{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "2021-07-18-real-time-in-session-personalization.ipynb",
      "provenance": [],
      "toc_visible": true,
      "mount_file_id": "16GrKMZiYuBlqPhTwLUIeLN4o4wmXI56Y",
      "authorship_tag": "ABX9TyMd0iA17gH6aAuFMcRV0J7I"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "JuP8_LTZpq6W"
      },
      "source": [
        "# Real-time In-session Personalization\n",
        "> Personalization with deep learning in 100 lines of code using Elasticsearch, Redis and Word2vec\n",
        "\n",
        "- toc: true\n",
        "- badges: true\n",
        "- comments: true\n",
        "- categories: [Elasticsearch, Redis, Word2vec, Visualization, Session, Sequential, Embedding, RealTime, Retail]\n",
        "- image:"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PfGYKsKLto_Y"
      },
      "source": [
        "### Introduction"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Etc2uAVCqzoW"
      },
      "source": [
        "| |  |\n",
        "| :-: | -:|\n",
        "| Vision | Personalizing Customer's In-session experience in real-time |\n",
        "| Mission | Learn embeddings for a small sequetial dataset and use efficient indexing and database techniques to retrieve results in real-time |\n",
        "| Scope | Synthetic dataset, Model training, indexing and retrieval, eCommerce space |\n",
        "| Task | Next-item Prediction |\n",
        "| Data | Synthetic Sequential |\n",
        "| Tool | Word2vec, Elasticsearch, Redis, Colab |\n",
        "| Technique | Word2vec for embedding, Elasticsearch for indexing, Redis for in-memory fast retrieval |\n",
        "| Process | 1) Setup environment by installing and starting elaticsearch, redis backend server, 2) Generate synthetic data, 3) Train the embedding model, 4) Visualize the results, 5) Index the embedding vectors, 6) Retrieve the top-k results in real-time |\n",
        "| Takeaway | Elasticsearch and Redis - a good combination for fast and efficient retrieval at scale, Word2vec is simple yet effective entry-level technique for learning sequential embeddings |\n",
        "| Credit | [Jacopo Tagliabue](https://github.com/jacopotagliabue/clothes-in-space) |\n",
        "| Link | [link1](https://github.com/jacopotagliabue), [link2](https://blog.coveo.com/clothes-in-space-real-time-personalization-in-less-than-100-lines-of-code/) |"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CGj9t27ro2uY"
      },
      "source": [
        "### Install Elasticsearch"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8XIwkKAEgo__",
        "outputId": "bda87def-02e4-45d1-a1c4-bec3562084eb"
      },
      "source": [
        "!wget https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-oss-7.0.5-linux-x86_64.tar.gz"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2021-07-18 07:27:11--  https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-oss-7.0.5-linux-x86_64.tar.gz\n",
            "Resolving artifacts.elastic.co (artifacts.elastic.co)... 34.120.127.130, 2600:1901:0:1d7::\n",
            "Connecting to artifacts.elastic.co (artifacts.elastic.co)|34.120.127.130|:443... connected.\n",
            "HTTP request sent, awaiting response... 404 Not Found\n",
            "2021-07-18 07:27:11 ERROR 404: Not Found.\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YUj0878jPyz7",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "d57caf6f-4902-4278-a2b2-6eba7b17ada9"
      },
      "source": [
        "%%bash\n",
        "\n",
        "wget -q https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-oss-7.0.1-linux-x86_64.tar.gz\n",
        "wget -q https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-oss-7.0.1-linux-x86_64.tar.gz.sha512\n",
        "tar -xzf elasticsearch-oss-7.0.1-linux-x86_64.tar.gz\n",
        "sudo chown -R daemon:daemon elasticsearch-7.0.1/\n",
        "shasum -a 512 -c elasticsearch-oss-7.0.1-linux-x86_64.tar.gz.sha512 "
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "elasticsearch-oss-7.0.1-linux-x86_64.tar.gz: OK\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vAzfu_WiEs4F"
      },
      "source": [
        "Run the instance as a daemon process"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "n9ujlunrWgRx",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "5584f918-2492-471f-9895-1c0866460228"
      },
      "source": [
        "%%bash --bg\n",
        "\n",
        "sudo -H -u daemon elasticsearch-7.0.1/bin/elasticsearch"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Starting job # 0 in a separate thread.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XyUa9r6MgWtW"
      },
      "source": [
        "# Sleep for few seconds to let the instance start.\n",
        "import time\n",
        "time.sleep(20)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "f6qxCdypE1DD"
      },
      "source": [
        "Once the instance has been started, grep for `elasticsearch` in the processes list to confirm the availability."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "48LqMJ1BEHm5",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "b64af411-85b9-468e-a343-2ff90b916d28"
      },
      "source": [
        "%%bash\n",
        "\n",
        "ps -ef | grep elasticsearch"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "root         244       1  0 05:33 ?        00:00:00 sudo -H -u daemon elasticsearch-7.9.2/bin/elasticsearch\n",
            "daemon       245     244  0 05:33 ?        00:00:34 /content/elasticsearch-7.9.2/jdk/bin/java -Xshare:auto -Des.networkaddress.cache.ttl=60 -Des.networkaddress.cache.negative.ttl=10 -XX:+AlwaysPreTouch -Xss1m -Djava.awt.headless=true -Dfile.encoding=UTF-8 -Djna.nosys=true -XX:-OmitStackTraceInFastThrow -XX:+ShowCodeDetailsInExceptionMessages -Dio.netty.noUnsafe=true -Dio.netty.noKeySetOptimization=true -Dio.netty.recycler.maxCapacityPerThread=0 -Dio.netty.allocator.numDirectArenas=0 -Dlog4j.shutdownHookEnabled=false -Dlog4j2.disable.jmx=true -Djava.locale.providers=SPI,COMPAT -Xms1g -Xmx1g -XX:+UseG1GC -XX:G1ReservePercent=25 -XX:InitiatingHeapOccupancyPercent=30 -Djava.io.tmpdir=/tmp/elasticsearch-278431634431622414 -XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=data -XX:ErrorFile=logs/hs_err_pid%p.log -Xlog:gc*,gc+age=trace,safepoint:file=logs/gc.log:utctime,pid,tags:filecount=32,filesize=64m -XX:MaxDirectMemorySize=536870912 -Des.path.home=/content/elasticsearch-7.9.2 -Des.path.conf=/content/elasticsearch-7.9.2/config -Des.distribution.flavor=oss -Des.distribution.type=tar -Des.bundled_jdk=true -cp /content/elasticsearch-7.9.2/lib/* org.elasticsearch.bootstrap.Elasticsearch\n",
            "root        7316    7314  0 07:31 ?        00:00:00 sudo -H -u daemon elasticsearch-7.0.1/bin/elasticsearch\n",
            "daemon      7317    7316 99 07:31 ?        00:00:08 /content/elasticsearch-7.0.1/jdk/bin/java -Xms1g -Xmx1g -XX:+UseConcMarkSweepGC -XX:CMSInitiatingOccupancyFraction=75 -XX:+UseCMSInitiatingOccupancyOnly -Des.networkaddress.cache.ttl=60 -Des.networkaddress.cache.negative.ttl=10 -XX:+AlwaysPreTouch -Xss1m -Djava.awt.headless=true -Dfile.encoding=UTF-8 -Djna.nosys=true -XX:-OmitStackTraceInFastThrow -Dio.netty.noUnsafe=true -Dio.netty.noKeySetOptimization=true -Dio.netty.recycler.maxCapacityPerThread=0 -Dlog4j.shutdownHookEnabled=false -Dlog4j2.disable.jmx=true -Djava.io.tmpdir=/tmp/elasticsearch-4253128636895970895 -XX:+HeapDumpOnOutOfMemoryError -XX:HeapDumpPath=data -XX:ErrorFile=logs/hs_err_pid%p.log -Xlog:gc*,gc+age=trace,safepoint:file=logs/gc.log:utctime,pid,tags:filecount=32,filesize=64m -Djava.locale.providers=COMPAT -Dio.netty.allocator.type=unpooled -Des.path.home=/content/elasticsearch-7.0.1 -Des.path.conf=/content/elasticsearch-7.0.1/config -Des.distribution.flavor=oss -Des.distribution.type=tar -Des.bundled_jdk=true -cp /content/elasticsearch-7.0.1/lib/* org.elasticsearch.bootstrap.Elasticsearch\n",
