYJ8tLYcKlGbx8tf1CwLfm+4w4kE4Kf9WbYfDBERETXLT/uzG9zG93NyZJIk4arYdnhxXDfccVUUvNxNLSOqdEYsXH8Cg9/aiK93noHBaL/+UgfOlWDC4m1yMSrUV4vHRyWyGNWGTLwqGhqVacHox1tOIT2/3Or9dmWch85gyho3giBqOhakqFlOFVQ0uE0ADjldmoiIiKyz9r7N93NqC9zUKozoGo5Xb+mB4UmhUNcWD/LLqvHED2m4aeEf2Hws3+bj2p5eiElLtqOoogYAEB3oicdHJSKYM2falBBfrbwDos4g8NLqQ1bvt9V8uR77RxE1GQtS1Cwdghv2nJBg2hGFiIiI2gZrO5nx/ZzaEh+tBpOujsGLY7uhb0yAfPuRnDJM+Xgn7vl4p7wJT2v77VAupny8U+451DnUB4+N7AJ/Tzeb/HxqWWO6hyPQy/T/3Yaj+Vh3OLfBfcwbmnOGFFHTsSBFzTJreGcA9TufSDBdUR3bK8JeQyIiIqImUu5kxvdzaqvC/Dwwc3An/HtkF8QFecm3bzqWj1ELNuPJH9KQX1bdaj//+9SzmP55KmpqN/zpGemPWcM7w8vd/v2sqHm0bmpMSK5vcP7i6kOo1hvkr4sra3Agy9SwPjrQE74eLDwSNRULUtQso7q3x+K7+qJzmA80KgmRgZ6YObgj+sawASoREVFbodzJjO/n1NYlhPniqTFJuP/6Dmjn7Q4AMArgq51nMPjNDXhv/XFcqDFc5lma5uMtp/Doin1y36prO7TDzCEdodWoW/TnkO1dHReIhDAfAMDpwkp8tOWU/L3t6YUQta3KErlcj6hZJCGE/Tr+ubjS0lL4+/ujpKQEfn5t80WsqKIGe06ft/cwiIiIiIgs1OiN+P1wLn45kI0qnVG+vb2/Bx4b2QXje0dCpVIuWm08IQTm/3YMC9efkG8b2iUUk66Jhkpq/vOSY8k8X4kXVx+CEICXuxrrHx2McH8PPLPyAD7bfhoA8MjQTugZFWDfgZLdxQZ5oXNY2126aY/6BGdIERERERGR03HXqDCmR3u8Or4HBieEoK72lF1ShTnf7sO4RX9ie3rhpZ/kIoxGgedWHbQoRqX0bI/JLEY5nehALwxOCAEAVNYYMHfNYQD1Dc3VkoSENlyEILInFqSIiIiIiMhp+Xm64a5+sXg+pRt6RvrLt6edK8GkJdtx/6e7kZ5f3ujn0xmMmP3tX/h022n5tklXR2Nc70hILEY5pXG9I+HtblqC+eNfWVi9Pwsn8027jscFe8HDjcsziZqDBSkiIiIiInJ6EQGeeGRYZzx6YwKiAz3l2387lIsR8zfj+VUHUVRRc8nnuFBjwD8+3Y0f/8oCAKgk4L7rOmB4Ulirjp3sy0erwS19IuWvH/12n/zvJPaPImo2FqSIiIiIiMhlJLX3wzM3dcXUAXHw9zTtjKY3CizfmoFBb27Aks0nLXZTq1NyQYcpH+/AhqP5AACNSsLMwZ3Qv2OQTcdP9jGwc4hcyKzW1/cku4I2ZEQujwUpIiIiIiJyKSqVhOs7BePV8d0xrlcE3DWmj0VlVXq8+ssRDJ+3Cav3Z6Fu/6f8smpMWrIduzJMm/l4uKkwe3gCekcH2OtXIBtTqSQkxzbcgXTVvmykcpMnombR2HsARERERERE9qB1UyOlVwRu6ByMlX9l4c8TBRAAMosu4KEv9+Lt4KOo0RtxrrhKfoyvhwazhnVGbJC3/QZOdrHbSuFJAvDT/iyrxSoiujTOkCIiIiIiIpcW4OWOqQPi8OzNXZHUvn7HtFMFlRbFKAC4uUd7FqNcVE5JVYPbxEVuJ6LLY0GKiIiIiIgIQHQ7L8wZnoBHhnaCxkpzIAnAHycKbD8wcgjh/h5QpkKqvZ2Imo4FKSIiIiIiolqSJKFnVIDV73E2jGtL6RkBAchFKQmmTIztFWG/QRG1YSxIERERERERKXA2DCklxwZixqCOiAz0hEYlITLQEzMHd0TfGPaPImoONjUnIiIiIiJSSOkZgQ82nZRnwXA2DAGmohQbmBO1DM6QIiIiIiIiUuBsGCKi1sUZUkRERERERFZwNgwRUevhDCkiIiIiIiIiIrIpFqSIiIiIiIiIiMimWJAiIiIiIiIiIiKbYkGKiIiIiIiIiIhsigUpIiIiIiIiIiKyKRakiIiIiIiIiIjIpliQIiIiIiIiIiIim2JBioiIiIiIiIiIbIoFKSIiIiIiIiIisikWpIiIiIiIiIiIyKZYkCIiIiIiIiIiIptiQYqIiIiIiIiIiGyKBalGmDp1qr2HQERERERERETkNFiQaiYhBJ599lm0b98enp6eGD58OI4fP27vYREREREREREROTwWpC6ioKAA99xzD2JiYvDVV1+hU6dOmDBhAmpqagAAb7zxBhYuXIjFixdjx44d8Pb2xsiRI1FVVWXnkRMREREREREROTYWpC5i9uzZ2L59Oz777DOMGTMGS5cuRXx8PIxGI4QQWLBgAZ5++mmMGzcOPXv2xKeffoqsrCysXLnS3kMnIiIiIiIiInJoLEhdxN69ezFlyhQMGjQI/v7+GDJkCF5//XV4eHjg1KlTyMnJwfDhw+X7+/v749prr8W2bdsu+pzV1dUoLS21+I+IiIiIiIiIyNVo7D0AR3Xddddh2bJl6NWrV4Pv5eTkAADCwsIsbg8LC5O/Z83cuXPxwgsvtOxA7czPQ4N+HYPsPQwiIiIiIiIiu9GoJHsPoc1hQeoi5s2bh1dffRWzZ8/GyZMn8ddff2H69OmYPn16s5/zySefxJw5c+SvS0tLER0d3RLDtRuNWgUfNSfaEREREREREVHjsSB1Ed7e3njllVfwyiuvYPz48Rg9ejRmz54NlUolL9XLzc1F+/bt5cfk5uaid+/eF31OrVYLrVbb2kMnIiIiIiIiInJonNrSCAEBAXjggQcwevRo/PHHH+jQoQPCw8Oxbt06+T6lpaXYsWMH+vfvb8eREhERERERERE5PhakLmL27NnYtGkTSkpKYDAYsGHDBmzatAnJycmQJAmzZs3Cyy+/jFWrViEtLQ1TpkxBREQExo8fb++hExERERERERE5NC7Zu4iYmBjMmTMHx48fR0VFBTZu3Ih7770XDz/8MADg3//+NyoqKvCPf/wDxcXFuP7667F27Vp4eHjYeeRERERERERERI5NEkIIew/C0U2dOhXLly9v8ectLS2Fv78/SkpK4Ofn1+LPT0RERERERER0OfaoT3DJHhERERERERER2RQLUo3QGrOjiIiIiIiIiIhcFQtSRERERERERERkUyxIERERERERERGRTbEgRURERERERERENsWCFBERERERERER2RQLUkREREREREREZFMsSBERERERERERkU1p7D0AVyaEAACUlpbaeSRERERERERE5Krq6hJ1dQpbYEHKjsrKygAA0dHRdh4JEREREREREbm6srIy+Pv72+RnScKW5S+yYDQakZWVBV9fX0iSZO/hNEtpaSmio6ORmZkJPz8/ew+HHAAzQUrMBFnDXJASM0HWMBekxEyQEjNR70qOhRACZWVliIiIgEplm+5OnCFlRyqVClFRUfYeRovw8/Nz+T9+ssRMkBIzQdYwF6TETJA1zAUpMROkxEzUa+6xsNXMqDpsak5ERERERERERDbFghQREREREREREdkUC1J0RbRaLZ577jlotVp7D4UcBDNBSswEWcNckBIzQdYwF6TETJASM1GvrR0LNjUnIiIiIiIiIiKb4gwpIiIiIiIiIiKyKRakiIiIiIiIiIjIpliQIiIiIiIiIiIim2JBioiIiIiIiIiIbIoFKSIiIiJyGUaj0d5DICIiIrAgRQ6MG0CSUl5enr2HQA6OHzRJiZkgADhw4ADuuOMOAIBKxdNfMuG5JpnjeSY1hiufV7TGa6Yk+EpMDub8+fPw8PCAp6cnhBCQJMneQyIHsHfvXiQnJ2Pjxo0YOHCgvYdDDuDUqVPYsmULioqK0LVrV9x4440AwNcNF3by5El88sknKC4uRmxsLB599FF7D4kcwL59+zBs2DAUFRVh1apVuPnmm/k64eJ4rklKPM8ka3iuaVJeXg6tVgs3N7cW/915iYgcyuHDhzFixAi8+eabqKyshCRJvHpF2LdvHwYNGoTZs2fzJIEAAGlpabjmmmvwww8/4P3338cTTzyBIUOGoLS0lK8bLiotLQ39+/fH4cOHsX//fnz55ZeYN2+evYdFdrZv3z7069cPd911F/r164cVK1YAgEt9kCBLPNckJZ5nkjU81zQ5fPgwbrnlFnzzzTeoqalp8d+dBSlyGGfOnMHkyZORmZmJX3/9FYsWLeKJAuHAgQMYMGAA/vnPf+Ltt9+GEALHjx/Hpk2bkJ2dbe/hkR0UFRVhypQpuO+++/C///0Pu3fvxqOPPopNmzZhzJgxyM7OhiRJLj2l2tUcP34cKSkpuP/++7FixQr89NNPiIqKQnV1tcX9mAnXsnfvXvTv3x+zZs3CggULMGfOHKxcuRIbN26099DITniuSUo8zyRreK5pcvr0adx2223YvHkzFi1ahFWrVrV4UYoFKXIIQgisWbMG4eHh+Pnnn9GzZ0+sWLHC4kTB2f/gqaHq6mo8/fTTuHDhAl566SUAwM0334yJEydiyJAhSElJwaxZs+w7SLK5rKws6PV63HfffQAAX19fDB06FN26dUN6ejpuuukmAOwT4yoMBgO+/PJLXH/99Xj66acBmDIREhKCbdu24e6778bMmTOh1+uhUqn4XuIi8vPzcdddd+HBBx/E3LlzAQA9e/ZEbGwsNm3aBIAFSlfDc01S4nkmXQzPNU3nV99//z06deqEnTt3IiAgAK+++mqLF6Wc9whSmyJJEsaOHYsHHngAycnJ+OCDD5CcnCyfKFRUVEClUvHqlYtxd3fHU089haSkJFx77bW48cYboVar8eabbyItLQ0pKSnYuHEjXnzxRXsPlWysrKwMaWlp8tclJSVQqVSYP38+iouL8frrr9txdGRLarUad999Nx599FF4enoCAF577TUsW7YMnTt3RkhICDZs2ID+/ftDCOHUJ49Uz93dHUuWLMGbb74p35aQkIDx48djwYIFyMnJYRZcDM81SYnnmXQprn6uqVarMXToUEyZMgW9evXCzz//jLCwMLkoVV1d3SJFKTY1J4dhNBotTg71ej0efvhhpKamYsKECXjwwQfh5eWF5cuXY+rUqfYbKNmU0WjEvn37cM8998DNzQ0//vgjoqKiAAAXLlzAAw88gKysLPzyyy9wd3e382jJFvLy8nDnnXfC29sb/fv3R/fu3XH33Xdj2rRpmDdvHiZNmgRPT08sW7bM3kOlVlbXWNO8wWZmZiYmTpyIZ599FqNGjQIArF+/HhMnTsTKlStx3XXX2XPIZAPK8wnz206ePInx48fjb3/7Gx5//HEA7CflSpTNeHmuSXXnmVOnToVGo+F5JgEwzbKdPHkyfHx80K9fP5c919TpdHBzc5O/rqmpwbhx45Cbm4unnnoK48aNkz+fjRs3rlk/Q9NSgyVqquzsbBw9ehQajQadOnVCeHi4/D29Xg+NRoOFCxfikUcewYoVK2A0GpGeno6PPvoIQ4YMQWxsrB1HT63FPBcdO3ZE+/bt0bt3b3z++efIysqSc2IwGODp6YkuXbrg4MGDnGbvxMwzER8fj4iICLz77rt49tlnsWzZMkiShIceekiebh8aGopjx47ZedTUmqqrq6HVagE0/IAZHR2NNWvWwN/fX/6eJEkICQmxeJ8h51OXC2sFproCVXx8PLp27Yrvv/8eTzzxBADX2y3JlRQVFeHcuXMAgKioKAQGBsrFSYPBwHNNF2SeicjISLRr1w49evTAZ599huzsbJ5nuijzXERERCAkJATvvPMOnnvuOXzyyScQQrjEuWZBQQEyMzPh5eWF0NBQi9dMvV4Pd3d3rFy5EuPHj8err74Kg8GADRs2YNWqVbj66qsRERHR9B8qiOxg3759IjY2VnTq1ElERESI8PBw8d1334nq6mr5PjqdTv7f6dOnC61WK/z8/MSePXvsNWxqZdZysWLFCqHX64UQQhiNxgaPmTZtmpg6daqcF3Iu1jLxzTffCCGEqKysFKWlpSIjI0O+v9FoFLfddpt49NFH7TVkamWHDh0S119/vdiwYYMQwvrrgvK2xx9/XAwePFgUFRXZYohkB43JhcFgEEIIcfToUdGuXTvxwQcf2HKIZGP79+8Xffv2FV26dBHR0dFi7Nix4vTp0xb3qTu/4Lmma7CWibpzCL1eL79GmON5pvNT5iIlJUWcPHlSCCFESUmJKC0ttXjtcNZzzX379omEhATRsWNHERUVJZKTk8W2bdss7lP3d1BdXS3GjBkj3NzchLe3t0hNTW32z+XiebK5/Px8TJgwAZMmTcLmzZuxatUqTJ48GRMnTsT8+fNRVlYGANBoNPLVK5VKBS8vL2zduhV9+vSx829AreFiuZg0aRLeeustlJeXW1zBLioqwlNPPYXVq1fjscceg0bDCZ/O5mKZmDx5MubOnQudTgdfX1/5Cvbx48fxn//8B+vXr8ff//53O4+eWkNGRgbGjh2Lw4cPy7vdWOtfUPdacfbsWTzxxBNYsmQJ3nnnHQQGBtpj2NTKGpuLullSERERSEhIwNatW1FTU2OPIVMrO3bsGIYNG4Zhw4bhu+++w8svv4zy8nKsWLECAORsqNVqGI1Gnmu6gItl4rvvvgNgen0wX+rL80zXYC0XFRUV+OGHHwAAPj4+8PX1RUxMDADnPdfMyclBSkoKxo8fj19++QXvvvsuOnfujIEDB+Lrr7+W71f3+dzd3R2xsbHw9fXFjh070Ldv3+b/8ObX0IiaJz09XXTp0kXs3r3b4vb58+cLSZLEu+++K4Sov5L58ccfC0mSeLXKyTUlF2vWrBH33HOPiIqKYi6cWFMykZubK1588UURExMj9u7da4fRUmurqqoSDz30kLj11lvFV199Je644w7Rs2dPsXHjRiGE6Yql+ayYP//8Uzz00EMiISGBmXBijcmFNWvWrBGHDx+25VDJRsrLy8XkyZPFfffdZ3H71KlTxfXXX2/1MTzXdG5NzcTatWt5nukCmpqLvLw8pz3X3Lt3r+jevbs4deqUfFtlZaX417/+Jdzd3cXq1auFEPWfzxctWtRir5mcIUU2V1pailOnTslXp3Q6HQBg1qxZePXVVzF79mykpqbKVymmTZuG9PR0Xq1yck3JRZ8+fTBw4EBs3ryZuXBiTclEUFAQpk2bhu3bt6N37952HDW1Fq1WizFjxmDUqFGYNGkSZs+ejS5duuCRRx6RZ8SYz6Ls2rUrUlJS8PvvvzMTTqwxuTBX93oyatQoJCYm2mPI1MqqqqoQHh6OoUOHAjD1AgKAW2+9FQaDAXq9vkE/IJ5rOremZqJXr148z3QBTc1FYGAgpk6d6pTnmiUlJTh48KD8Hmk0GuHp6Yk33ngD999/P+68804cP35c/nw+ceJEnDhxokX+PrjLHtlFXXf+VatWITQ0FHq9Hmq1GgAwduxYtGvXDh9++CGEENzRwoVcLheBgYFYunQptFotm9C6iMa+Vmg0GubBBf35559YuHAhjhw5goULF2LQoEGorq7GiRMn0K1bN3sPj+zkYrlIT09HUlKSvYdHNrB7925cddVVAOqb1q9ZswZPPvkkdu7cCTc3N0iShPPnz3Mpr4tobCYKCwsRFBRkdbdOcj6NzUVRURHatWtn59G2HoPBgKFDh6J9+/Z4//330a5dO/lv4Ny5c7jzzjsxbNgwPPPMMxBCtOjfBv/KyC4eeOABuLm54bHHHkNBQQE0Go38IhAeHo6CggK4ubmxGOViLpeLwsJCeWctFh9cQ2NfK5gH11J3xfK6667DI488gsTERDzyyCNYt24dHnvsMQwbNkzuR0iu43K5GDJkCHPhIpQfMAGgoqIC5eXlUKvVkCQJTz/9NEaNGsU+Yi6isZkYM2YMampqeF7hIhqbi9GjR6OmpqZBb0JnoVarMXHiRGRkZGDhwoUoLS2Vi06RkZHw8fHBkSNHIElSixdq2Z2N7GL06NE4efIkPv30U8yYMQPvvfcewsLCAJiaCgYEBKCmpoYfNF0Mc0FKzASZqzthVKlU0Ol0cHNzw3XXXQcAePfddzFy5Ej4+vri119/ha+vr51HS7bCXJBS3ZV9SZJgMBigVqvh5+cHT09PqNVqPP3005g3bx42b97Mi58ugpkga5iL+vfQGTNm4OTJk/jxxx9x4cIF/Oc//4Gfnx8AICgoCIGBgTAYDPLxaiksSJFN1f2hV1dX4+GHH0ZAQAA+/PBDdOvWDWPGjEFJSQnWr1+PrVu3Ou0fPTXEXJASM0FKdZmomzbv5uYmn0Rdd911eO211+Dr64stW7ZwuZ4LYS5ISZmJumXeWq0WgYGBmDNnDhYtWoStW7ciOTnZzqMlW2AmyBrmwsRoNEKtVkMIgbfffhsvvfQSfv75Z/z4448YO3YsMjMzsXr1amzfvl0+Ri2JS/ao1SinNNb90Z8+fRqxsbH44YcfcPfdd2PZsmWYNWsWACAuLg47duxAjx497DBisgXmgpSYCVK6VCa6d++OL774AgDkK5pz587F+vXrsWHDBhYdnBhzQUqNzQQAFBYWYsuWLVi6dCm2bdvm1B8wXRkzQdYwFw0JISyOQ48ePbBx40Y888wzeP311zFixAikpaVBq9Vi27Ztrfc+esX79BEpFBYWyv9Wbrd85swZERERIaZPny50Op2th0Z2xFyQEjNBSo3NhPJ7a9asEYcOHbLJGMn2mAtSak4m9u3bJ0aPHi0OHjxos3GS7TATZA1zYXLmzBmxfPlyMX/+fLFu3TohRP3xyMjIEJGRkeKBBx5ocM5tNBqFwWBo1bFxlz1qUYcOHUKvXr3w4IMPYsGCBQAsm8T95z//wYULF/D222/Ltwnulub0mAtSYiZIqTmZIOfHXJDSlWSiuLgYAQEBNh4xtTZmgqxhLkzS0tKQkpKCyMhInD9/HidOnMDy5ctx5513QgiBe++9F+7u7li8eLFdzrlZkKIWk5WVhVtuuQV6vR5Hjx7F/fffj/nz5wOoD7Ver4dGw9ZlroS5ICVmgpSYCbKGuSCl5maCFzScFzNB1jAXJqdOncKQIUMwadIkPP/88ygrK8N7772H1atX45dffkFYWJi8GYi98B2cWoQQAhs2bEBsbCxmzZqF06dPY9q0aZAkCfPmzeNJo4tiLkiJmSAlZoKsYS5I6Uoy4UwfMKkeM0HWMBcmer0eH3/8Mfr06YPnnnsOHh4e8PDwwIABA7B06VL5fvYsRgEsSFELkSQJAwcOhK+vLwYMGIABAwbIUwCFEJg/fz40Go3TVZ3p0pgLUmImSImZIGuYC1JiJkiJmSBrmAsTjUaDnj17wtPTE56envLt1157LTQaDQoKChAWFmbHEZqwIEUtJjo6GtHR0fLXEydOhCRJFhVpg8GAb775Br169UL37t3tOFqyFeaClJgJUmImyBrmgpSYCVJiJsga5sIkJSUFHh4eAOqXI9bNDNPpdPL9du3ahauvvtouY2RBiprtcutN1Wo1JkyYAACYNm0aANMWmx988AFOnDhhkzGS7TEXpMRMkBIzQdYwF6TETJASM0HWMBcmyuNQV4wCIC9VLC8vh16vh5eXFwDgySefxOuvv468vDwEBwfbfMxokb36yOUcOHBAjB07tlHbYer1evHZZ58JSZJEYGCg2LVrlw1GSPbAXJASM0FKzARZw1yQEjNBSswEWcNcmDTmOBiNRlFQUCAiIiJERkaGeOGFF4SPj4/YuXOnDUdqiQUparJTp06J+Ph4IUmS6N27tzh69Ogl728wGMR9990n/Pz8xKFDh2w0SrI15oKUmAlSYibIGuaClJgJUmImyBrmwqQpx6GyslJ0795djBgxQri7u4vdu3fbcKQNqWw/J4vasurqanzyySfo1asXdu7cCXd3d4wfPx7Hjh276GN+/fVXrF+/HuvWrUNSUpINR0u2wlyQEjNBSswEWcNckBIzQUrMBFnDXJg05TgIIZCfn4+DBw9i06ZN2LVrF5KTk+0w6nqSEELYdQTUphiNRqxcuRJGoxG33347iouLMXLkSJSVlWHlypVISEho8JisrCyoVCqEh4fbYcRkC8wFKTETpMRMkDXMBSkxE6TETJA1zIVJc47DW2+9hTFjxqBr1652GLElFqSoyQwGA9Rqtfx1YWEhxowZg7KyMvz444/o3Lkz9Ho9du7cib59+1o0UyPnxVyQEjNBSswEWcNckBIzQUrMBFnDXJg05ThcffXV0Gg0kCTJjiOux4IUNZuo3ToSAAoKCnDTTTehrKwMK1aswHvvvYc9e/bgl19+QVBQkJ1HSrbEXJASM0FKzARZw1yQEjNBSswEWcNcmFzuOOzevRu//vor2rVrZ+eR1mNBiprMPOjmXxcWFiIlJQU7duyAVqvF5s2bcdVVV9lxpGRLzAUpMROkxEyQNcwFKTETpMRMkDXMhUlbPg5sak6XZDQaLb42GAyQJAmlpaUoLi4GADn8QUFBSExMRGBgIHbv3u1wYaeWw1yQEjNBSswEWcNckBIzQUrMBFnDXJg423FgQYqsKigoAACoVCoYDAYA9WtTMzIykJSUhG3btsn3F0Lg3XffxfLly/Hbb785RIM0annMBSkxE6TETJA1zAUpMROkxEyQNcyFibMeBxakqIFjx44hPj4e//jHPwAAarVaDntmZiauvvpqjBkzBqNGjbJ4XI8ePXDs2DH06dPHHsOmVsZckBIzQUrMBFnDXJASM0FKzARZw1yYOPNxYEGKGjh06BA8PT2RlpaGBx54AIAp9DU1NVi1ahXuvvtuLF682GKdqiRJGDx4MDp16mSvYVMrYy5IiZkgJWaCrGEuSImZICVmgqxhLkyc+TiwIEUNaLVaBAQEYPz48di2bRumT58OAHB3d8e4ceMwb948i20lyTUwF6TETJASM0HWMBekxEyQEjNB1jAXJs58HDT2HgA5nh49eiA5ORl///vf4e7ujuXLl2POnDkoKSnBNddcg3vvvRdubm72HibZGHNBSswEKTETZA1zQUrMBCkxE2QNc2Hi1MdBEClUVFSInj17ir1794qKigqxZMkSERQUJCRJEvv37xdCCKHX6+08SrI15oKUmAlSYibIGuaClJgJUmImyBrmwsSZjwOX7JEFnU4HrVaL8PBwlJeXw8vLC+vWrYNOp0OnTp3w4YcfAkCbnRJIzcNckBIzQUrMBFnDXJASM0FKzARZw1yYOPtx4JI9F5aVlYU9e/agpqYGcXFx6Nu3rzzVLzk5GSdOnMCSJUuwefNm/PTTT0hLS8Nrr70GjUaDt99+286jp9bCXJASM0FKzARZw1yQEjNBSswEWcNcmLjicWBBykWlpaVh/PjxCA4ORnp6OuLi4vD444/j9ttvB2BqnHbvvfciLi4Oq1evRt++fdGzZ0+oVCqMHDnSzqOn1sJckBIzQUrMBFnDXJASM0FKzARZw1yYuOxxsPeaQbK9EydOiKioKPHvf/9bFBcXi927d4t77rlH3HvvvUKn0wkhhNDpdGLmzJli586dQgghjEajEEIIg8Fgt3FT62IuSImZICVmgqxhLkiJmSAlZoKsYS5MXPk4sCDlYqqrq8WcOXPEHXfcIaqrq+XbP/roIxEUFCQKCgrsODqyF+aClJgJUmImyBrmgpSYCVJiJsga5sLE1Y8Dl+y5GKPRiKioKCQlJcHd3R1CCEiShAEDBsDHxwc6nc7qY1Qq9r93ZswFKTETpMRMkDXMBSkxE6TETJA1zIWJqx8HFqRcjIeHB8aPH48OHTpY3B4QEAA3NzeLwO/duxd9+vRxmrDTxTEXpMRMkBIzQdYwF6TETJASM0HWMBcmrn4cnOc3oYvKzs7Gzp07sXbtWhiNRjnsBoMBkiQBAEpKSnD+/Hn5Mc8++yyGDRuGwsJCCCHsMm5qXcwFKTETpMRMkDXMBSkxE6TETJA1zIUJj4MZW68RJNvat2+fiI2NFQkJCcLf318kJiaKL7/8UhQWFgoh6puhHT16VISEhIiioiLx0ksvCU9PT7F79257Dp1aEXNBSswEKTETZA1zQUrMBCkxE2QNc2HC42CJBSknlpeXJxITE8VTTz0lTp48Kc6dOycmTpwokpKSxHPPPSfy8vLk++bm5oo+ffqIiRMnCnd3d6cMO5kwF6TETJASM0HWMBekxEyQEjNB1jAXJjwODbEg5cQOHjwo4uLiGoT38ccfFz169BBvvPGGqKioEEIIcejQISFJkvD09BR79+61w2jJVpgLUmImSImZIGuYC1JiJkiJmSBrmAsTHoeG2EPKiel0Ouj1elRWVgIALly4AAB47bXXMGTIEHzwwQc4ceIEACAwMBAzZ87Enj170Lt3b3sNmWyAuSAlZoKUmAmyhrkgJWaClJgJsoa5MOFxaEgSwpk6YpHSNddcAx8fH6xfvx4AUF1dDa1WCwC4+uqr0alTJ3z11VcAgKqqKnh4eNhtrGQ7zAUpMROkxEyQNcwFKTETpMRMkDXMhQmPgyXOkHIiFRUVKCsrQ2lpqXzbf//7Xxw8eBB33nknAECr1UKv1wMABg4ciIqKCvm+zh52V8VckBIzQUrMBFnDXJASM0FKzARZw1yY8DhcHgtSTuLQoUO49dZbMWjQICQlJeGLL74AACQlJeGdd97Bb7/9hgkTJkCn00GlMv3fnpeXB29vb+j1eufaOpJkzAUpMROkxEyQNcwFKTETpMRMkDXMhQmPQ+No7D0AunKHDh3CwIEDMWXKFFx11VVITU3FtGnT0LVrV/Tp0wdjx46Ft7c3Zs6ciZ49eyIxMRHu7u74+eefsX37dmg0jIEzYi5IiZkgJWaCrGEuSImZICVmgqxhLkx4HBqPPaTauKKiIkyePBmJiYl455135NuHDBmCHj16YOHChfJtZWVlePnll1FUVAQPDw/MmDEDXbt2tcewqZUxF6TETJASM0HWMBekxEyQEjNB1jAXJjwOTeM6pTcnpdPpUFxcjNtvvx0AYDQaoVKp0KFDBxQVFQEAhBAQQsDX1xevv/66xf3IOTEXpMRMkBIzQdYwF6TETJASM0HWMBcmPA5N43q/sZMJCwvD559/jhtuuAEAYDAYAACRkZFyoCVJgkqlsmimJkmS7QdLNsNckBIzQUrMBFnDXJASM0FKzARZw1yY8Dg0DQtSTqBz584ATFVVNzc3AKaqa15ennyfuXPn4sMPP5Q7+Ltq4F0Jc0FKzAQpMRNkDXNBSswEKTETZA1zYcLj0HhcsudEVCoVhBBymOsqsM8++yxefvll7N2716UapJEJc0FKzAQpMRNkDXNBSswEKTETZA1zYcLjcHmcIeVk6nrUazQaREdH46233sIbb7yB3bt3o1evXnYeHdkLc0FKzAQpMRNkDXNBSswEKTETZA1zYcLjcGmuXY5zQnVVVzc3NyxduhR+fn7YsmUL+vbta+eRkT0xF6TETJASM0HWMBekxEyQEjNB1jAXJjwOl8YZUk5q5MiRAICtW7fiqquusvNoyFEwF6TETJASM0HWMBekxEyQEjNB1jAXJjwO1kmibg4ZOZ2Kigp4e3vbexjkYJgLUmImSImZIGuYC1JiJkiJmSBrmAsTHoeGWJAiIiIiIiIiIiKb4pI9IiIiIiIiIiKyKRakiIiIiIiIiIjIpliQIiIiIiIiIiIim2JBioiIiIiIiIiIbIoFKSIiIiIiIiIisikWpIiIiIiIiIiIyKZYkCIiIiIiIiIiIptiQYqIiIjIAUydOhWSJEGSJLi5uSEsLAw33ngjPv74YxiNxkY/z/LlyxEQENB6AyUiIiJqASxIERERETmIUaNGITs7GxkZGVizZg2GDBmCf/7zn7j55puh1+vtPTwiIiKiFsOCFBEREZGD0Gq1CA8PR2RkJPr27YunnnoKP/74I9asWYPly5cDAObNm4cePXrA29sb0dHRmDlzJsrLywEAGzduxLRp01BSUiLPtnr++ecBANXV1fjXv/6FyMhIeHt749prr8XGjRvt84sSERGRy2NBioiIiMiBDR06FL169cIPP/wAAFCpVFi4cCEOHjyITz75BOvXr8e///1vAMCAAQOwYMEC+Pn5ITs7G9nZ2fjXv/4FAHjooYewbds2fP3119i/fz8mTJiAUaNG4fjx43b73YiIiMh1SUIIYe9BEBEREbm6qVOnori4GCtXrmzwvUmTJmH//v04dOhQg+999913mD59OgoKCgCYekjNmjULxcXF8n3OnDmD+Ph4nDlzBhEREfLtw4cPxzXXXINXX321xX8fIiIiokvR2HsARERERHRpQghIkgQA+P333zF37lwcOXIEpaWl0Ov1qKqqQmVlJby8vKw+Pi0tDQaDAQkJCRa3V1dXIygoqNXHT0RERKTEghQRERGRgzt8+DA6dOiAjIwM3HzzzZgxYwZeeeUVtGvXDlu2bMF9992HmpqaixakysvLoVarkZqaCrVabfE9Hx8fW/wKRERERBZYkCIiIiJyYOvXr0daWhpmz56N1NRUGI1GvP3221CpTK1Av/32W4v7u7u7w2AwWNzWp08fGAwG5OXl4YYbbrDZ2ImIiIguhgUpIiIiIgdRXV2NnJwcGAwG5ObmYu3atZg7dy5uvvlmTJkyBQcOHIBOp8O7776LlJQU/Pnnn1i8eLHFc8TFxaG8vBzr1q1Dr1694OXlhYSEBPztb3/DlClT8Pbbb6NPnz7Iz8/HunXr0LNnT9x00012+o2JiIjIVXGXPSIiIiIHsXbtWrRv3x5xcXEYNWoUNmzYgIULF+LHH3+EWq1Gr169MG/ePLz++uvo3r07vvjiC8ydO9fiOQYMGIDp06dj4sSJCAkJwRtvvAEAWLZsGaZMmYJHH30UXbp0wfjx47Fr1y7ExMTY41clIiIiF8dd9oiIiIiIiIiIyKY4Q4qIiIiIiIiIiGyKBSkiIiIiIiIiIrIpFqSIiIiIiIiIiMimWJAiIiIiIiIiIiKbYkGKiIiIiIiIiIhsigUpIiIiIiIiIiKyKRakiIiIiIiIiIjIpliQIiIiIiIiIiIim2JBioiIiIiIiIiIbIoFKSIiIiIiIiIisikWpIiIiIiIiIiIyKZYkCIiIiIiIiIiIpv6f5wtvaoA/97KAAAAAElFTkSuQmCC", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "daily_metrics = metrics.groupby(\"date\").agg({\"ordered_units\": \"sum\", \"revenue\": \"sum\"}).reset_index()\n", "plot_daily_metrics_trend(daily_metrics)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "How do shoppers move through the purchase journey?" ] }, { "cell_type": "code", "execution_count": 13, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:51.264213Z", "iopub.status.busy": "2026-03-01T03:45:51.264049Z", "iopub.status.idle": "2026-03-01T03:45:51.353167Z", "shell.execute_reply": "2026-03-01T03:45:51.352395Z" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plot_conversion_funnel(metrics)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "Things to explore\n", "
    \n", "
  • What fraction of impressions become visits?
    • \n", "
  • What fraction of visits add to cart?
    • \n", "
  • What fraction of cart adds become orders?
    • \n", "
  • What is the overall conversion rate?
    • \n", "
\n", "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Enrich Product Data\n", "\n", "Real-world e-commerce teams constantly face a critical question: **Does improving product content quality increase sales?** To help you explore this with known ground truth, the simulator lets you inject treatment effects into a subset of products starting at a specific date, creating synthetic experiments where you know the true causal impact.\n", "\n", "The enrichment process operates through two coordinated mechanisms:\n", "\n", "1. **Content Enhancement**: Generate upgraded product copy with premium brand positioning, compelling descriptions, and detailed feature highlights\n", "2. **Sales Impact**: Apply the corresponding sales boost that this improved content would realistically produce\n", "\n", "The **ENRICHMENT** configuration controls the treatment parameters. The `effect_size` controls the magnitude of the boost (0.5 = 50% increase), `enrichment_fraction` determines what share of products receive treatment (1.0 = 100%), and `enrichment_start` sets when the intervention begins. The `ramp_days` parameter models a gradual rollout where the effect linearly increases over the specified number of days, mimicking real-world deployment patterns." ] }, { "cell_type": "code", "execution_count": 14, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:51.354914Z", "iopub.status.busy": "2026-03-01T03:45:51.354735Z", "iopub.status.idle": "2026-03-01T03:45:51.470674Z", "shell.execute_reply": "2026-03-01T03:45:51.469731Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "IMPACT:\r\n", " FUNCTION: product_detail_boost\r\n", " PARAMS:\r\n", " effect_size: 0.5\r\n", " ramp_days: 0\r\n", " enrichment_fraction: 1.0\r\n", " enrichment_start: \"2024-11-15\"\r\n", " seed: 42\r\n", " backend: \"mock\"\r\n" ] } ], "source": [ "! cat \"config_tutorial_enrichment.yaml\"" ] }, { "cell_type": "code", "execution_count": 15, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:51.472918Z", "iopub.status.busy": "2026-03-01T03:45:51.472719Z", "iopub.status.idle": "2026-03-01T03:45:51.495765Z", "shell.execute_reply": "2026-03-01T03:45:51.494968Z" } }, "outputs": [ { "data": { "text/html": [ "
def product_detail_boost(metrics: list, **kwargs) -> tuple:\n",
       "    """\n",
       "    Product detail regeneration and metrics boost for enrichment experiments.\n",
       "\n",
       "    Selects a fraction of products for treatment, regenerates their product\n",
       "    details (title, description, features) while preserving brand/category/price,\n",
       "    and applies metrics boost effect.\n",
       "\n",
       "    Args:\n",
       "        metrics: List of metric record dictionaries\n",
       "        **kwargs: Parameters including:\n",
       "            - job_info: JobInfo for saving product artifacts (required for saving)\n",
       "            - products: List of product dictionaries (required for product details)\n",
       "            - effect_size: Percentage increase in ordered units (default: 0.5)\n",
       "            - ramp_days: Number of days for ramp-up period (default: 7)\n",
       "            - enrichment_fraction: Fraction of products to enrich (default: 0.3)\n",
       "            - enrichment_start: Start date of enrichment (default: "2024-11-15")\n",
       "            - seed: Random seed for product selection (default: 42)\n",
       "            - prompt_path: Path to custom prompt template file (optional)\n",
       "            - backend: Backend to use for regeneration ("mock" or "ollama", default: "mock")\n",
       "\n",
       "    Returns:\n",
       "        Tuple of (treated_metrics, potential_outcomes_df):\n",
       "            - treated_metrics: List of modified metric dictionaries with treatment applied\n",
       "            - potential_outcomes_df: DataFrame with Y0_revenue and Y1_revenue for all products\n",
       "    """\n",
       "    job_info = kwargs.get("job_info")\n",
       "    products = kwargs.get("products")\n",
       "    effect_size = kwargs.get("effect_size", 0.5)\n",
       "    ramp_days = kwargs.get("ramp_days", 7)\n",
       "    enrichment_fraction = kwargs.get("enrichment_fraction", 0.3)\n",
       "    enrichment_start = kwargs.get("enrichment_start", "2024-11-15")\n",