            "root        7404    7402  0 07:31 ?        00:00:00 grep elasticsearch\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wBuRpiyf_kNS"
      },
      "source": [
        "query the base endpoint to retrieve information about the cluster."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ILyohKWQ_XQS",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "41c4d0e8-1888-4b9b-9975-5163756d7dc5"
      },
      "source": [
        "%%bash\n",
        "\n",
        "curl -sX GET \"localhost:9201/\""
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "{\n",
            "  \"name\" : \"106eaf552057\",\n",
            "  \"cluster_name\" : \"elasticsearch\",\n",
            "  \"cluster_uuid\" : \"tbMnpTtSTOeqwDan3GKphg\",\n",
            "  \"version\" : {\n",
            "    \"number\" : \"7.9.2\",\n",
            "    \"build_flavor\" : \"oss\",\n",
            "    \"build_type\" : \"tar\",\n",
            "    \"build_hash\" : \"d34da0ea4a966c4e49417f2da2f244e3e97b4e6e\",\n",
            "    \"build_date\" : \"2020-09-23T00:45:33.626720Z\",\n",
            "    \"build_snapshot\" : false,\n",
            "    \"lucene_version\" : \"8.6.2\",\n",
            "    \"minimum_wire_compatibility_version\" : \"6.8.0\",\n",
            "    \"minimum_index_compatibility_version\" : \"6.0.0-beta1\"\n",
            "  },\n",
            "  \"tagline\" : \"You Know, for Search\"\n",
            "}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BpdcqLnYh_ks",
        "outputId": "080872ed-e2e5-4909-f897-9dc5109b7b9b"
      },
      "source": [
        "%%bash\n",
        "\n",
        "curl -sX GET \"localhost:9201/\""
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "{\n",
            "  \"name\" : \"106eaf552057\",\n",
            "  \"cluster_name\" : \"elasticsearch\",\n",
            "  \"cluster_uuid\" : \"tbMnpTtSTOeqwDan3GKphg\",\n",
            "  \"version\" : {\n",
            "    \"number\" : \"7.0.1\",\n",
            "    \"build_flavor\" : \"oss\",\n",
            "    \"build_type\" : \"tar\",\n",
            "    \"build_hash\" : \"e4efcb5\",\n",
            "    \"build_date\" : \"2019-04-29T12:56:03.145736Z\",\n",
            "    \"build_snapshot\" : false,\n",
            "    \"lucene_version\" : \"8.0.0\",\n",
            "    \"minimum_wire_compatibility_version\" : \"6.7.0\",\n",
            "    \"minimum_index_compatibility_version\" : \"6.0.0-beta1\"\n",
            "  },\n",
            "  \"tagline\" : \"You Know, for Search\"\n",
            "}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jloHOErjLNwe"
      },
      "source": [
        "### Install Redis"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RnbYNpCpHrwM"
      },
      "source": [
        "!wget http://download.redis.io/releases/redis-stable.tar.gz --no-check-certificate\n",
        "!tar -xf redis-stable.tar.gz && cd redis-stable/src && make"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9Ze3TwogHy4I",
        "outputId": "25b9f210-e9bc-48eb-f9ec-1243d8d68182"
      },
      "source": [
        "! nohup ./redis-stable/src/redis-server > redis_nohup.out &\n",
        "! cat redis_nohup.out"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "nohup: redirecting stderr to stdout\n",
            "5863:C 18 Jul 2021 05:46:12.258 # oO0OoO0OoO0Oo Redis is starting oO0OoO0OoO0Oo\n",
            "5863:C 18 Jul 2021 05:46:12.258 # Redis version=6.0.5, bits=64, commit=00000000, modified=0, pid=5863, just started\n",
            "5863:C 18 Jul 2021 05:46:12.258 # Warning: no config file specified, using the default config. In order to specify a config file use ./redis-stable/src/redis-server /path/to/redis.conf\n",
            "5863:M 18 Jul 2021 05:46:12.259 * Running mode=standalone, port=6379.\n",
            "5863:M 18 Jul 2021 05:46:12.259 # Server initialized\n",
            "5863:M 18 Jul 2021 05:46:12.259 # WARNING you have Transparent Huge Pages (THP) support enabled in your kernel. This will create latency and memory usage issues with Redis. To fix this issue run the command 'echo never > /sys/kernel/mm/transparent_hugepage/enabled' as root, and add it to your /etc/rc.local in order to retain the setting after a reboot. Redis must be restarted after THP is disabled.\n",
            "5863:M 18 Jul 2021 05:46:12.260 * Ready to accept connections\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EmkCMB9gH0yi",
        "outputId": "eedda604-f247-425a-ec4b-e08643eb451c"
      },
      "source": [
        "!pip install redis\n",
        "\n",
        "import redis\n",
        "r = redis.StrictRedis(host=\"127.0.0.1\", port=6379)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Collecting redis\n",
            "\u001b[?25l  Downloading https://files.pythonhosted.org/packages/a7/7c/24fb0511df653cf1a5d938d8f5d19802a88cef255706fdda242ff97e91b7/redis-3.5.3-py2.py3-none-any.whl (72kB)\n",
            "\r\u001b[K     |████▌                           | 10kB 17.8MB/s eta 0:00:01\r\u001b[K     |█████████                       | 20kB 22.6MB/s eta 0:00:01\r\u001b[K     |█████████████▋                  | 30kB 17.1MB/s eta 0:00:01\r\u001b[K     |██████████████████▏             | 40kB 15.0MB/s eta 0:00:01\r\u001b[K     |██████████████████████▊         | 51kB 7.2MB/s eta 0:00:01\r\u001b[K     |███████████████████████████▎    | 61kB 8.4MB/s eta 0:00:01\r\u001b[K     |███████████████████████████████▉| 71kB 8.0MB/s eta 0:00:01\r\u001b[K     |████████████████████████████████| 81kB 5.6MB/s \n",
            "\u001b[?25hInstalling collected packages: redis\n",
            "Successfully installed redis-3.5.3\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "XZTmvkfgLSQU"
      },
      "source": [
        "### Generate Synthetic Data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "TooXXcTNIC1T",
        "outputId": "5a2f67cc-0f47-4114-c09b-de154ec242c3"
      },
      "source": [
        "%%writefile catalog.csv\n",
        "sku,name,target,image\n",
        "01111,shoesA,men,\"https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/e706a692-be9e-4d5a-9537-3d64bd2d0e34/blazer-mid-77-shoe-9SvVxB.png\"\n",
        "01112,shoesB,women,\"https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/b98cdc08-0691-48a8-99f7-bd4856ab5628/air-force-1-07-shoe-AKTdww3y.png\"\n",
        "01113,shoesC,men,\"https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/fe07ff2a-e259-4664-b83e-003f68dfaf62/air-max-96-2-shoe-6l7J5T.png\"\n",
        "01114,shoesD,men,\"https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/9313fb42-b9fb-4a13-bd94-46137fd19c12/air-jordan-1-mid-se-shoe-zgPD6z.png\"\n",
        "01115,shoesE,women,\"https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/e1eaf5d3-dfca-4bba-b0f3-2f356520b655/blazer-mid-77-se-shoe-97H9PZ.png\"\n",