       "    seed = kwargs.get("seed", 42)\n",
       "    prompt_path = kwargs.get("prompt_path")\n",
       "    backend = kwargs.get("backend", "mock")\n",
       "    quality_boost = kwargs.get("quality_boost", 0.0)\n",
       "\n",
       "    rng = np.random.default_rng(seed)\n",
       "\n",
       "    # 1. Save original product details\n",
       "    if job_info and products:\n",
       "        job_info.save_df("product_details_original", pd.DataFrame(products))\n",
       "\n",
       "    # 2. Select treatment products\n",
       "    if products:\n",
       "        unique_product_ids = sorted(set(p.get("product_identifier", p.get("product_id")) for p in products))\n",
       "    else:\n",
       "        unique_product_ids = sorted(set(record["product_id"] for record in metrics))\n",
       "\n",
       "    n_treatment = int(len(unique_product_ids) * enrichment_fraction)\n",
       "    treatment_ids = set(rng.choice(unique_product_ids, size=n_treatment, replace=False))\n",
       "\n",
       "    # 3. Regenerate product details for treatment products\n",
       "    if products and job_info:\n",
       "        updated_products = _regenerate_product_details(\n",
       "            products, treatment_ids, prompt_path, backend, seed, quality_boost\n",
       "        )\n",
       "        job_info.save_df("product_details_enriched", pd.DataFrame(updated_products))\n",
       "\n",
       "    # 4. Apply metrics boost effect and calculate potential outcomes\n",
       "    treated_metrics = []\n",
       "    potential_outcomes = {}  # {(product_id, date): {'Y0_revenue': x, 'Y1_revenue': y}}\n",
       "    start_date = datetime.strptime(enrichment_start, "%Y-%m-%d")\n",
       "\n",
       "    for record in metrics:\n",
       "        record_copy = copy.deepcopy(record)\n",
       "        product_id = record_copy.get("product_id", record_copy.get("product_identifier"))\n",
       "        record_date_str = record_copy["date"]\n",
       "        record_date = datetime.strptime(record_date_str, "%Y-%m-%d")\n",
       "\n",
       "        is_enriched = product_id in treatment_ids\n",
       "        record_copy["enriched"] = is_enriched\n",
       "\n",
       "        # Calculate Y(0) - baseline revenue (no treatment)\n",
       "        y0_revenue = record_copy["revenue"]\n",
       "\n",
       "        # Calculate Y(1) - revenue if treated (for ALL products, with ramp-up)\n",
       "        unit_price = record_copy.get("unit_price", record_copy.get("price"))\n",
       "        if record_date >= start_date:\n",
       "            days_since_start = (record_date - start_date).days\n",
       "            ramp_factor = 1.0 if ramp_days <= 0 else min(1.0, days_since_start / ramp_days)\n",
       "            adjusted_effect = effect_size * ramp_factor\n",
       "\n",
       "            original_quantity = record_copy["ordered_units"]\n",
       "            boosted_quantity = int(original_quantity * (1 + adjusted_effect))\n",
       "            y1_revenue = round(boosted_quantity * unit_price, 2)\n",
       "        else:\n",
       "            # Before treatment start, Y(1) = Y(0)\n",
       "            y1_revenue = y0_revenue\n",
       "\n",
       "        # Store potential outcomes for ALL products\n",
       "        key = (product_id, record_date_str)\n",
       "        potential_outcomes[key] = {"Y0_revenue": y0_revenue, "Y1_revenue": y1_revenue}\n",
       "\n",
       "        # Apply factual outcome (only for treated products)\n",
       "        if is_enriched and record_date >= start_date:\n",
       "            record_copy["ordered_units"] = boosted_quantity\n",
       "            record_copy["revenue"] = y1_revenue\n",
       "\n",
       "        treated_metrics.append(record_copy)\n",
       "\n",
       "    # Build potential outcomes DataFrame\n",
       "    potential_outcomes_df = pd.DataFrame(\n",
       "        [\n",
       "            {\n",
       "                "product_identifier": pid,\n",
       "                "date": d,\n",
       "                "Y0_revenue": v["Y0_revenue"],\n",
       "                "Y1_revenue": v["Y1_revenue"],\n",
       "            }\n",
       "            for (pid, d), v in potential_outcomes.items()\n",
       "        ]\n",
       "    )\n",
       "\n",
       "    return treated_metrics, potential_outcomes_df\n",
       "
\n" ], "text/latex": [ "\\begin{Verbatim}[commandchars=\\\\\\{\\}]\n", "\\PY{k}{def}\\PY{+w}{ }\\PY{n+nf}{product\\PYZus{}detail\\PYZus{}boost}\\PY{p}{(}\\PY{n}{metrics}\\PY{p}{:} \\PY{n+nb}{list}\\PY{p}{,} \\PY{o}{*}\\PY{o}{*}\\PY{n}{kwargs}\\PY{p}{)} \\PY{o}{\\PYZhy{}}\\PY{o}{\\PYZgt{}} \\PY{n+nb}{tuple}\\PY{p}{:}\n", "\\PY{+w}{ }\\PY{l+s+sd}{\\PYZdq{}\\PYZdq{}\\PYZdq{}}\n", "\\PY{l+s+sd}{ Product detail regeneration and metrics boost for enrichment experiments.}\n", "\n", "\\PY{l+s+sd}{ Selects a fraction of products for treatment, regenerates their product}\n", "\\PY{l+s+sd}{ details (title, description, features) while preserving brand/category/price,}\n", "\\PY{l+s+sd}{ and applies metrics boost effect.}\n", "\n", "\\PY{l+s+sd}{ Args:}\n", "\\PY{l+s+sd}{ metrics: List of metric record dictionaries}\n", "\\PY{l+s+sd}{ **kwargs: Parameters including:}\n", "\\PY{l+s+sd}{ \\PYZhy{} job\\PYZus{}info: JobInfo for saving product artifacts (required for saving)}\n", "\\PY{l+s+sd}{ \\PYZhy{} products: List of product dictionaries (required for product details)}\n", "\\PY{l+s+sd}{ \\PYZhy{} effect\\PYZus{}size: Percentage increase in ordered units (default: 0.5)}\n", "\\PY{l+s+sd}{ \\PYZhy{} ramp\\PYZus{}days: Number of days for ramp\\PYZhy{}up period (default: 7)}\n", "\\PY{l+s+sd}{ \\PYZhy{} enrichment\\PYZus{}fraction: Fraction of products to enrich (default: 0.3)}\n", "\\PY{l+s+sd}{ \\PYZhy{} enrichment\\PYZus{}start: Start date of enrichment (default: \\PYZdq{}2024\\PYZhy{}11\\PYZhy{}15\\PYZdq{})}\n", "\\PY{l+s+sd}{ \\PYZhy{} seed: Random seed for product selection (default: 42)}\n", "\\PY{l+s+sd}{ \\PYZhy{} prompt\\PYZus{}path: Path to custom prompt template file (optional)}\n", "\\PY{l+s+sd}{ \\PYZhy{} backend: Backend to use for regeneration (\\PYZdq{}mock\\PYZdq{} or \\PYZdq{}ollama\\PYZdq{}, default: \\PYZdq{}mock\\PYZdq{})}\n", "\n", "\\PY{l+s+sd}{ Returns:}\n", "\\PY{l+s+sd}{ Tuple of (treated\\PYZus{}metrics, potential\\PYZus{}outcomes\\PYZus{}df):}\n", "\\PY{l+s+sd}{ \\PYZhy{} treated\\PYZus{}metrics: List of modified metric dictionaries with treatment applied}\n", "\\PY{l+s+sd}{ \\PYZhy{} potential\\PYZus{}outcomes\\PYZus{}df: DataFrame with Y0\\PYZus{}revenue and Y1\\PYZus{}revenue for all products}\n", "\\PY{l+s+sd}{ \\PYZdq{}\\PYZdq{}\\PYZdq{}}\n", " \\PY{n}{job\\PYZus{}info} \\PY{o}{=} \\PY{n}{kwargs}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{job\\PYZus{}info}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\n", " \\PY{n}{products} \\PY{o}{=} \\PY{n}{kwargs}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{products}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\n", " \\PY{n}{effect\\PYZus{}size} \\PY{o}{=} \\PY{n}{kwargs}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{effect\\PYZus{}size}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{l+m+mf}{0.5}\\PY{p}{)}\n", " \\PY{n}{ramp\\PYZus{}days} \\PY{o}{=} \\PY{n}{kwargs}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{ramp\\PYZus{}days}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{l+m+mi}{7}\\PY{p}{)}\n", " \\PY{n}{enrichment\\PYZus{}fraction} \\PY{o}{=} \\PY{n}{kwargs}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{enrichment\\PYZus{}fraction}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{l+m+mf}{0.3}\\PY{p}{)}\n", " \\PY{n}{enrichment\\PYZus{}start} \\PY{o}{=} \\PY{n}{kwargs}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{enrichment\\PYZus{}start}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{2024\\PYZhy{}11\\PYZhy{}15}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\n", " \\PY{n}{seed} \\PY{o}{=} \\PY{n}{kwargs}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{seed}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{l+m+mi}{42}\\PY{p}{)}\n", " \\PY{n}{prompt\\PYZus{}path} \\PY{o}{=} \\PY{n}{kwargs}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{prompt\\PYZus{}path}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\n", " \\PY{n}{backend} \\PY{o}{=} \\PY{n}{kwargs}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{backend}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{mock}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\n", " \\PY{n}{quality\\PYZus{}boost} \\PY{o}{=} \\PY{n}{kwargs}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{quality\\PYZus{}boost}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{l+m+mf}{0.0}\\PY{p}{)}\n", "\n", " \\PY{n}{rng} \\PY{o}{=} \\PY{n}{np}\\PY{o}{.}\\PY{n}{random}\\PY{o}{.}\\PY{n}{default\\PYZus{}rng}\\PY{p}{(}\\PY{n}{seed}\\PY{p}{)}\n", "\n", " \\PY{c+c1}{\\PYZsh{} 1. Save original product details}\n", " \\PY{k}{if} \\PY{n}{job\\PYZus{}info} \\PY{o+ow}{and} \\PY{n}{products}\\PY{p}{:}\n", " \\PY{n}{job\\PYZus{}info}\\PY{o}{.}\\PY{n}{save\\PYZus{}df}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{product\\PYZus{}details\\PYZus{}original}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{n}{pd}\\PY{o}{.}\\PY{n}{DataFrame}\\PY{p}{(}\\PY{n}{products}\\PY{p}{)}\\PY{p}{)}\n", "\n", " \\PY{c+c1}{\\PYZsh{} 2. Select treatment products}\n", " \\PY{k}{if} \\PY{n}{products}\\PY{p}{:}\n", " \\PY{n}{unique\\PYZus{}product\\PYZus{}ids} \\PY{o}{=} \\PY{n+nb}{sorted}\\PY{p}{(}\\PY{n+nb}{set}\\PY{p}{(}\\PY{n}{p}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{product\\PYZus{}identifier}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{n}{p}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{product\\PYZus{}id}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\\PY{p}{)} \\PY{k}{for} \\PY{n}{p} \\PY{o+ow}{in} \\PY{n}{products}\\PY{p}{)}\\PY{p}{)}\n", " \\PY{k}{else}\\PY{p}{:}\n", " \\PY{n}{unique\\PYZus{}product\\PYZus{}ids} \\PY{o}{=} \\PY{n+nb}{sorted}\\PY{p}{(}\\PY{n+nb}{set}\\PY{p}{(}\\PY{n}{record}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{product\\PYZus{}id}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]} \\PY{k}{for} \\PY{n}{record} \\PY{o+ow}{in} \\PY{n}{metrics}\\PY{p}{)}\\PY{p}{)}\n", "\n", " \\PY{n}{n\\PYZus{}treatment} \\PY{o}{=} \\PY{n+nb}{int}\\PY{p}{(}\\PY{n+nb}{len}\\PY{p}{(}\\PY{n}{unique\\PYZus{}product\\PYZus{}ids}\\PY{p}{)} \\PY{o}{*} \\PY{n}{enrichment\\PYZus{}fraction}\\PY{p}{)}\n", " \\PY{n}{treatment\\PYZus{}ids} \\PY{o}{=} \\PY{n+nb}{set}\\PY{p}{(}\\PY{n}{rng}\\PY{o}{.}\\PY{n}{choice}\\PY{p}{(}\\PY{n}{unique\\PYZus{}product\\PYZus{}ids}\\PY{p}{,} \\PY{n}{size}\\PY{o}{=}\\PY{n}{n\\PYZus{}treatment}\\PY{p}{,} \\PY{n}{replace}\\PY{o}{=}\\PY{k+kc}{False}\\PY{p}{)}\\PY{p}{)}\n", "\n", " \\PY{c+c1}{\\PYZsh{} 3. Regenerate product details for treatment products}\n", " \\PY{k}{if} \\PY{n}{products} \\PY{o+ow}{and} \\PY{n}{job\\PYZus{}info}\\PY{p}{:}\n", " \\PY{n}{updated\\PYZus{}products} \\PY{o}{=} \\PY{n}{\\PYZus{}regenerate\\PYZus{}product\\PYZus{}details}\\PY{p}{(}\n", " \\PY{n}{products}\\PY{p}{,} \\PY{n}{treatment\\PYZus{}ids}\\PY{p}{,} \\PY{n}{prompt\\PYZus{}path}\\PY{p}{,} \\PY{n}{backend}\\PY{p}{,} \\PY{n}{seed}\\PY{p}{,} \\PY{n}{quality\\PYZus{}boost}\n", " \\PY{p}{)}\n", " \\PY{n}{job\\PYZus{}info}\\PY{o}{.}\\PY{n}{save\\PYZus{}df}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{product\\PYZus{}details\\PYZus{}enriched}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{n}{pd}\\PY{o}{.}\\PY{n}{DataFrame}\\PY{p}{(}\\PY{n}{updated\\PYZus{}products}\\PY{p}{)}\\PY{p}{)}\n", "\n", " \\PY{c+c1}{\\PYZsh{} 4. Apply metrics boost effect and calculate potential outcomes}\n", " \\PY{n}{treated\\PYZus{}metrics} \\PY{o}{=} \\PY{p}{[}\\PY{p}{]}\n", " \\PY{n}{potential\\PYZus{}outcomes} \\PY{o}{=} \\PY{p}{\\PYZob{}}\\PY{p}{\\PYZcb{}} \\PY{c+c1}{\\PYZsh{} \\PYZob{}(product\\PYZus{}id, date): \\PYZob{}\\PYZsq{}Y0\\PYZus{}revenue\\PYZsq{}: x, \\PYZsq{}Y1\\PYZus{}revenue\\PYZsq{}: y\\PYZcb{}\\PYZcb{}}\n", " \\PY{n}{start\\PYZus{}date} \\PY{o}{=} \\PY{n}{datetime}\\PY{o}{.}\\PY{n}{strptime}\\PY{p}{(}\\PY{n}{enrichment\\PYZus{}start}\\PY{p}{,} \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{\\PYZpc{}}\\PY{l+s+s2}{Y\\PYZhy{}}\\PY{l+s+s2}{\\PYZpc{}}\\PY{l+s+s2}{m\\PYZhy{}}\\PY{l+s+si}{\\PYZpc{}d}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\n", "\n", " \\PY{k}{for} \\PY{n}{record} \\PY{o+ow}{in} \\PY{n}{metrics}\\PY{p}{:}\n", " \\PY{n}{record\\PYZus{}copy} \\PY{o}{=} \\PY{n}{copy}\\PY{o}{.}\\PY{n}{deepcopy}\\PY{p}{(}\\PY{n}{record}\\PY{p}{)}\n", " \\PY{n}{product\\PYZus{}id} \\PY{o}{=} \\PY{n}{record\\PYZus{}copy}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{product\\PYZus{}id}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{n}{record\\PYZus{}copy}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{product\\PYZus{}identifier}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\\PY{p}{)}\n", " \\PY{n}{record\\PYZus{}date\\PYZus{}str} \\PY{o}{=} \\PY{n}{record\\PYZus{}copy}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{date}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\n", " \\PY{n}{record\\PYZus{}date} \\PY{o}{=} \\PY{n}{datetime}\\PY{o}{.