        "01116,tshirtA,women,\"https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/a7f6f9ff-8fd2-4acc-b547-8ea5497e9def/nikelab-t-shirt-Q89CVt.png\"\n",
        "01117,tshirtB,women,\"https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/a7b40e60-9626-4538-bdbb-51a0e85da6d3/dri-fit-race-short-sleeve-running-top-h2qbZD.png\"\n",
        "01118,tshirtC,women,\"https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/46752967-b4eb-4d9b-8150-75901f3e87a8/dri-fit-run-division-running-tank-K7wCpp.png\"\n",
        "01119,tshirtD,men,\"https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/f20341c3-0620-4430-9c04-783f9b507789/sb-skate-t-shirt-LHlzwG.png\"\n",
        "01120,tshirtE,men,\"https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/259734bd-93b6-420e-adc6-31cb187e4442/dri-fit-superset-short-sleeve-training-top-6LKnxr.png\""
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Overwriting catalog.csv\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pBXL-Q7CJydq",
        "outputId": "e5434b3f-4668-40cc-f278-774c674cafbd"
      },
      "source": [
        "%%writefile sessions.txt\n",
        "0\t01112 01115 01115 01112\n",
        "1\t01118 01116 01120 01118 01117 01116 01118\n",
        "2\t01113 01114 01114 01111 01111 01113\n",
        "3 01116 01117\n",
        "4 01111 01113 01114 01118 01113 01114 01111\n",
        "5 01117 01117 01118 01116\n",
        "6\t01112 01112 01115 01112\n",
        "7\t01120 01120 01119 01120 01119 01120\n",
        "8\t01113 01114 01114 01111 01113\n",
        "9 01116 01117 01116 01117 01118\n",
        "10 01111 01113 01114 01119 01120 01119\n",
        "11 01117 01117 01118 01116"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Overwriting sessions.txt\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "RchMHrYbmxS2"
      },
      "source": [
        "### Import and global variables"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YattbDl9J2Ey"
      },
      "source": [
        "import os\n",
        "import csv\n",
        "from datetime import datetime\n",
        "import string\n",
        "import requests\n",
        "import json\n",
        "import random\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.manifold import TSNE\n",
        "from sklearn.metrics.pairwise import cosine_similarity\n",
        "import redis  # to communicate with redis\n",
        "import gensim # to talk to gensim\n",
        "from IPython.display import Image  # to display URL in noteboook for visual debug\n",
        "from IPython.core.display import display # to display URL in noteboook for visual debug\n",
        "from elasticsearch import Elasticsearch, helpers # remember to !pip install elasticsearch"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "3-FB2-l4J66f"
      },
      "source": [
        "DATA_FOLDER = ''\n",
        "CATALOGUE_FILE = os.path.join(DATA_FOLDER, 'catalog.csv')\n",
        "SESSION_FILE = os.path.join(DATA_FOLDER, 'sessions.txt')"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QaPInbJdKGPa"
      },
      "source": [
        "EMBEDDING_DIMS = 5 # specify embedding size"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0EDt629Qncz0"
      },
      "source": [
        "### Python clients for Redis and ElasticSearch"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VRHMIUj4KM_n"
      },
      "source": [
        "# redis credentials here!\n",
        "REDIS_HOST = 'localhost'\n",
        "REDIS_PORT = 6379\n",
        "REDIS_DB = 0\n",
        "REDIS_PWD = None\n",
        "# redis data structure\n",
        "REDIS_HASH_FORMAT = 'product_h'\n",
        "# start redis client\n",
        "redis_client = redis.StrictRedis(host=REDIS_HOST, \n",
        "                                 port=REDIS_PORT, \n",
        "                                 db=REDIS_DB, \n",
        "                                 password=REDIS_PWD)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9ViWcJFwKSu_"
      },
      "source": [
        "INDEX_NAME = 'catalog'\n",
        "ES_HOST = {\"host\": \"localhost\", \"port\": 9200}\n",
        "# if you're running with the docker-compose option and not the manuel docker instance, use:\n",
        "# ES_HOST = {\"host\": \"elasticsearch\", \"port\": 9200}\n",
        "es_client = Elasticsearch(hosts=[ES_HOST])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "50p-Clm1njtG"
      },
      "source": [
        "### Train Embeddings"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "AyFHo7ADmxTL"
      },
      "source": [
        "_First of all, get products from the catalogue dump into a usable form_"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-D8h1wzvKXpd"
      },
      "source": [
        "# First of all, get products from the catalogue dump into a usable form\n",
        "def get_products_from_catalogue(catalog_file):\n",
        "    \"\"\"\n",
        "    parse catalogue file into a map SKU -> properties (sku, name, target, image url)\n",
        "    \"\"\"\n",
        "    products = {}\n",
        "    with open(catalog_file) as csvfile:\n",
        "        reader = csv.DictReader(csvfile)\n",
        "        for row in reader:\n",
        "            if row['sku'] and row['image'].endswith('.png'):\n",
        "                products[row['sku']] = row\n",
        "    \n",
        "    return products"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IfdN790_Kebj",
        "outputId": "83342647-da5a-403b-a765-4ed5f4697d6f"
      },
      "source": [
        "products = get_products_from_catalogue(CATALOGUE_FILE)\n",
        "print('{} products in catalog!'.format(len(products)))\n",
        "dict(list(products.items())[0:3])"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "10 products in catalog!\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'01111': OrderedDict([('sku', '01111'),\n",
              "              ('name', 'shoesA'),\n",
              "              ('target', 'men'),\n",
              "              ('image',\n",
              "               'https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/e706a692-be9e-4d5a-9537-3d64bd2d0e34/blazer-mid-77-shoe-9SvVxB.png')]),\n",
              " '01112': OrderedDict([('sku', '01112'),\n",
              "              ('name', 'shoesB'),\n",
              "              ('target', 'women'),\n",
              "              ('image',\n",
              "               'https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/b98cdc08-0691-48a8-99f7-bd4856ab5628/air-force-1-07-shoe-AKTdww3y.png')]),\n",
              " '01113': OrderedDict([('sku', '01113'),\n",
              "              ('name', 'shoesC'),\n",
              "              ('target', 'men'),\n",
              "              ('image',\n",
              "               'https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/fe07ff2a-e259-4664-b83e-003f68dfaf62/air-max-96-2-shoe-6l7J5T.png')])}"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 228
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yhpq7QZnKjNm"
      },
      "source": [
        "def train_embedding_model(training_data):\n",
        "    \"\"\"\n",
        "    training_data is a list of lists (list of words, products, etc.)\n",
        "    \n",
        "    \"\"\"\n",
        "    # train model with standard params\n",
        "    model = gensim.models.Word2Vec(training_data,\n",
        "                                   min_count=1,\n",
        "                                   sample=1,\n",
        "                                   size=EMBEDDING_DIMS,\n",
        "                                   workers=-1,\n",
        "                                   window=3,\n",
        "                                   iter=20)\n",
        "    vectors = model.wv\n",
        "    # remove model from memory\n",
        "    del model\n",
        "    \n",
        "    # return vectors as TOKEN -> VECTOR map\n",
        "    return vectors\n",
        "\n",
        "def solve_vector_analogy(vectors, man, king, women):\n",
        "    # MAN : KING = WOMAN : ? -> QUEEN\n",
        "    return vectors.most_similar_cosmul(positive=[king, women], negative=[man])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2tnJ5TyYKpU9"
      },
      "source": [
        "def get_products_from_sessions(session_file):\n",
        "    \"\"\"\n",
        "        Our file from the analytics service conveniently dumps, line by line,\n",
        "        user sessions. We just read the file and return a list of lists!\n",
        "        \n",
        "        Every line is:\n",
        "        \n",
        "        LINE_ID (as INT) TAB PRODUCT 1 TAB PRODUCT 2 ...\n",
        "        \n",
        "        P.s.: our file has been pre-processed to include only session with length >= 3 and < 200\n",
        "    \"\"\"\n",
        "    sessions = []\n",
        "    with open(session_file) as session_f:\n",
        "        for line in session_f:\n",
        "            products = line.strip().split(' ')[1:]\n",
        "            sessions.append(products)\n",
        "        \n",
        "    return sessions"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "igHQ_LXBMKds",
        "outputId": "8ae8cf2b-e033-42a3-9763-0aded9fa441b"
      },
      "source": [
        "training_session_data = get_products_from_sessions(SESSION_FILE)\n",
        "print('Total sessions: {}, first is: {}'.format(len(training_session_data), training_session_data[0]))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Total sessions: 12, first is: ['01115', '01115', '01112']\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QszMVo-KKrtI"
      },
      "source": [
        "product_embeddings = train_embedding_model(training_session_data)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9yrviXSInoOE"
      },
      "source": [
        "### Visualize Results"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "lRiw5i-8mxTU"
      },
      "source": [
        "_Check item-item similarity by looking at product vectors close together in the space_"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 552
        },
        "id": "U1iolrrhKv-7",
        "outputId": "0a6c9d12-8074-4f01-fcba-88f59033ce91"
      },
      "source": [
        "TEST_PRODUCT = '01111'\n",
        "\n",
        "matches = product_embeddings.most_similar_cosmul(positive=[TEST_PRODUCT])\n",
        "# display top N\n",
        "print(\"For this:\")\n",
        "display(Image(products[TEST_PRODUCT]['image'], width=150, unconfined=True))\n",
        "for i,m in enumerate(matches[:2]):\n",
        "    print(\"\\nRecommendation {}:\".format(i+1))\n",
        "    display(Image(products[m[0]]['image'], width=150, unconfined=True))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "For this:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Recommendation 1:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Recommendation 2:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 552
        },
        "id": "Bb4cfcMWVX4W",
        "outputId": "412147da-7d81-4021-ecb0-6c6ece0f0d02"
      },
      "source": [
        "TEST_PRODUCT = '01112'\n",
        "\n",
        "matches = product_embeddings.most_similar_cosmul(positive=[TEST_PRODUCT])\n",
        "# display top N\n",
        "print(\"For this:\")\n",
        "display(Image(products[TEST_PRODUCT]['image'], width=150, unconfined=True))\n",
        "for i,m in enumerate(matches[:2]):\n",
        "    print(\"\\nRecommendation {}:\".format(i+1))\n",
        "    display(Image(products[m[0]]['image'], width=150, unconfined=True))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "For this:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Recommendation 1:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Recommendation 2:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 552
        },
        "id": "7usD5zN9YX-1",
        "outputId": "80cac680-b125-40d4-9ec6-bbced7a6b72e"
      },
      "source": [
        "TEST_PRODUCT = '01114'\n",
        "\n",
        "matches = product_embeddings.most_similar_cosmul(positive=[TEST_PRODUCT])\n",
        "# display top N\n",
        "print(\"For this:\")\n",
        "display(Image(products[TEST_PRODUCT]['image'], width=150, unconfined=True))\n",
        "for i,m in enumerate(matches[:2]):\n",
        "    print(\"\\nRecommendation {}:\".format(i+1))\n",
        "    display(Image(products[m[0]]['image'], width=150, unconfined=True))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "For this:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Recommendation 1:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Recommendation 2:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 552
        },
        "id": "OosIQItKYMyK",
        "outputId": "3b53c612-f8b0-4d9a-ebbe-776ea86a1072"
      },
      "source": [
        "TEST_PRODUCT = '01116'\n",
        "\n",
        "matches = product_embeddings.most_similar_cosmul(positive=[TEST_PRODUCT])\n",
        "# display top N\n",
        "print(\"For this:\")\n",
        "display(Image(products[TEST_PRODUCT]['image'], width=150, unconfined=True))\n",
        "for i,m in enumerate(matches[:2]):\n",
        "    print(\"\\nRecommendation {}:\".format(i+1))\n",
        "    display(Image(products[m[0]]['image'], width=150, unconfined=True))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "For this:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Recommendation 1:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "/9j/4AAQSkZJRgABAQAAAQABAAD/4gJASUNDX1BST0ZJTEUAAQEAAAIwQURCRQIQAABtbnRyUkdCIFhZWiAH0AAIAAsAEwAzADthY3NwQVBQTAAAAABub25lAAAAAAAAAAAAAAAAAAAAAAAA9tYAAQAAAADTLUFEQkUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAApjcHJ0AAAA/AAAADJkZXNjAAABMAAAAGt3dHB0AAABnAAAABRia3B0AAABsAAAABRyVFJDAAABxAAAAA5nVFJDAAAB1AAAAA5iVFJDAAAB5AAAAA5yWFlaAAAB9AAAABRnWFlaAAACCAAAABRiWFlaAAACHAAAABR0ZXh0AAAAAENvcHlyaWdodCAyMDAwIEFkb2JlIFN5c3RlbXMgSW5jb3Jwb3JhdGVkAAAAZGVzYwAAAAAAAAARQWRvYmUgUkdCICgxOTk4KQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAWFlaIAAAAAAAAPNRAAEAAAABFsxYWVogAAAAAAAAAAAAAAAAAAAAAGN1cnYAAAAAAAAAAQIzAABjdXJ2AAAAAAAAAAECMwAAY3VydgAAAAAAAAABAjMAAFhZWiAAAAAAAACcGAAAT6UAAAT8WFlaIAAAAAAAADSNAACgLAAAD5VYWVogAAAAAAAAJjEAABAvAAC+nP/bAIQABgYGBgcGBwgIBwoLCgsKDw4MDA4PFhAREBEQFiIVGRUVGRUiHiQeHB4kHjYqJiYqNj40MjQ+TERETF9aX3x8pwEODg4ODw4QEhIQFhgVGBYhHhsbHiExIyYjJiMxSy83Ly83L0tCUEE9QVBCd11TU113iXNtc4mmlZWm0cfR/////8IAEQgCUAJQAwERAAIRAQMRAf/EADQAAQABBQEBAAAAAAAAAAAAAAABAgMEBQYHCAEBAQEBAQEBAAAAAAAAAAAAAAECAwQFBv/aAAwDAQACEAMQAAAA+mAAAAAAAAAAAAAAAAAAAAAAAAACQAAAAAAQAAACCogAAAAAAAAAAAAAAAAkwJdQcPLtTu7nOqAAAAASAAAAAACAAAACSAAAAAAAAAAAAAAAAeQ5vj68Lm9GuuTGXCT1mz6b3nIAAAAJAAAAAAAIAAABJAAAAAAAAAAAAAAABwMvylz3ratp2MunrBNhm4Wp6zZ9N7xAAAAJAAAAAAAIAAABIIAAAAAAAAAAAAAAB8O8+l3N6PO1nYS4hmSwcprPG6x9Wbx3eoAAAJAAAAAAABAAABIIAAAAAAAAAAAAAANfL8X8unrWOnQS7Yy6uRYNaaizwjpz+ht49j1mAAACQAAAAAAAQAAASCAAAAAAAAAAAAAADz6XxDj29HxvYrsUzaqSiMdbFnhW8ehdOfv2swAAASAAAAAAAAQAACQQAAAAAAAAAAAAAAeeZvlPH0dhnW1NszsdS/c4udYc1jp5vqX+3D6G1IAAAJAAAAAAAAIAABIBAAAAAAAAAAAAAAOFl8+4ejZ51tk21my1mopjXTWNHOV4b34fbWswAAASAAAAAAAACAACQCAAAAAAAAAAAAAAeWy8bx7153sTpk3dzJqF1ct0xU8j0+se3nybJBAAJAAAAAAAABAABIAIAAAAAAAAAAAAAOQx18o57nO9ZrG3murjds+eN6Hry7fl036Xq9B9HmgAAAkAAAAAAAAAgAEgAgAAAAAAAAAAAAAsS/P3D1bCM653hoc76+4rsuFuax7N3vHT9eIAAAkAAAAAAAAAgAEgAEAAAAAAAAAAAAAk8o5ejWcunSM11yc311xmpmWaua5TWfQ+vHo9ZAAAkAAAAAAAAAEAAkAAEAAAAAAAAAAAAA8JxvL5ejfZtxOSuu3uMqSzLaXy7ePRt8vS+mAAAJAAAAAAAAAAIAJAAAIAAAAAAAAAAABJ878utvPXscbvM65re6zbi9JkHj+8dfvHt/TmAABIAAAAAAAAABABIAAAIAAAAAAAAAAALB4bx6+bTfr2Ols02uu+uNtnO6mNVXztvPodz9HdeQkgAEgAAAAAAAAAAgEgAAAEAAAAAAAAAAA1MvI418uZ361jv1Mm6Z2lmSmBL5xueO6z6kn0n15CQQASAAAAAAAAAACASAAAAQAAAAAAAAAADWS8rjfybjfq2O23a6643bMFlPNtTwnU95ufobpykEAAkAAAAAAAAAAEAkAAAAEAAAAAAAAAAGozr505dtNL3eel0vW9GxtLnJTkV+cN5+yN8uq1kAACQAAAAAAAAAACCQAAAAAQAAAAAAAAAWzy/h35rHXdJsrMitqyLRgy9L34b7WQAABIAAAAAAAAAABBIAAAAAAIAAAAAAAMCdPN59XA4+bPzxyEvVmGbZUV2WJbRYOi9E3jgAAJAAAAAAAAAAAAAAAAAAAAIAAAABhOnJz28Vj381n2TXZz5nQ4+eS4KzKiLllsgxY896fd7i/P9B38ytAJAAAAAAAAAAAAAAAAAAAAAABAANa6+az7HPT2Y8qLpbWqztufxbueGXZcMyiRCrRqnfh9/c1md7Fw9Q6/I6e+PLYAAAAAAAAAAAAAAAAAAAAAAAAg5h7PLc/bhbUsLUVlJBkOfcc/i5s50rJKYk1VZJyHT62neuAQbXpy7x8rsnzwAAAAAAAAAAAAAAAAAAAAAAAPBs/paHSCstSyVJJBbtznLt8fJ2ufPiy2Jb1ka1yO/qaa+qmXJvnw56aDJuarOovi9MfGqQAAAAAAAAAAAAAAAAAAAAAAaJ6fF8/obyUGUlmaFRUkCqStM6cs+cRit6/Xe21cYyrxrY1mfbSXLLllR6Ffkdm+cAAAAAAAAAAAAAAAAAAAAAAPMp9nip9CIvVUUS0FwqSSSmoAJAJJOl38fmef2KShbtlyyTd3y+tvgwAAAAAAAAAAAAAAAAAAAAACTwrP6XBdKYzbMZbcsrcSsrQKgpJBIKhApIFVlVlEZGs+uvgbRwAAAAAAAAAAAAAAAAAAAAAGsdvEc/pCFybjEzuhbiVlRUklVSAABEAAVUlpaTLue3vzO7fLAAAAAAAAAAAAAAAAAAAAAHET6HnU+ziS5NlJblsrloKiUqKialBCxAkACq0gsS025VxuL5/XHwAAAAAAAAAAAAAAAAAAAAAB47Pu6WezGlugxyyuekldiJKiqwUkhZCQskpUQKsS2ZczWR7lfzFaAAAAAAAAAAAAAAAAAAAACk+es/qcpaImyswJbC7BK0uVJUSVJJABIIBJIshcaWmXK1ketX4G7eYAAAAAAAAAAAAAAAAAAAActPX5FPv5SW5a7KjGlxlvF+5uklRUlRIAJBBBULAMWaoly9YpX0e/H6988AAAAAAAAAAAAAAAAAAAAeVz7HIT6WUlMWlqrHZsTdRlWXkkqK0qBKASCSBQAwM7uWZFlJ21+b6A+SAAAAAAAAAAAAAAAAAAAANE9Hnufp8+9l1q3LXZiy4stdZNl9JKytJBJVYJEACKA1mel9MnWRvr5PV3woAAAAAAAAAAAAAAAAAAAABJ8s+X6O26eva9PXDWJFhYMispBeQVEgmyQCQCADV56Z1xWx39+V2l+fIAAAAAAAAAAAAAAAAAAAAAPAPL7ejZ3Oppt+/kde61OlusqzMQZCCsICrBJIACCxjPVzwZ+PH0/Xjtt4AAAAAAAAAAAAAAAAAAAAAFB4P5fZ1TO+1m1bwL3c7r6jTLTY2UmQklRIQKAkAkJFvXcPh7xzxI2PTHQ9OYAAAAAAAAAAAAAAAAAAAAAtS+HeX1dbc7necVvjHs43X1srUyUzLKzISCSoWSCCQSAAdZw+HvbyoIrrO3EAAAAAAAAAAAAAAAAAAAAAYsvjPm9XV3O21nCb0GvTwT7Fus6y+ZtzdKCorIFkgkAEglnpuHxugvKUw5e19HAAAAAAAAAAAAACAAAAAACQAa6Xyzzeno9Z2VmudLG+vnz7WlXZ2XjYXNRJJUUgqsAkAAqY6Lh8jodcbzOnxv0H0+ekAAAAAAAAAAAgAAAAAAAAEg0GN8Rw77jWcqte6ZPXfnE+1oZrOszLM9BUSSQBRJAAAmd/x+T0muF9nW512ffhkWAAAAAAAAAACAAAAAAAAAADkefXTcumdZerUum27b8yz9rSTVdm5syEkkqAFAEkAAiY3vH5XTa4X2caXe9Oe23gAAAAAAAAAAQAAAAAAAAAAec8O+bLXVw1DpsO3XzjP2NM0TfWXEFQKgRUpAJJAIE59Fx+X0N43GaUy9TpOvIAAAAAAAAAAQAAAAAAAAACzL5h5/R0es2VyLNNOtfX08HPqaaaybNvZWhaglYIJsEAkKQUTl1XH5m9vCtKS2dl34SAAAAAAAAAAQAAAAAAAAADmsdNFx6bPUtCtHOsdfbxk+hpJrb2ZdlZUSVJIKSqgQSsEoKJy6vj8ve3jVc25ZNt057reAAAAAAAAAAIAAAAAAAAABxXHtaxrN1K6tLqmtL0+lyc92tl3OpfSoqKkkAgmgCSFBNzPLd4eLpdcMm4plgpTru/EAAAAAAAAACAAAAAAAAAAef+fvkLl2KxprJ3nzq/Z0E9VNbCr6SVlSCQUk0BBWAE6t4aePj2TGcxJBrs3u/T56gAAAAAAAAAQAAAAAAAAAVHkHl9PQ2Z2pFuqm+57ebwnP6LUztnVfq8kguJJJBAoAVEoB3D5m9z4OR5encXlks1GLG9689xvAAAAAAAAAAgAAAAAAAAAxJfIPL6us1jP1LduOve9vL818/wBVQuw0rLqSC4SkgpJoQCoqIJTqHh9OvxvNOHsyJNjcXksRk7z1PXkAAAAAAAAAIAAAAAAAAAOZxvhuHfornN1LFudvPT74/MnP9XsLMirhUlRUVElSCgmhTE1UVkA2Th7Lfz2mz04Pj6t7eWXc0Rinc9+AAAAAAAAAAgAAAAAAAAA4bn15vj16DWcqzXN9z28+K3874/S7bSorK0kuAqJQULKTVuWoqsuApCe638zWnm/H13o2LFaabOvRfT5pAAAAAAAAAIAAAABAAJAAPN+PbV8+nQ6zXUadz18/KT1+G4+/udIKisFxILhUgtrUSWCorsukkBPX78DbvPgTfnHD1768su51ub1HXlt94AAAAAAAAAgAAAgAAAEgA8u497ON7zWbK9L15bzXPgM/Q8pz9rNtqKrKwVoLhWQlC1EmMslxLtlaUrWndX5fdPmDguXpxs62rnZlz946jryAAAAAAAAAgAAAgAAAAAHkPD0Z+dbbWcJr0fv5R5ln6vBT6t1QJsqLhBcSsgoKioxlpi7ZesuJStadDfH6m+IMeXzHj7N6532bR2/fgAAAAAAAABAAAIAAAAABJ4p5vT0NZtmXudl04Dw3H39K9eSUKILiVVJdSsFoqKiwtqWuzIS/c0LKba+f2F8CAcZz763G9kxrZr0X0eYAAAAAAAACAACAAAAAACDw3y+vsNZV1fXjudYx2vmzl+o2Gl1RQQXQmRZWVEFskqLS2pYMm5v2SlJcZ9Yvw928wsr5dw9W9YiOl68dtvIAAAAAAAAgAAgAAAAAA1GdeU8PT1dzbr0fv5hzM9Pg/P7+31YakoJKi5ZeSoApJJLa0lEtdmVc3EpJTsnzvQ78oDiOffW43sWc3WOr68gAAAAAAABAAIAAAAAABx+OnEce3Sam03jr+nGTy/H0uo6eDks/S009Wknqst3ErLyV1WkkEgpW2IgrL1k2Zjl0d8PfvmZbIFhfKuHq37Ep3vfhAAAAAAAABABAAAAAAAB5ly7avHTd2db147nWJMcxdKCS+mA7amd8KdrTV9nKYv3GQzUzls1pjLaaxpqwt1NreGfcYpmRkQAOL59tNjpfk7vv58uwAAAAAAACACAAAAAAADxLh6djNZienejy1A15h259lRIBJUSVEgtlZUUAEAFBjmSZWQAx5fKOPq3jO23z63pyAAAAAAAAgEAAAAAAAFZ8/wDn9XQrsbn0zv5oBp1r1MoAFRJUSCogpJAIAALZaLhm5ACTyrh6s+S7rPoHbgAAAAAAABBAAAAAAAAMGXwnz+vrrMqz03v5YJOct2Fl0kqJJAJKiogoJIAABJaLZdL8XYAqPKOHq2Emql9X9PlrAAAAAAABSAAAAAAQCQcRjp57x9HRJl6nb9eG11ig57Vz0ulwqBJIJJJKASQACQC2WysvxfgDDl844+rIkxTsuvHo98wAAAAAABSAAAAAQAADyzl25/n23tl+um3x6rpyxzVaZpUSVEgkkqKi0WS6VAEAkAtlsrMyK4AxpfMOPqrig2WsejdvOAAAAAAAKQAAAACCAASeD8PVXnezq+nX9OPW9OOrXX1s7KyoEkkklRJhloyS4SQQSCQWS2ZJl5ACTyXh69VNZ5iJ7B38mVYAAAAAABSAACAAAQADVTXz/wAPX0a4xknoHTz9p0488tOpnF8qBJJJIBiGOZJkEggEkkFBSZcXYAA814+m7Zos9JPSevm3GsAAAAAAAUgAAgAgAAA1M1868PZ3qdv14cJz7dxvj2PTjorbllReKyokFRIBixYFZhWACSSCguGVkAAPKuPq6LXPyTHfqpPZPR4wAAAAAABSAAQACAAAC4fLvn9vaL2nbzcLy79hvl6F04eZ46d115wVFwkqJABbMiMSNfbsLKyokAkktxdjJAAB5dy9PGZ6ck16kx7F38dAAAAAAAIIAAIBAAAAAPnnj693neq3jfR3O+PW75fNnn9fX3HddeW+3i5UFJABQUGxypOYt2Vl8qBUXCDm+XTlue/WO/CAAAcNjv57z781L3lx7N38dIAAAAAAIIABAIAAAAAPOcduG5ejGrsd8fUN8MFfBvP6enTaWZqbTWc7WcjUyLLtlVWi5GxQc5Ll7meWY1Od4Gda7GtVnUGyufSu/AAADj8dvOeffKPXO3k2FyAAAAAAIBABABAAAAAAOPx0825end3Pr/fycdjp5Xx775NlZskv2UkKIjHMGX1D0efIsHF46eScuxcmN0bCzJS2tqTpOnPuevIAADXTWBNbzWLiAAAAAACACAQCAAAAAACCSox1vJwWOnn/AB77izZpmpfpEEFJjGTc+i9+IHM4349y73JckzK2KXooLVnc9ePT7wAAAAAAAAAABAAIBBAAAAAAIAJAB5xz68Vz67c2qZ1lYikgpMJe168d/vAGlzrxPj6JlyDYVnJdILSem9uG01kAAAAAAAAAAQAAf//EADgQAAEEAgEDAwIFAwMDBAMAAAEAAgMEBREGEBIhEzFQICIHFDAyQRUjUSQzQBY0YCU1RIFSYXH/2gAIAQEAAQkA/wDLACToDI5TGYthdfvt5nxA1bFr/qCf8Z+PCTspYXHfirxa2Ym2IMbk8bloDNjrxBB0R/4Fyj8YsPhcpbxdKjmvxX5Rl2uZ/VBO90znOjkYJcVfmco6LI4BLI+S8Ax7QIJrsW3On4h+KuQw0bYciaVutfpVbtWX/wAA5nz7HccZYqQK0RaiZI98kAAkLQYjDI1zU2GWbCTzsWQc8zdjJYYJNyOLJqji6GRokZ+WnL3O/CjODC24YrFsj5/7w15YszdmnyVlyo4+aSJvYytxgzfzPwqcv2x+G46YMfYpuUXBIImEvTeGwx7cBHx6CJgEp5PiYPy/qxDF2mw27FaVn4a5WS/xSKGZ/wA9mLzcdhstfJxNCWzM3vbQxcUADAyCtGP4ijYG67WsaPZuvClYHe7ZKxmlIcL+J9aB5DrsH5PIRhzfwWsGTEZuLv8AnvxWlkj/AA+zYYuNRNa5uh4Gu1V27CYz2TIl6Tv8Oj8Is8qSPYcFziiYroexv4GTEycjZ8/+KkUk3CbcbG4mu2PWo2jbQVWPgKLShaCixulLpHWynjRXNaPqR+oGfghYLMpyKn8/+JksEfCMl6ofExjmuaIPuYq7FGD4UAIG0NklPi8FSuij2Xve5o8g8iY1+JuFy4ryB+A5HjsiHhzHAOY7538ZJ3wcFmeGY+U2sPQlUWSsA9lWCDMXISDZxtDKUrQ/tysl0D59UNb3E3MvaO4qdcYt8pEl64KEUModCM1A6XF3owOHcbdPnRZnkosbFRpxtPzvP6jbnDMxCWYCmauPsQF1jK0sLWfPO6fnN98dWcYurlJ5WixYx+IyLp5ZIJVfikDhFGcpFlbTp64v4zjd+0wPsP45jMpTrBl+5PGHEtVDHCtFWY2Q+CR89bgbYp24HN45YE1bJs3axNPIsZ61eDDxOELJIzXrR1fSfSou7LfcQXl7gSjUitNBL247taB67o2x+Am/fKdqlAZrcJ388DogrO4U4jMQ3WMrExyFpUD+4DQuRONdzlQPfdITna0VXe1gJRkZ2khTv91yPk0fGsUL7qXD8g/JcVxF2QfP/jRkbUT8bFA+eVrrDZ2Km8doWQsuZUl7BirlSKzWbJNZnjbGGxxRAelpwDyw9jjN5avxQ0cJiIyvwpsd3GpK5HzzR3OAX4gn+o081d1xu2L3G8a/uqv1C96mIP2uJrUHvDpXQzYyHZjdNm8dEDt8Mz7jHA1nhfiVYEtylTYvwet7qWIC756zI6OvKWO5NXEWIq02M4DlPQyd3FTPoTsjkcx6y+Ho5Ete9tHFQ15CyWnVw+FbYEr4qeKja/vdE53a0q9fqUali7ckffs5ObKZGyPwkmdBm5ID8/mO41XMasjTZb5HjWOGarsp8rvxMfi8m+3SpyzGvN3taoQ3wVXhY73a4Nb7CWQBfiC10uCl7hUtD07TV+GxDuYQBiHt89k9iKMhZTJQ4vL17Uqs1rk77c3fXgayKFjFVtmNwD1UnY8AhzJmge8llgafMbC/b3D8TY3/APTFpzDA/wDtTaX4Q0ppeQXr/wA/m5oIcdYMp5nzDI5+4yvQhr40OypZFTZU+0AD8i5zT4qUZmu+ySvRlOu+ZtZrdJ3hoAXKsczI4XIVJH4OhNlrkNGg3iXHq+DxrYIo/nrEMEleVtqPJ5SpBDJj+PU6+OaxzfDYPIUFbbVFUDXJkYAXYNp/urcXewtWCxuCgrG1j8WPncjlcbjIvUvW8lzmTImWrQhx2PeIC6ax+WDj7NhAO9CL+VX/AMLsHuECVrQ2Vrb1MzbSAq1y3Re98Bwedr5eAuDPmACVcyWPogm1bv8APcfWb/p6dzl/IrZfq46N8sjpZHNY6N7XtGHtNcBASIk5vboJrNhRjT0SEHeU8/Yox42UQnxAu0nzz1zFPXmw/P60hbBlY45I5Y2SRyfK5LL4/GMBtTZbkWQyLi0ODWN9mzsDnxoRN/wYwjGoHmJ+iamTEsWntNlrio7EYb5LHsc8Fp1sp0faNokdgXqN7V3t/wA5iz6VRzWG0QXBoT2klYnM5nDP3UmxfNsZcj/1EVa9Rt6/L2vkuS5/+lxMgroue975JHlEFEbLUOrmg/xUsSQSMLHUnwXYvUYvyYKfUZG3YTHEa2ppdN0EC+Qhod6EoP7jAA0uc67ZZNO6RoLdkkoMATmAhQnslDgiQ1zXg4Xk7JWsgvv+Q8fybd9+Sv27zk0g7Up7I+9dvuERpyPugtIotVO5LXka9slHMVrGmSmRh0Nh0ewjEd+THGAfa3dq1Gj15Mhk5rX2uD3ly1tCER0jO4ELS7i9rU1xBWGz82PLY5SySORjJI3/AB3JbwpYW24GMBrBpRkieViub/JW9InbiUf3dB1IWk172+Aq+Ts1wBHM3kN3+ZX8guHepJcvckbo2vVI32NJKij75I2d1elNPb/LOblpmSXDHGEQEwpqcTpcOvF8M9Jx+O5xc9S7BVBa3cbggf71ORWhurZCadshKcfucUED9B/Q14Rz+6+zGiVolMHQ/wCVgLop5KpI4n40DZAWWt/nbl61uE7EwUjXFkzGo9skbi1VXd1Sq5OcfKCCHXS0fo0u0rXUrtRTAnf5UjvsIQ0W6Kwt03sXVmcfjM3bNLD5CyC5gbXDAoP96Rqc4s7ZAqwa3+2FB9tGMJzvLQgOg6DrpaC1+holftVh3awI7JYEPHauG2tS3qZPxnPLnpYypUBPlrwg7tMEqcO2QbUH2OMa8CCQKU6jBQGh/wAABewQH8lWBuNqYdzOKb+4LCWfy2boSE/Gcyt/meRmIEfvTYw/8xCS0unrtcnEujZM0Tv7ZPBncduYvdDqOgRXn/GitID6NdT7KdwEYKrg+mCVGdkuUrnN+9pilbNFHM34t8kcTHyyO9d9qaS1Ij7gp/22I3If27JCaOyQhX2BkAK16t8Iu86Q6gLSH0aWuhQ+gq0wPYxhOz27AADWAKT30uNzGbBUCT8VzS4avHbTWmIeycPCnaXQbCeGzRNO2OMjCHi75qybFYaEshTd7TeoQQWv1J2d8kLVETK8zFHy4J5+5cMl3jrMPxfPrwmyFGgwsHkIpv8ALVB4EkJTnEal1kSRSsFoaOxjWpr9piHQIfoaWlr6Lr/vEaidvxpv7tr3cuGylty5D8Xk+N4bJvfLPWynEs3jdzVmQ2YLDdxSa/lSNIcJGgkBwe0zN/sTxA/cWjaY3RTSh0CH0D9EK2XC0oW9rQm+AUPHkrjguuylaerX+LB0HFQ03uZHI0QWbsWm2Io5Y5AHMcGBu2q4Ca8oTXbaCUXqM+EOg9ug+gI/WE9u7k7ym+w8DZJaxuF4tRdWrW7Za1rGNYxvxZ/bIsa0fl4lXhHc06ONpWR/er3MDLEC+rNdlb6cse3uAcWoeXIHQCb7Be7wOoH6gVbH37c0joa9XjcYAdasS1oYYuyGPBP78TWHxsp1DOVjBuCFVG+yhHhT+xWTbpmVcA495l0oiUP3NCHsovLnnoPqKHUdXftKouPYxNP2qw3bVx1/9q9EfjbHiraWLb/YhCqt8BRDQVo/aVeZ3V8sU5xjslM1tM8yhOdppKhGmN6bQPTa30KH1H2VF22tTPZSt+1YOTsycsfxt89uPvuWMafRiVcaTfZWSNFOi78NyWZXYtem9RPUB24lSnfa1M6FDoUP0f5Cou+xqiOwn+yqvEWVouPxuak9LB5h6x0ZDGBQIH7VaOmlOi7ODZSUqyzcWtN21yreGkqP7pk326jqFv6T1HuFQO2NUJGk5XCY9SBOILiR8Zyp/ZxzJKlHpgUY0U1yvO7Y3krMNMHCKcSl8xvTxp7go3ajCqt+1zkP0B9B6BH2VB3gKH2R9lbZ3MeFTl9alUk+M5pL/osfVCqxkNC0m+yyp1UmXMx6eCoxqUfbKph/cXd4AUDe1jR1H6Y6E6BKx47QFD7IqYbCwcnfjms+Mzc35zkMzQYY9NTgmg6WW/7WVc6k/wBLimKQ+JFN+5QDvlaE326j9AdR0d+16pN8BReyciNrDSencngPxdmyynVs2nrFxSFgfIfIYvcoeyy3/Z2VzWTvbx5THRkUp8qi3Zc5N+o/pSnUT1RH2hRDwn+yaQQnySQPinYGuY9rXsPxPLZS3DthCpM00LSER3tfcs64jHXCFzPbMpjK6sH96kOyqzOyMBDqP1ZWvfDL2tpH7Qo/ZO9kG+dJ7dtKwU/fUfAT8Typ4fexsCrjQC90E8p8H53I0KhXLLHq8qttVh3iRRN75Qm+Ah0CHU/pYCqbt+xVBxUpdCwOEfkBDSKPssZL6GWjb8VmJPVz91Q+yC2pXaXG4TLZuXXDksu+Z5Qq07w9U2eC5BDqOp+ofRwtp/q9p6zFb8hnp2gV5NtCa7oVbD2jvYWSsmjjmZ8QwAvaCobht3blhQO3rofZXZSyN5Ax1MU6Vaqs/J3Z27OrB2QFE3tYAm+6HQHyh1P6XB2N78o9clxzruO9SJtGy2SNhaWPTelhuwsBN6mMjZv4fI2fymNyFneJj7IYwq/sEPZOVSIWctTiKbsvCnm9X1XlRt7pWg9GodB0HQ9Aj1HULiM/pZl8SBIIIWYpf0vLERthfsBMO+k3kFcdl7bN6An4fmMwi43casc37QoR7dJHaBXHIC6S5bKfIIo5JCovvZAxQAHuf0b0CHQdSgj1CCKB0Vipvy+Vxs2yNEhZ7GnI4yWNgx9kSMaVG5NUo8FY+T0MzScfiOfTEUMXAqB0GqLfhE6CtTFrDoUan5OnBXWdmEGCy8u6/tIQmNDWgDoOoRKHUoIrfQIIohPJ7e5qjlbNHHK1A6Wfo/07KieMQybaCoyn+yuba0vamyNlayVp+G568G5ho1RB0FD7BPdoFYiv+aysZcFziX0uKZQ7rNPp9Ah035QKKb1KanIlAoJqKC1trgsBZFnC0H7WVxzMlj56rjRneNskbE/YXuFaYNFYCYy4irs/Dc2d38grMBpM0wKPwFYfoFcerejjxM4L8SH6wVWNV2H0WhEoIdAgegW+hTU5O90CmoI9eFWdfnqhPTlNA1rkeSjFaXYHljtqduwuMyFrshWPw/J5BJynI6NIfa1A6anxOszw1moNa0BrAvxKd/bwsSjAAQaj02h7dB7oFBFFNRT0CgUF/HQLizyzOQDfS3Uhu1Z6syrCetPNVnELthSDbVipPRzVb4Zo7nNCsvFnM5OwDTaAE86auPQmW/NOQgCTpctykWTzpfEmHzroem0ESmoe3QopqKf0aU1N9kemDlbDmKMjjrXXldL05K+TYK0uwEw7arRdEWysXc14a9p+EdK2Fkkrli/uijJUAIarDwGbWArGDGROd0uWDVpXLIVVoZRxLSuzRcg7o4pvQeSgPCHQlFAooohAqMpiPRmi4gjjls2sRXLndLNaK3WnrSiAz15ZIJlA7asM23awM/q4qEE/CchnNfj+XkWNYGtA1F+xegbdqtVB8ezQuZXhR4rmJFHDaOIxdsxMcJIw5Dt6aQCJUfhE9AUVpAI9CEWqMkHRQ90ejSe7wuGXNXblU/Ryul6U8GQY2rL7eT9zffjk3bbvV9/Cc6n9LBRxBY0fsCJ0PB45B33LNo9fxN9W4zjmFiL8bDHH6DWTcOdp5pWJuNZmM/bVsQz1jqxAx/cNscHn+R3N/wAtP/7/APvYW0Sv/raLmra2f8ed+w//AJ5QeP8A8q9K7aOoKuP4rM6Vj70lGCGtWEEEfW1VhuVZ60qj9atPLWnEDu5qrTitlacxJBBIPwnPbIdkMfVBxrToKU9rffAQ+lia7iOlkRdjXyNeRINKMEN7SvSG9oFwGgbGHxFnZnxsvDuPyD7Kr+DY4genf/6GYPbKx8JrD/cyTeGYke8zeIYQe8J4rgBrVBvHcED/AO1swmEb7Yl2Gw59sW7A4gnzj3cbwp/+A/i+J/ikeMYv+KTONYwHZp18ZSr69Ou4nWtvKhPk/Ty2gAYckwUpQQBu4wSRuG8fb/OUK1g/CchnNjkmUKoDt0rZJic0JkTYY44W9csdUJSqcvqtKAKAH1a6uK/nqfqciPKiP3D6blOO9Ts1JFi5XOH3CYjsXFbBJv1T8HH5kYpJm2MrcmCp7G1C0SXaDCHHbieucmLa0MDRTqivAxm1v6NfQE/3Q6b+tyPume7kDsA/Q06IKbE2LM5ZjU8/asBOIM1G0/B5S5+Rxd+3ulF2FqgaCwFMd6VipMS4acR0CtSCfIF4TZO72XugP0An+6H6Liv5Ufu5Qu2z6GkNPcVScZTLO5P8DSsOez74zFMyxDDOz4LntksxNSuDVZ9oUPgNC0HNWFtGzj2dxUsnpRSyKnHqJvchGAhsIH9AHynO24ofoFFD9xTDpryq/wCz6Mk4sxeScFTHbEwKR4Kk9lxa0Jca+sT8Dz6yH5ajVBrEeAo9dqYQB78Zk+/IRdLQ3WnCi8aQ6D6R0AROig7ygf0XIe7kP9sKMab9F6B9ihegYqMrZa8cgM0rWgEovD2ghcct/l8zGwn4AAkgBZ64LfJctMDV2XBTzNgjDioZg8BcYG58jJ0ys744ooY1E7wxMPj9EKeUMHkhyYf0XkjyhsMTW/tCA0PoB0QVMyOnk8lWaMvLKKj/AEWMLewBpmLxotNC429Sr2gP+fnsm7E4TIX2ti1FOIyajHePHIvX/pFxsCo2mywiVg4c9z/6pvpkS92XpNTW+GOCjdsIfQPpuxF8Di0E6cVE79EoDbgmN87+rkLGsz91YLE0cnJfjtx/lZMbm7uKfJda1gO1x6B8GDoh4/5+fx5yeDydJqrTiS05xVIhzNriuNrtqsyckNphjz2eYVwtwNXJ9Mg3d9r2prR3PYgNFNPQfRvp7Jwlmt1WBXAW2ZdmFyB/QcdKNhQGh9XK7TY+SThcKeJn5JQclmyvIchclWU2YUGCNrIx8BF/uxqzX/LZeaFU9moQFxHl2GykEGMa/N2iOR8gewfh298lPMvPS3yeCpzTJQWXiHw0tXnWyAEFpAfS521A3TA4rJM3GyUKF3lNKB+rRPsp5a1OF9izJTkdNVgncz6uW1ieTsDYM1cyUVKShSlwDIK2XZXBtOY6Os4iT/cf8Dy+AVOW5NipkiltWvw6tZWqMxgbuRxWRZmZ4Jn8Mpvp4aVrnMIDwXFk3563ZsyLEZHJYxjWVJ6nJqcrQLUEMlS0N15/QIXY8fxo/wCOm0SvdzWDo9jZGPjcmF8cr2PDHeECgVtacfYNjcfdWJalNnfant8rrt2yjWsTW787HWJuwMAYPr5Fl6Mebq12sys8Ek5ka+rhmy2vWbNFVtXZqtGs15Be4j4Hm3DbOVsNytBY+wySIwlZBmQoF2QxVrhuZtcq9Stm1VrRVa8cES5Ba/KcfzFhY6MNY0KEeyiCaxhIJbDbuRaEdpmavM/cxufGvvpMzmNd4eBlMU7YNsXcY79tx1/FsG3XqU0FxosQSdM5Yq1LFN8z4IXvY1zQ2tLr/bmkqVhuxal5Px2IkNvycwLvFTFzZnN2P3XPQiDjIQ5YKIS5iiHfoclGEpzR2pcXPcuZVjYtYXGuMkVClHjMbDjq3pMPwO1yHikeQkkvUHWJrlCUR5PH8dyb6mVrPrgjRIXPpuzi1lipN0GqHag90xD2R91sjab5RO9qQhTvIae0VKradStWb153GHYqlIu18btxSGKSbYlnhoVQd+lA1rfYN86QP8J2tpxC4pEXX7Ev6F/D43JPgdbgh4pgIe8NqwQQV4hFXh+EbI9o0HQ1qsEs0sFVfiJJ/wCl4yFVRoDxEFENHabtAhFeECNolSEAFYGr+by8T3DrzGIv4/K5EeU0BNPsoXbGkz9oXt7py91xOHtpW5vkvxEf92FjVfY0q/kKEJh6FefbWvZEkbUz9NK41UNfFtlcOvJGd/H8oE/3KYmn9o1CExApxRHlYOEw4ei0j/l//8QAPhEAAQICBwUGBAQFBAMAAAAAAQACAxEEECExUFFxEiAyQEEFIjBhgZETM1OxQmBykhRDUoKhFSM0Y2Kywf/aAAgBAwEBPwD82yKkVsrZKkR+RAFZvED8hAb8kUQT+QSgCtlbKkpBSFTghenCRx8XoCZ8AoiRTsfCbduz3HI3DHxem3bhqFbrkcfF6G4ap7ksfCabFJSRBrDc1PIKdTWycE61ztTj0H5gncZqUpBWSmUXsGaa4OFhRCBsmnTDe6JlAxJ3yU6p3+WPgyIKeO82ohpvAQErlJSVyn5VFOMmnz/IDXBzJdQrwiEKypVO6J/EfyA3NNuRQCIMt1yiXj8gN6DyTa5OyUnoMcVd1qKfcPyALyU2xyKDiFOYsKJfLoE5/QH13HcJx8IWNNTbQEaia3XVO4Tj4QExUFeiFJSRR4V0T+H1x9omQg0NCYCRtH0qmpouU6rwQhwpxmcfHkms6uU6iVPcCdMGU8ehQYsYyhsLlA7PEMh8Uhx6NFwUdzjEcHADZMgAJeEQHXqPAdB2TPaY4