}\\PY{n}{strptime}\\PY{p}{(}\\PY{n}{record\\PYZus{}date\\PYZus{}str}\\PY{p}{,} \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{\\PYZpc{}}\\PY{l+s+s2}{Y\\PYZhy{}}\\PY{l+s+s2}{\\PYZpc{}}\\PY{l+s+s2}{m\\PYZhy{}}\\PY{l+s+si}{\\PYZpc{}d}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\n", "\n", " \\PY{n}{is\\PYZus{}enriched} \\PY{o}{=} \\PY{n}{product\\PYZus{}id} \\PY{o+ow}{in} \\PY{n}{treatment\\PYZus{}ids}\n", " \\PY{n}{record\\PYZus{}copy}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{enriched}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]} \\PY{o}{=} \\PY{n}{is\\PYZus{}enriched}\n", "\n", " \\PY{c+c1}{\\PYZsh{} Calculate Y(0) \\PYZhy{} baseline revenue (no treatment)}\n", " \\PY{n}{y0\\PYZus{}revenue} \\PY{o}{=} \\PY{n}{record\\PYZus{}copy}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{revenue}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\n", "\n", " \\PY{c+c1}{\\PYZsh{} Calculate Y(1) \\PYZhy{} revenue if treated (for ALL products, with ramp\\PYZhy{}up)}\n", " \\PY{n}{unit\\PYZus{}price} \\PY{o}{=} \\PY{n}{record\\PYZus{}copy}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{unit\\PYZus{}price}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{,} \\PY{n}{record\\PYZus{}copy}\\PY{o}{.}\\PY{n}{get}\\PY{p}{(}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{price}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{)}\\PY{p}{)}\n", " \\PY{k}{if} \\PY{n}{record\\PYZus{}date} \\PY{o}{\\PYZgt{}}\\PY{o}{=} \\PY{n}{start\\PYZus{}date}\\PY{p}{:}\n", " \\PY{n}{days\\PYZus{}since\\PYZus{}start} \\PY{o}{=} \\PY{p}{(}\\PY{n}{record\\PYZus{}date} \\PY{o}{\\PYZhy{}} \\PY{n}{start\\PYZus{}date}\\PY{p}{)}\\PY{o}{.}\\PY{n}{days}\n", " \\PY{n}{ramp\\PYZus{}factor} \\PY{o}{=} \\PY{l+m+mf}{1.0} \\PY{k}{if} \\PY{n}{ramp\\PYZus{}days} \\PY{o}{\\PYZlt{}}\\PY{o}{=} \\PY{l+m+mi}{0} \\PY{k}{else} \\PY{n+nb}{min}\\PY{p}{(}\\PY{l+m+mf}{1.0}\\PY{p}{,} \\PY{n}{days\\PYZus{}since\\PYZus{}start} \\PY{o}{/} \\PY{n}{ramp\\PYZus{}days}\\PY{p}{)}\n", " \\PY{n}{adjusted\\PYZus{}effect} \\PY{o}{=} \\PY{n}{effect\\PYZus{}size} \\PY{o}{*} \\PY{n}{ramp\\PYZus{}factor}\n", "\n", " \\PY{n}{original\\PYZus{}quantity} \\PY{o}{=} \\PY{n}{record\\PYZus{}copy}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{ordered\\PYZus{}units}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\n", " \\PY{n}{boosted\\PYZus{}quantity} \\PY{o}{=} \\PY{n+nb}{int}\\PY{p}{(}\\PY{n}{original\\PYZus{}quantity} \\PY{o}{*} \\PY{p}{(}\\PY{l+m+mi}{1} \\PY{o}{+} \\PY{n}{adjusted\\PYZus{}effect}\\PY{p}{)}\\PY{p}{)}\n", " \\PY{n}{y1\\PYZus{}revenue} \\PY{o}{=} \\PY{n+nb}{round}\\PY{p}{(}\\PY{n}{boosted\\PYZus{}quantity} \\PY{o}{*} \\PY{n}{unit\\PYZus{}price}\\PY{p}{,} \\PY{l+m+mi}{2}\\PY{p}{)}\n", " \\PY{k}{else}\\PY{p}{:}\n", " \\PY{c+c1}{\\PYZsh{} Before treatment start, Y(1) = Y(0)}\n", " \\PY{n}{y1\\PYZus{}revenue} \\PY{o}{=} \\PY{n}{y0\\PYZus{}revenue}\n", "\n", " \\PY{c+c1}{\\PYZsh{} Store potential outcomes for ALL products}\n", " \\PY{n}{key} \\PY{o}{=} \\PY{p}{(}\\PY{n}{product\\PYZus{}id}\\PY{p}{,} \\PY{n}{record\\PYZus{}date\\PYZus{}str}\\PY{p}{)}\n", " \\PY{n}{potential\\PYZus{}outcomes}\\PY{p}{[}\\PY{n}{key}\\PY{p}{]} \\PY{o}{=} \\PY{p}{\\PYZob{}}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{Y0\\PYZus{}revenue}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{y0\\PYZus{}revenue}\\PY{p}{,} \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{Y1\\PYZus{}revenue}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{y1\\PYZus{}revenue}\\PY{p}{\\PYZcb{}}\n", "\n", " \\PY{c+c1}{\\PYZsh{} Apply factual outcome (only for treated products)}\n", " \\PY{k}{if} \\PY{n}{is\\PYZus{}enriched} \\PY{o+ow}{and} \\PY{n}{record\\PYZus{}date} \\PY{o}{\\PYZgt{}}\\PY{o}{=} \\PY{n}{start\\PYZus{}date}\\PY{p}{:}\n", " \\PY{n}{record\\PYZus{}copy}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{ordered\\PYZus{}units}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]} \\PY{o}{=} \\PY{n}{boosted\\PYZus{}quantity}\n", " \\PY{n}{record\\PYZus{}copy}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{revenue}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]} \\PY{o}{=} \\PY{n}{y1\\PYZus{}revenue}\n", "\n", " \\PY{n}{treated\\PYZus{}metrics}\\PY{o}{.}\\PY{n}{append}\\PY{p}{(}\\PY{n}{record\\PYZus{}copy}\\PY{p}{)}\n", "\n", " \\PY{c+c1}{\\PYZsh{} Build potential outcomes DataFrame}\n", " \\PY{n}{potential\\PYZus{}outcomes\\PYZus{}df} \\PY{o}{=} \\PY{n}{pd}\\PY{o}{.}\\PY{n}{DataFrame}\\PY{p}{(}\n", " \\PY{p}{[}\n", " \\PY{p}{\\PYZob{}}\n", " \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{product\\PYZus{}identifier}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{pid}\\PY{p}{,}\n", " \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{date}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{d}\\PY{p}{,}\n", " \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{Y0\\PYZus{}revenue}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{v}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{Y0\\PYZus{}revenue}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\\PY{p}{,}\n", " \\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{Y1\\PYZus{}revenue}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{:} \\PY{n}{v}\\PY{p}{[}\\PY{l+s+s2}{\\PYZdq{}}\\PY{l+s+s2}{Y1\\PYZus{}revenue}\\PY{l+s+s2}{\\PYZdq{}}\\PY{p}{]}\\PY{p}{,}\n", " \\PY{p}{\\PYZcb{}}\n", " \\PY{k}{for} \\PY{p}{(}\\PY{n}{pid}\\PY{p}{,} \\PY{n}{d}\\PY{p}{)}\\PY{p}{,} \\PY{n}{v} \\PY{o+ow}{in} \\PY{n}{potential\\PYZus{}outcomes}\\PY{o}{.}\\PY{n}{items}\\PY{p}{(}\\PY{p}{)}\n", " \\PY{p}{]}\n", " \\PY{p}{)}\n", "\n", " \\PY{k}{return} \\PY{n}{treated\\PYZus{}metrics}\\PY{p}{,} \\PY{n}{potential\\PYZus{}outcomes\\PYZus{}df}\n", "\\end{Verbatim}\n" ], "text/plain": [ "def product_detail_boost(metrics: list, **kwargs) -> tuple:\n", " \"\"\"\n", " Product detail regeneration and metrics boost for enrichment experiments.\n", "\n", " Selects a fraction of products for treatment, regenerates their product\n", " details (title, description, features) while preserving brand/category/price,\n", " and applies metrics boost effect.\n", "\n", " Args:\n", " metrics: List of metric record dictionaries\n", " **kwargs: Parameters including:\n", " - job_info: JobInfo for saving product artifacts (required for saving)\n", " - products: List of product dictionaries (required for product details)\n", " - effect_size: Percentage increase in ordered units (default: 0.5)\n", " - ramp_days: Number of days for ramp-up period (default: 7)\n", " - enrichment_fraction: Fraction of products to enrich (default: 0.3)\n", " - enrichment_start: Start date of enrichment (default: \"2024-11-15\")\n", " - seed: Random seed for product selection (default: 42)\n", " - prompt_path: Path to custom prompt template file (optional)\n", " - backend: Backend to use for regeneration (\"mock\" or \"ollama\", default: \"mock\")\n", "\n", " Returns:\n", " Tuple of (treated_metrics, potential_outcomes_df):\n", " - treated_metrics: List of modified metric dictionaries with treatment applied\n", " - potential_outcomes_df: DataFrame with Y0_revenue and Y1_revenue for all products\n", " \"\"\"\n", " job_info = kwargs.get(\"job_info\")\n", " products = kwargs.get(\"products\")\n", " effect_size = kwargs.get(\"effect_size\", 0.5)\n", " ramp_days = kwargs.get(\"ramp_days\", 7)\n", " enrichment_fraction = kwargs.get(\"enrichment_fraction\", 0.3)\n", " enrichment_start = kwargs.get(\"enrichment_start\", \"2024-11-15\")\n", " seed = kwargs.get(\"seed\", 42)\n", " prompt_path = kwargs.get(\"prompt_path\")\n", " backend = kwargs.get(\"backend\", \"mock\")\n", " quality_boost = kwargs.get(\"quality_boost\", 0.0)\n", "\n", " rng = np.random.default_rng(seed)\n", "\n", " # 1. Save original product details\n", " if job_info and products:\n", " job_info.save_df(\"product_details_original\", pd.DataFrame(products))\n", "\n", " # 2. Select treatment products\n", " if products:\n", " unique_product_ids = sorted(set(p.get(\"product_identifier\", p.get(\"product_id\")) for p in products))\n", " else:\n", " unique_product_ids = sorted(set(record[\"product_id\"] for record in metrics))\n", "\n", " n_treatment = int(len(unique_product_ids) * enrichment_fraction)\n", " treatment_ids = set(rng.choice(unique_product_ids, size=n_treatment, replace=False))\n", "\n", " # 3. Regenerate product details for treatment products\n", " if products and job_info:\n", " updated_products = _regenerate_product_details(\n", " products, treatment_ids, prompt_path, backend, seed, quality_boost\n", " )\n", " job_info.save_df(\"product_details_enriched\", pd.DataFrame(updated_products))\n", "\n", " # 4. Apply metrics boost effect and calculate potential outcomes\n", " treated_metrics = []\n", " potential_outcomes = {} # {(product_id, date): {'Y0_revenue': x, 'Y1_revenue': y}}\n", " start_date = datetime.strptime(enrichment_start, \"%Y-%m-%d\")\n", "\n", " for record in metrics:\n", " record_copy = copy.deepcopy(record)\n", " product_id = record_copy.get(\"product_id\", record_copy.get(\"product_identifier\"))\n", " record_date_str = record_copy[\"date\"]\n", " record_date = datetime.strptime(record_date_str, \"%Y-%m-%d\")\n", "\n", " is_enriched = product_id in treatment_ids\n", " record_copy[\"enriched\"] = is_enriched\n", "\n", " # Calculate Y(0) - baseline revenue (no treatment)\n", " y0_revenue = record_copy[\"revenue\"]\n", "\n", " # Calculate Y(1) - revenue if treated (for ALL products, with ramp-up)\n", " unit_price = record_copy.get(\"unit_price\", record_copy.get(\"price\"))\n", " if record_date >= start_date:\n", " days_since_start = (record_date - start_date).days\n", " ramp_factor = 1.0 if ramp_days <= 0 else min(1.0, days_since_start / ramp_days)\n", " adjusted_effect = effect_size * ramp_factor\n", "\n", " original_quantity = record_copy[\"ordered_units\"]\n", " boosted_quantity = int(original_quantity * (1 + adjusted_effect))\n", " y1_revenue = round(boosted_quantity * unit_price, 2)\n", " else:\n", " # Before treatment start, Y(1) = Y(0)\n", " y1_revenue = y0_revenue\n", "\n", " # Store potential outcomes for ALL products\n", " key = (product_id, record_date_str)\n", " potential_outcomes[key] = {\"Y0_revenue\": y0_revenue, \"Y1_revenue\": y1_revenue}\n", "\n", " # Apply factual outcome (only for treated products)\n", " if is_enriched and record_date >= start_date:\n", " record_copy[\"ordered_units\"] = boosted_quantity\n", " record_copy[\"revenue\"] = y1_revenue\n", "\n", " treated_metrics.append(record_copy)\n", "\n", " # Build potential outcomes DataFrame\n", " potential_outcomes_df = pd.DataFrame(\n", " [\n", " {\n", " \"product_identifier\": pid,\n", " \"date\": d,\n", " \"Y0_revenue\": v[\"Y0_revenue\"],\n", " \"Y1_revenue\": v[\"Y1_revenue\"],\n", " }\n", " for (pid, d), v in potential_outcomes.items()\n", " ]\n", " )\n", "\n", " return treated_metrics, potential_outcomes_df" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from online_retail_simulator.enrich.enrichment_library import product_detail_boost\n", "\n", "Code(inspect.getsource(product_detail_boost), language=\"python\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Calling `enrich(\"config_tutorial_enrichment.yaml\", job_info)` applies the treatment effect to the simulated data. The `product_detail_boost()` function performs two operations: first, it regenerates product details for the treatment group using `treatment_mode=True` (creating enhanced content with premium brand names, compelling descriptions, and specific feature highlights); second, it applies a 50% sales boost to these products starting November 15th. The function automatically saves three artifacts: the original product details, the enhanced treatment product details, and the enriched sales data—which we can load and examine to understand both the content improvements and their impact on sales metrics." ] }, { "cell_type": "code", "execution_count": 16, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:51.497471Z", "iopub.status.busy": "2026-03-01T03:45:51.497282Z", "iopub.status.idle": "2026-03-01T03:45:51.644945Z", "shell.execute_reply": "2026-03-01T03:45:51.643980Z" } }, "outputs": [], "source": [ "enriched_job = enrich(\"config_tutorial_enrichment.yaml\", job_info)\n", "results = load_job_results(enriched_job)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Changes" ] }, { "cell_type": "code", "execution_count": 17, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:51.646766Z", "iopub.status.busy": "2026-03-01T03:45:51.646604Z", "iopub.status.idle": "2026-03-01T03:45:51.656135Z", "shell.execute_reply": "2026-03-01T03:45:51.655444Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
product_identifiercategorypricetitledescriptionbrandfeaturesquality_scoredateimpressionsvisitscart_addsordered_unitsrevenueenriched
0B1P4DZHDS9Electronics686.37NexGen Smart Electronics ItemQuality electronics product for everyday use. ...NexGen[np.str_('Long battery life'), np.str_('Voice ...0.8002024-11-01254100.00True
1B1SE4QSNG7Toys & Games80.75PrimePick Premium Toys & Games ItemQuality toys & games product for everyday use....PrimePick[np.str_('High quality materials'), np.str_('D...0.8822024-11-01200339180.75True
2BXTPQIDT5CFood & Beverage42.02TrustMark Premium Food & Beverage ItemQuality food & beverage product for everyday u...TrustMark[np.str_('Easy to use'), np.str_('High quality...0.8792024-11-01101000.00True
3B3F1ZMC8Q6Food & Beverage33.42BestChoice Classic Food & Beverage ItemQuality food & beverage product for everyday u...BestChoice[np.str_('High quality materials'), np.str_('G...0.8852024-11-0100000.00True
4B2NQRBTF0YToys & Games27.52QualityFirst Classic Toys & Games ItemQuality toys & games product for everyday use....QualityFirst[np.str_('Great value'), np.str_('Easy to use'...0.8302024-11-01101000.00True
\n", "
" ], "text/plain": [ " product_identifier category price \\\n", "0 B1P4DZHDS9 Electronics 686.37 \n", "1 B1SE4QSNG7 Toys & Games 80.75 \n", "2 BXTPQIDT5C Food & Beverage 42.02 \n", "3 B3F1ZMC8Q6 Food & Beverage 33.42 \n", "4 B2NQRBTF0Y Toys & Games 27.52 \n", "\n", " title \\\n", "0 NexGen Smart Electronics Item \n", "1 PrimePick Premium Toys & Games Item \n", "2 TrustMark Premium Food & Beverage Item \n", "3 BestChoice Classic Food & Beverage Item \n", "4 QualityFirst Classic Toys & Games Item \n", "\n", " description brand \\\n", "0 Quality electronics product for everyday use. ... NexGen \n", "1 Quality toys & games product for everyday use.... PrimePick \n", "2 Quality food & beverage product for everyday u... TrustMark \n", "3 Quality food & beverage product for everyday u... BestChoice \n", "4 Quality toys & games product for everyday use.... QualityFirst \n", "\n", " features quality_score \\\n", "0 [np.str_('Long battery life'), np.str_('Voice ... 0.800 \n", "1 [np.str_('High quality materials'), np.str_('D... 0.882 \n", "2 [np.str_('Easy to use'), np.str_('High quality... 0.879 \n", "3 [np.str_('High quality materials'), np.str_('G... 0.885 \n", "4 [np.str_('Great value'), np.str_('Easy to use'... 0.830 \n", "\n", " date impressions visits cart_adds ordered_units revenue \\\n", "0 2024-11-01 25 4 1 0 0.00 \n", "1 2024-11-01 200 33 9 1 80.75 \n", "2 2024-11-01 10 1 0 0 0.00 \n", "3 2024-11-01 0 0 0 0 0.00 \n", "4 2024-11-01 10 1 0 0 0.00 \n", "\n", " enriched \n", "0 True \n", "1 True \n", "2 True \n", "3 True \n", "4 True " ] }, "execution_count": 17, "metadata": {}, "output_type": "execute_result" } ], "source": [ "enriched = results[\"enriched\"]\n", "enriched.head()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The enriched dataset contains the same metrics as before (`impressions`, `visits`, `cart_adds`, `ordered_units`, `revenue`) but now includes an `enriched` column indicating which products received the treatment. Products with `enriched=True` have enhanced content and boosted sales metrics starting from the treatment date." ] }, { "cell_type": "code", "execution_count": 18, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:51.657971Z", "iopub.status.busy": "2026-03-01T03:45:51.657801Z", "iopub.status.idle": "2026-03-01T03:45:51.662024Z", "shell.execute_reply": "2026-03-01T03:45:51.661398Z" } }, "outputs": [ { "data": { "text/plain": [ "product_identifier B1P4DZHDS9\n", "category Electronics\n", "price 686.37\n", "title NexGen Smart Electronics Item\n", "description Quality electronics product for everyday use. ...\n", "brand NexGen\n", "features [np.str_('Long battery life'), np.str_('Voice ...\n", "quality_score 0.8\n", "date 2024-11-01\n", "impressions 25\n", "visits 4\n", "cart_adds 1\n", "ordered_units 0\n", "revenue 0.0\n", "enriched True\n", "Name: 0, dtype: object" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" } ], "source": [ "example_enriched = enriched.iloc[0]\n", "example_enriched" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "This record shows all the sales funnel metrics plus the `enriched` indicator. The treatment assignment is random but deterministic (controlled by the `seed` parameter), allowing us to create a reproducible experiment where we know the ground truth effect size." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The enrichment process modifies both sales metrics and product content. Let's compare the original and enhanced product details to see the transformation." ] }, { "cell_type": "code", "execution_count": 19, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:51.663928Z", "iopub.status.busy": "2026-03-01T03:45:51.663774Z", "iopub.status.idle": "2026-03-01T03:45:51.667079Z", "shell.execute_reply": "2026-03-01T03:45:51.666288Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "======================================================================\n", "CONTROL GROUP (Original Content)\n", "======================================================================\n", "Brand: NexGen\n", "Title: NexGen Smart Electronics Item\n", "Description: Quality electronics product for everyday use. Long battery life. Voice control.\n", "Features: [np.str_('Long battery life'), np.str_('Voice control'), np.str_('Bluetooth connectivity'), np.str_('Fast charging')]\n" ] } ], "source": [ "control_products = results[\"product_details_original\"]\n", "control_example = control_products.iloc[0]\n", "\n", "display_product_details(control_example, \"CONTROL GROUP (Original Content)\")" ] }, { "cell_type": "code", "execution_count": 20, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:51.668697Z", "iopub.status.busy": "2026-03-01T03:45:51.668543Z", "iopub.status.idle": "2026-03-01T03:45:51.671731Z", "shell.execute_reply": "2026-03-01T03:45:51.670967Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "======================================================================\n", "TREATMENT GROUP (Enhanced Content)\n", "======================================================================\n", "Brand: NexGen\n", "Title: NexGen Revolutionary Electronics Item\n", "Description: Premium electronics product with exceptional quality. Crystal-clear OLED display. Advanced Bluetooth 5.0.\n", "Features: [np.str_('Crystal-clear OLED display'), np.str_('Advanced Bluetooth 5.0'), np.str_('Lightning-fast charging'), np.str_('Intelligent voice assistant')]\n" ] } ], "source": [ "treatment_products = results[\"product_details_enriched\"]\n", "treatment_example = treatment_products.iloc[0]\n", "\n", "display_product_details(treatment_example, \"TREATMENT GROUP (Enhanced Content)\", add_newline=True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "
\n", "Things to explore\n", "
    \n", "
  • What differences do you notice in brand names?
    • \n", "
  • How do the titles differ between control and treatment?
    • \n", "
  • What makes the treatment description more compelling?
    • \n", "
  • How are the features enhanced in the treatment version?
    • \n", "
\n", "
" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "### Impact" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The treatment effect visualization compares daily revenue between the original and enriched datasets. Before the treatment start date, the lines should overlap perfectly. After November 15th, the enriched dataset should show higher revenue due to the 50% boost applied to all products." ] }, { "cell_type": "code", "execution_count": 21, "metadata": { "execution": { "iopub.execute_input": "2026-03-01T03:45:51.673475Z", "iopub.status.busy": "2026-03-01T03:45:51.673294Z", "iopub.status.idle": "2026-03-01T03:45:51.851398Z", "shell.execute_reply": "2026-03-01T03:45:51.850517Z" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "plot_treatment_effect(metrics, enriched, \"2024-11-15\")" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.12.3" } }, "nbformat": 4, "nbformat_minor": 4 }