AtdjUOFFi8DHOUPs2K7iexv+UygUVkps2zm4oSAAAAGQsCMiJKmQv5g6WO8A1shtdCDHtBGwAQVH7Ne2boJ2h/Sb0QQSCCCLwcWgwIsYyY3Um4KBRIUG0gPfmR9lah1U6ppwBCjwDCfZwm5SUq51Sqo8P4kVo6C06BNzqj0eDSONtvRwvUXs6Mw9wtePYp8KJD42ObqMTolF+O4ud8tt/mckAAAAAALgPAiQ2vaWuEwVGhugv2Td0OamgZ1AVTCmqNBMJne4jfuG0K8EdMlSaEWzfCEx1blpiUKEIMJkPIW6m+oX+BEhtiNLXCYUWixIdrQXNzF6BrJUOFEinuNn59AoFGbCt4n55aVl5fSRCabIY2n6m5tclJUmiNjTLZB+eeqIIJBEiDIjD6HC+LSIY6NO0dBWLxr4Mk+BDfxMaUaFBycNHIUGF/5/uTaLBaflt9bVKp7thj3SJ2WkyHko1JZCgfFBDpjueZKoEMsgbbrXxTtuO92hCk5sQdbDh/ZsOUJ8Tq50hoKj1XUarPXkP8ATD8WQc34U/7pZb9Lh/EgvAvHeHph8CH8OFCZky3U2mrKo3nXlDUL0VSYfwo72i6cxocNo7PiR4TM3ieivM0V5ZorryoqzXaLO7CiatOG9msnHc/oxh9zYskV0RXUaIco2/0RuXRUlm3RooyG0PTDez4exRtrq90/QWBdEehVxXWSH2Q5QXp1+iKABsNxsRaWuLTeCRhcibBeU1gY1rBc1oHsguhV40V4U7QfRdOUbYSfJf8AxXlBUxuzSovmZ++F0GHt0pmTe8fSoIXq4q7So8o0yBKNgl71BdotlGY7Nn2wvsyGQyJFP4u6NBuHoV5eyNztDyrLifOxFGrtFs4cN2TiPfC4NKjwRJj+7/SbQoXaEJ9jxsH3Ca5rxNpBFQyqMzbyrOH3miZmumfCEB7XvAJE2jrMYaHFpmCQVDpfR49Qg4OEwQRUbijyg4G1PeyGJvcAFHp8Qkth9wZ9SiSSSTM4cSmvfDM2OIKhU7pFZ/c1Nc14m0gg8q+kwIYAL5kC5tpUSnvNkNmz5m0ouc47TnEnMqJxnDzUFR/nwP1H/DTygvT73amoKJeDhovFRqaqP8+Bq7/1K6coUan8I1w0XhFGpqg/8qjj9X2Q5N3C7QpyNRtacNbxN1RraoBnToIyB+yHJxPlv/SU9GoW4azjCNbVQ+9TZ5B/2Qv5N/A7QpyNQRvOuGQhaT5I1svC7OtpLz/1uQ5N5kx58ino1BP4sMYJN13GXhdmDvxTkz7lDk4vyn6JyNb7gcLAmQEdxl4XZo/5GrQhycb5bk9Gu+YwuGO9ojWFD4guzvkxXZxPsEOTjSEMz6kBPR3IgtnnhUK4ndnsscfJUASooOb3FDk6a7ZgA5RWFRBajuPE26YUzgG4FFMg1vqqGJUOD6n3KHJ9on/YYM4n2Chu24QzbYiNwK7CQJAaI1tCe7acSqOJUeCP+tvKdpmyAP1FQX7D7bnWFOEkawonFhAEyAijW4yY4o3IDZDRk0D2HKU9u1Rw7+h4/wA1Q3bcPzbYUdyJcDhEPiG4FFPC31QEyBmU7iOvKR27cCM3NhPtbVCfsPBNxsKeEa3WsOEQrzpuAJx2nEqjjapEEZxGo2nlOqILSW5EiqE7bZI3tRCNQwiF1RqCeZM8zVQROlwfIk+w5alM2KRFGbpj1qY7YcCnBGoJ4k44PC4TqjUFEM3Syq7ME6Q45QyuvK9os+VE/tNcF02lp6XIhFBRBccHh8IRqnIE5V9l3xzk1oQ5WnCdGd5EGtpLSCOiMiAR1RqeJswdvCNEaohk0DOuhQjCo4ne87R8suTNdJaXUeKBluQXWFvqEar7MIKCiGbtLKmN23sbm4BG868kUayqXD2I7pCx3eHrWCWkEIyNoRQTxJxwVom4VzkCa6GzbpMEZOmdBahEY572Bw228Q3z4Z3Cu0GThw35GXvuQnWFqNUQWA4LDHeRqiGwCvs4iGKTHP8ALh/dB7p7cztzntdZlQ+0pACLDJObEyn0V17yz9TUx8OJwPY7QgoiV48KW8R5J8aFD44jR6qN2g2REJsz/Ubk97nuLnElx67gJBBVhEwinCbTgsMWEo1PM3GvMZ7khkmRo7OCNEGjim9oUsXxA7VoQ7TjdYUI+4X+qO+g39xR7TidILBqSUe0aTlDHojT6UfxgaNC/jaX9Y+wX8XSvrvRpNJP8+J+4r48f68T9xX8RH+tE/cUKVSPrP8AdCl0j6z/AHX8ZSPrPRpdI+s9OiPfxPcdT4EJ17UUE4SJGCMHdCO6LwjzgOyQUaoouOCNFg0RTuF2m4Oe/A3QVRBNuBgTICFRtBHludOeuAGVZEiRgUMd7ceJOrPOt4m6o1xBJ088ChCwnci/hwAWEaoqSkogm3TAmCTRUBOqL+Goc+LWt0Tb63CRIwBg2nAVhFRvwVC48/D4Aojy3Zkg7aYHIKIZvOAMOy4FdUVFcZ7INgTeBmijXt0wGEO56qMOD1TWhrQEMCFoafJdVEY4Em8EqGO43RR726VbBLARfy4BJkPAhHuX9VIEgm2V1WeBQzNjV1XxQ1xa8JrhsT6CxRTN/oKpSACcxpRhkXWogjkgCUIZ6qQaPAhsJYT0JsQBAAqcQAScChxALDUQDYRMKI34Ym242EFEkmaaJuGu6WtyWwFsHNbDlsuyUjkVI5FEEX1gEzlXIlBjz0QhZlBjR0rfwnwIe24EbRDQg1rbeuZT3ACZTnFxwNkQiw3IOBFhUQBzDmLaofH4hMyTXDvNVmQU96JcPAa9zZyKMV56okm84NMnrVCvOnhvMm67jOLwYl4xKH18N5m7TcbxBDwHWuPOf//EACsRAQABAgMIAgMAAwEAAAAAAAEAAhEQMVADEiAhMkBBcTBRYGGBE0LBIv/aAAgBAgEBPwD8tUJcm9N4gj+CLLvEL+BLwecbwygg/gJlFm9N6Xl2XcKXnHKUtzX3Ji2PhOZKdfZVn8NLDN1+rKPAYONMNfco8BhbE/AGPFfgauUp6T1r1fSy974FDEacGUgvOJRbKOFstfcmGTgVJ5i3xOct+8CUl0/X4BUWb+I8JL4Ul5R0n4BV4I/BTNnk/gC5v7jxOBKM3X2PgjlGDLzlFwMKeo19yiXqMHEMac8Keo19i2qPgpzwp6v5r9TYZvNTeVWG3GZmFJbX39yqsyp+GmyX15QzhVeVXu/ELSwbmttQTecMpWefjypphX960+OAlRu+vhC7KsBSDfXf0xpsy0txUlj98BydWfgzLMaU9cOcKbY25X4c9Tcn18XJzJu0/ublP2zdp+uAFZVniak4fUcn1PrsN7lxmnvAZHrtnUyGXauoPD4/vav/AGecPOmuB5JmYPauUMvcNPZ5nnBnntHKf9wdMcsXKZ4OUO0fBDnz054D6we1qhp9kgjk9zVnDE04r+4fZ271OChnCyDqBCr7ly1x7VS7N5jKOk1I8+u0MWbPJNNcnAxp8zz2zNn1PrTXJwMacqv52pgynlUabV0vrAxp6O0cn1DF5abX0sOA6TtHJhiwyPWmbV5BDFjkdo5PqGLNn06ZW3q9cDHx2lXS+ocFHJTS1sLDgY+MTsauir1Dgy56XtH/AMw4HKOf87Sroq9Q4dm8rfWlbXMIcBzSOb2lr3IcNDar3pVfWw4KDNj1OB2RKy1XvhZmaS5vuHAFgI59oSsucLNm3p0hbC4GJ1HbGFZZ98DNm800ivpcDGgze4qLkMWUtqjSNrkYGIWAjkw7ess++B0ja5nBSXcKsmHbpctwMob0mj7Tq/nBQWMKsodxWWb4s2bmaPX1OIXcavHcpctMsaOVZo7m+8aDnit3uqzzjlpBhSWMTKn13SXMWUN6TRaulxC7jV0sL2HtzgrPOOzzTRa8saDmuNf+p9stN2br2luFLmI2qNF2mZjR0/BY+puk3Zu/ubv7m6TdJuksSx9Sx9Sx9TdJuk3SbpLYvDWecGDcHRKup754UuY7PyaI5uBme9Cc3ChtVoa2FxMzgO5dKry4KW53jwOT64KG560LaPMOCjz3jwOTwUNqtCqb1PBR50FzY4jcHQFsLwE2fnDz3LxV9TKAbxLKYUFqTQEuJjQFrx6n3Nn5+Z7Ovqmz/wBot28dC+/eFKWCVdTNn5wK7Vo5fI/GoF2Xvx19UHOOh1ZuG6pclQ70oLGGcprqphtBz5QR7JQjtDxLtTz+Cp5kcAvoVVPkwFIO82YFiVNqXhKqjzDaM/yH1DaUzfp+5vH2Tep+ybwnLG4Z4qEdpSeY7X6I11ONPUfBVY8c2LAVgWNDqovzIiZkpUTCvp+QLAY15Hw0ZvwIM3KdJ2mR8dBer1wV9Pw0ZOpbTx8dBY98FXS/DT0nef/EAEcQAAIBAgIECQkFBgUEAwAAAAECAwARBCESIjFxECBBUFFhcoGREzAyM0BCkqGxBSOCwdFDUlNgYqIUVHOywhUkY5NEg+H/2gAIAQEACj8A/my5rD4fIHRd9Y36FFzWC8lAqtLrHTUMbDUtc19sYscsgRI07tI19pYBnJAM8F08UqDFRjaY3uRvBsRVj/Ib43E4ZmSaRmMcKyptQWBJp8Eh24fBL5AW63N3rTcnNmdnJ3k8taMiCJRYAnWepTc9JUeNCVtCyStPp6F+ipdNRe6sVZ+rVrE47BGXQ0LmWeDVuCjObkf0GhLh8REssMg2Mri4/kE4n7W8jqQp6ELPkjTMaMszs8uLxBuC8rm9lGwDlJ5TRC3FurfVlLDIG+iKJHloQ2iLZAmiAF11sMuW1X1cr/1apI3A1YKgBOyiEZhdAbat/rSD7PxN8DNyCGdLywSkdDqShP8AIA0gjFb9IBtWsX0pDpXJkYa1/G1beSrA7SeQ0LXokSAZ9YN6LMWpRdWFusi1KIQhVgwvewtVgo/dJ6q+7xEFmF/4LaeXXapHm+z8TNg3d9rqjEowPZ5/AGFwM83eiGwotp6JPSTQvyseTi3qy8orXRTo3JAv3EUcpNa/Qa2YmGYJ0eVS1950ef8AORsNCezLMqmrhTYEixIHmU0SRsrLyOFb5sOfwb4zB6W4S0U6j5m+jXp4LDuu6KRgf9/P5JbE4JU7ZnWrBkVhy+kL8dVHSTagR0g3q1ojY9BqUBZkjkWNdPyscrqskbDlBX5gVdGAKnpB5+LL/wBRwWn1KHvVyEKHlzQ1plCQxYGx6hUiIdrq6t8jY1n0HI8IZuWRswNwG2tNzsQGwFaJvmNl6uWhaisWAQTcoLy3tGPzokLBGATe5y6+fg5AhcDsyCnISUEaW0HRsR8qso5Ok0qYPEm0E0uxxcIWBcoNEE7aOHQyyRxTrnGxjbRO8dYJqzoe49YoFj4Z02AREcqxRwZnGxFPX07Kxsarg3WKZNBg86P92SsqAKoGTCopnH8NCijuJNbRR0JkGkgNhfn/AEhLh5U0em6mrshVuuo5GXZpAG1OUiN0QHIEixqIwIDaJ7uuZv6JJG2s9G2VclOGHXTkDk4Mq0hEAWPuqAb2H8gE4PFytCGUjUaW7FGFZqbcHJV9HinFPLikw8MWnoLpurPpOegBajEssJ8sEFh5VGKt/IDofs/CyfaBYfxpD5GGtWVElG6RQ358F2KG1KGlAW3KpHI1GRwbWjzY33kDKrFvEVuPTwZH7QlkO5ISP+Veqn013SqCef8AabUqiX7UXDgf+DBLa28s9XeAPhJe1AbDxQir6C3PADnVl6FG2mJAuQo0j4CpYkKho2ewPUbDZwXOGwckrj+qYi3ySti2t1xtb6Nz+FcqVRjsUnK/dRRZEMpYHNjLKZvG0a0AuNjSaG/+YiFio7a1dJFKkb6YSACzq7rn3GntfNgzm/zrETEL6su1r0kagkrEgy7+DQw8CF5Dy2GxR1nYKtLOTOyA30AQFCDqULavWxvz/mwtbewH51pRI7lUv6RiugFqsMNjpURl93RcqCK++kgR3NrBiRtrfxToRlpCOQFBqk99W04yoPW1ZHBSOd+gDz/slhPV65KJjiwi2VRcyMrsxHiAtac02LVC976TyMdYdIvcg1lGiop6lFhXfV+HcK2YnDafYMlHZaj5GHBvEp5CToJz/JYxuAEQsSbdWymUIHIiQaYRWy8pKResTh8LgwkMMeJUpNIyIFM0iH0CeReThdNxpzwsiTRazqpcoAb6QUbbVpGSTS032LFHtd7bBRUsEuWFn0V2afQxLFjz/E2Ht96JlBjt/VpZVBg8KXLvPHEEJY/wQALdo1mTck7T1nj7QRWEw0uIYNiGijCkyrz9HCOQMbs3ZUZmpIIbAtM5tK4vYhQPRqaYyOXZpHZ79GRyAHIKy8yp0xZke+gx5CbbD115Gdb+UgZgSLGxI6Rz1FF1FrsdyjM1iJ2JIGlaJT43NDDo37OBQLfja7Uzu212JZjvJq5U7Okcoq4bOI/UeayvTRyrKzo67RpEmvISf5hBeJu0Nq0ro4urqQysOkEc7WYi6RKLyNuFNh8PyRIxBPW7DbQ3108QgE3B2aLUdNRrWFxvo+HmNeU+TXv9I9wrJRwfdE3fDvrRN3ch6xUuFkBs4I8og3FahlNr2RwT4c5g4yZbpfMRp++R9KZ5HN3djdmPWeHp4pDqdU9PUaAYemvKp4thRrRUAliTYAdJo6CjRjvtI5T31meHLYdxohlN1YZEHpBoK+xZ9it2+g842HKaP30h0B0Rrkg8K2G1ZBlvuJsa2cfRcbG5D1GhDJ0MdU7jwZ8Nm5Ixm53CvJxA3EQNyetzy8Ocz+ThB/dXN3/IcVpMLyptKdafpQZHUMrDYQeb9eVfIx9qQW+QuayGVbY1YdxKmsxC58BetoB8fMAjoNTRjoVtJfA1Ge1D+lqhG6E/mansdoW0Y+VBb7TtJ3ngA03VbnYNI2uaZCpIl6UA2+PJVo4FESKNgtt43q7SRdljZhzfqwQl27cuQ8AK94ivS04m3sLj5rW2GQf2mtsafT2BjiNG18tAn96rnlPGsjfdP1CT9DzfcS4glewp0V+QrZJ9QK1haWPtIb2q6uhI3MK/YJ9K2n2Xbl41kcqvJo6EnbTI82kMkD6HbbVX5mvRCjwrJ4ww7jY1nGb93LWqj2XsNrLXoqU8G0fZfetXLwZELOg/tbm3WxWKW46Ui1z87VyV6LlG3PlWRrNQFHZJuh7jcVb/ALh8vxFq95V8fZPfWslX68Gq7+SbdJq/Xm26YSFYh23124PSGVWcbephtq7KDpL0ryirrMyyKek20WrNsRERuIv7JsNZudLu4LMtmG9cxWrIiuNzC/NYVEUsx6Aoua155Xlb8ZvwZMLVqy5jtj9a1W+tW8nMjoe2dBh86yjX57B7JYFxc9QzNWMmSjoUcNysZjP/ANZK812fElcOu6T0v7QeHNcx3VkbZjkqzqbOOvp762DOs3c+Ay9kstiW3Csjkm7hzjxF+5wDzXcYdGml7cgsoPWBxNmY3Gs1GjIOlR+lXPkyR1kC4rYPZLKUu56FvsHWatyW6OrhyeBX+Brf8ua9CdhniIjoSd52N31/1GDl8kAmIQdaHJ/w0rjlttHBcr815RV1baa2Lq7mGXgaztn7JsVQi9LbbndW49J5TwvIiMVlbYgRsjcnmzYjH5UQ2jcMMiKaQfvqNbvHLQNavIOitsLr7L7wUdQArICixHIBc0Z2kjWRYzlGt87EctBVUWCgWA5s/Zv9DW1RwIT+8NVh3ivKL/DksrdzbDVnUqHQ5MoLAEkVyn2RiDI13fUXxNF//HHdV72OdKiD3VFerLx/C3NuyGQ+CmvcX6cQaRwSID1PioqzV2+ZvxOW3n+SuThzWZXG51t+XNv7CX/Ya9xeJ6MOF/uxSfpWTAez+tgbxQhubdmEnP8AYa90cTJf8FGN4l065LcO0+wcork4cjMEO5xo827MDMBvYW4uc+LB7hIqCvePz4dg9h5OH0GD/Cb1kcx382ZuscQ3u4HE2Ams2GG0t7vpmveHDtPsm0GvSgT5C3Nmc2K0z2IR+p4nuGv/AJMCfBGTXID7RnFK6d19IfXmy8eEQQL2vSc+JtxNiGvSxBbwirk9j908XKZNJe1H+o5r1YImkI6dEZCryOS7npZjc8T9k/0rbh3c96oK932Pk+p4utEwa3TbaKurKGU9RzHNWeJxMcf4U12+nF/YuBvItXqfs1R4m3/Gvd9jJ0V02tyKrC542th30R2HzHNXoQySne7aI+S8XVfEI0nYj12+leqwkC+Ok1coHsdvLfZ+Jj+MAVZwLMOhhkRxdXEKYjv2r8xzV6oRxfCtz8zxckXyMe9tZzWTLGvwItbW9j9DCW+Nx+lWixI8vHvb0x3NxbOhDLvU3FasiK4/EL80ZEivW4mVxuLG3Ev0DpNZouuel2zY1sx0y9wOjW1j7HnaBe7WNXnwpMsYG1l99O8VcEXHFzhkeLuvpL8jzRYw4WVxvC5VsUDiXVX8o+6PWrlr05ZH+Jia9ED2PKfDsPxJrDgth8TeWLoVvfTi5OiSrvU6J+vNGc0kMI3M1z8hxeiFP9zVkiM3gL1tVSa9I+xgBcQqNufU/Pg+/jPlYO2u1fxDLi5OxibdILfXmj1mMdzujT/94lzsA6SdgrNF1z0uc2PjXo4KYjfoGvRSw+nsmYGW8ZisnRWG5hfgtBjCWAGxZffXv2jiayEON6m4rVkVXG5hfmf0YJn+JwOJdMOPKtv2KPHgzdY4x+NwtbZFv4+y3KxCNt6ZcADMNKJ/3JFzU0VdGKup2hlyI4mcReE/gOXyPM/q8DH4uzNxNfEt5T8IyXg9bj0B3IjPVtZT4Z+ynaJk+jcP3eIIScdEoGTfiHE5UmXv1DzPcReSi+BBw600gTcDtPcKsqgBR1DIcHpTTN4KB7Nk6Oh+Enh1JUKk9B5GHWDVpYXKOOscu48OUqvCd7C4+Y5m2kCrrJjJSNwaw4coI7DtSZfTh+5wyPh0a99NkfXbxrl9lsolAJ7QtxOiHEf8G/Lh1o3VxvQ3q6uoZdzC45lyjjdz+AE1m2s29szw685Mp3Nkvy4PU4eWX4ELVm2FRjvYaRrZ7LkRY1d4rwvvTZ4jh1JUKHqvsPcatLE5Rx1rw60LPCdym6/I8y5/4VkG+QhPz4fWyBT1LtJ8KsBkB0AcGcmHMCDpaf7sfWmWIwqsbMCNIx6psfpW1fZvWRCUdTJkfEHiaslop+0PQbvGXDk6LKu9Don5HmXPEYyNbdSXc8Pqk8mvak2/Lhs2P+0No2gJZflp3qM4VYxEISt18mosFNKsbA2jnBOj1BluSKWb/RkBPg2ianh/1Y2T5kWpW3G/CKHnBU0m5DbxOVCNA3q0a7neRsoJGhNlHE1JUKk9B5GG41aWGQo46xy7jwWXymg/Zk1TzL6nDtI3alP6Lw60xaU7mNh8hwqSjXRiASpOWqeQng2cBrCyHlYxLfxFSwnpimcfIkisau/ybj6CpO+FaxB7KIv1vWMffIq/7VFStvmel73c/nWHO9SawXfCprBjdCorDf8ArWoPgFQfDUPgai8KiXcorLg2ji2OUM/X+435cG0VrOlnHQ66rcyejijGN0QCDgzbIbzkKyjRUH4Rbh2MnyatntmUsZUHobap7jVmG3eODJdCZerS1W5j94V63FTP4uTwZNiYgfiBraeG7zy2A6lH6mrta7HpPtmw1qpiplHxcFlxEbQ/i9JfmOY7GLDuV7Z1VHia2cGUU8bncGz4lxBqJ3bfadhPEyW7HcM61p3aRt7nS4LOhDof6lNxWrLGsg/EL8xEGfFXbrWFb/U8S7xMYnPZ2Hw4PQRm8BWZFzvPs/JW08XMYOYj4DWwDh1sLJojsPrLzF6nCabb5mv9F4nJFJ9V4P2bfT2nbxdaTDSqu8qa2gHhsmKUwt2tqHmHaauv+JMSdmHU4Ljgy8nEnfcng15mN+wou1bCD5vabec21sHF2GrJFipFQdCk3A+dFpGYIi7M3NrnqAzq4AtRV1YMrdBBuDVjImuvQ4yYePMF3ijAj/1JCETwJokgXJO0ngBnCBo1J9IqwbRHWbUw0s9FhYg9BHIRWV4beDcGQjK/EGrIgeb1lzA3ex+sjhkO9ksfpTMkcC6BVijK7ki4IoyJEUMUrek6OoYE8FmkVpiOjypuB4cwa8kBMfbj11+a1tVCN1ZWoPiZXcxO+ehGDZdDovWz7SnPxnT/ADrP/Ex/7ODWRUYd1ZXuu45jzdoVLSy/1BLaK7i1ekdL4vN5nj7MNh/oaytB9WokT4uQYcAejBHZVFbRWSIq/CLcwe8KsElnj+CQirddDDY/DRiFsLI3pmPVvEfeo2OPYfAirWRxsYHdHwBcJ93Asp9GOSNbHS6FauTV3bR5rIVmRXomx3HzQRF8SehRymtEyoH0eVQdgPHdxNgIWJUXsUZkowrjUIxAtokrGbAE7QMzejopCWJO1mZhcjoFXHlY7j8Qr3jzDtmaVd0yh6GyhNKJiZcPLaGSKYG5KvURxE4jxEjH0CXQaZXR2gMprSMmOme+jo3AsgoADWJPQMzVzPM8hv8A1sTQMH+WlGnF+HlTuqTDv0j72M96i47xUUnYYE+G3zG08GTAg1Zkax48UK9MjBaaZv4kg0IxuG00ZJGIRORV0jayqNlZIAo3KLcd3mgw58uwsFQOQ6KeulkmkbQRAb2vnmadiHJ8opsWbZfqUcgotISrMeRFUglm6BW0nmHTxKQqkuG5ZUS9mT+rqojK3QRWIw2LWwYwEnygHSuYagcdhIzJBi4VETlGIV0dabQW/pG5JJuSTWaYKUL2nGgPrWwAcI0hsPLUoHQTpD+69QSbwUPyNPe/uSA/UCp0P9Udx4qTSL2lZfqKw/8A7APrWGA/1FP0rTiDsoaxFyu21wOERicMumclDJbaeS4NMQdhGsPEU/gagiHTJKifU0s7dECNL8wLU5/rnkCD4UuaEKcqQIE8WN2rSkO12JZjvJueDJZDIdyAt5iLE4/ECyaZIQrGLF3Gw2pIMLp6SwYdBBFvIXM1pso/DGvS55AK0nc6U0trF2/JRyDmNIMbtZDlFOev91uusRhX/wDImqdzLcGjKmLdYZV0bsVJ2g/08HrsTh4/7tP8vN3Y5KOknZXqowpPS21j3nhzXFkdzJUkfYdk/wBpFTOOh5Xb6mk8BWXF9Vh7DfIbeYLmHSC2crdW2q1top2Qm4jeViqjoFrVHFGPcRQo5lNuioIpJbeVeONUZ7fvEcHp41n+BK5PNXjww8s+8ZIPHierxMLfMr5nOScKNyL+p5y9yd/mq+bIkxTeVPUmxBxNkIb4XB8zYuhkO+Q39s//2Q==\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Recommendation 2:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 552
        },
        "id": "sB9_68_TYU0Q",
        "outputId": "de4939e3-da83-48df-8b68-d4339d0af56d"
      },
      "source": [
        "TEST_PRODUCT = '01118' \n",
        "\n",
        "matches = product_embeddings.most_similar_cosmul(positive=[TEST_PRODUCT])\n",
        "# display top N\n",
        "print(\"For this:\")\n",
        "display(Image(products[TEST_PRODUCT]['image'], width=150, unconfined=True))\n",
        "for i,m in enumerate(matches[:2]):\n",
        "    print(\"\\nRecommendation {}:\".format(i+1))\n",
        "    display(Image(products[m[0]]['image'], width=150, unconfined=True))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "For this:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Recommendation 1:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Recommendation 2:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 552
        },
        "id": "xnHKYorjYr52",
        "outputId": "c05932ee-1626-4b36-ada6-f446da06331e"
      },
      "source": [
        "TEST_PRODUCT = '01120' \n",
        "\n",
        "matches = product_embeddings.most_similar_cosmul(positive=[TEST_PRODUCT])\n",
        "# display top N\n",
        "print(\"For this:\")\n",
        "display(Image(products[TEST_PRODUCT]['image'], width=150, unconfined=True))\n",
        "for i,m in enumerate(matches[:2]):\n",
        "    print(\"\\nRecommendation {}:\".format(i+1))\n",
        "    display(Image(products[m[0]]['image'], width=150, unconfined=True))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "For this:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "/9j/4AAQSkZJRgABAQAAAQABAAD/4gJASUNDX1BST0ZJTEUAAQEAAAIwQURCRQIQAABtbnRyUkdCIFhZWiAH0AAIAAsAEwAzADthY3NwQVBQTAAAAABub25lAAAAAAAAAAAAAAAAAAAAAAAA9tYAAQAAAADTLUFEQkUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAApjcHJ0AAAA/AAAADJkZXNjAAABMAAAAGt3dHB0AAABnAAAABRia3B0AAABsAAAABRyVFJDAAABxAAAAA5nVFJDAAAB1AAAAA5iVFJDAAAB5AAAAA5yWFlaAAAB9AAAABRnWFlaAAACCAAAABRiWFlaAAACHAAAABR0ZXh0AAAAAENvcHlyaWdodCAyMDAwIEFkb2JlIFN5c3RlbXMgSW5jb3Jwb3JhdGVkAAAAZGVzYwAAAAAAAAARQWRvYmUgUkdCICgxOTk4KQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAWFlaIAAAAAAAAPNRAAEAAAABFsxYWVogAAAAAAAAAAAAAAAAAAAAAGN1cnYAAAAAAAAAAQIzAABjdXJ2AAAAAAAAAAECMwAAY3VydgAAAAAAAAABAjMAAFhZWiAAAAAAAACcGAAAT6UAAAT8WFlaIAAAAAAAADSNAACgLAAAD5VYWVogAAAAAAAAJjEAABAvAAC+nP/bAIQABgYGBgYGBgcHBgkKCQoJDQwLCwwNFA4PDg8OFB8TFhMTFhMfGyEbGRshGzEmIiImMTgvLS84RD09RFZRVnBwlgEGBgYGBgYGBwcGCQoJCgkNDAsLDA0UDg8ODw4UHxMWExMWEx8bIRsZGyEbMSYiIiYxOC8tLzhEPT1EVlFWcHCW/8IAEQgCUAJQAwEiAAIRAQMRAf/EADYAAQEAAgMBAQEAAAAAAAAAAAABAgQDBQYHCAkBAQADAQEBAQAAAAAAAAAAAAABAwQFAgYH/9oADAMBAAIQAxAAAAD76UlUUBQolAoUAChQKJQAAt4fnB73f/K/bn6Rv437FP62v5t/RqOVRFEWmNoiiKIoiiKJLAACKIokoijFlDFlicKhUKlKoAAoAKUKAAACgeaOu/E/bdJEy4ZE5MRydh1fJL9h/T/wF+yUe0AAURRAFEURYJRFEURRFGKjFYASUcBQBZkCkWBQUSsiVQAAAAC/mX9IfkZPyzn2uwpux7Cdvm3Trvbon5d1vv8Apbs/mO86eacf9Cuf439knyAAKRRFEUQACURRFEUSUYqIoxmUOApJkFCgKItJVAKUgCiVSKIofk39ZfmSJ+Q8+GefXs9zo+izbe70e36xHl9Xe1/fjz2p6fpNOL9KfZvP+huz4qIoAAAAAAFIsAEokoiiLCTKGvQAVQAUWUAKKAUlBQLAB8K+7fE/Pr83Tf16b970Hku+o2bPT7eijm2On7L3V23Q99wTH7M2vm/0nVjkyk+cVEUS0RRFEURRKCURRJYRRFEWEWGvQAWUFJkBQUCkqkoCgAApfxx+xvzl4s+L7nJz5OhvcnTehru5+v7Ts/Fnkdz3PVeq/La/Y7M+X7B/On6I14gvyQCgAAAAAAABJlCKJMoSURRrAqiMhKoAoALKUBQAAKAPlv1Lqon8cbmnp4eq9T4TaR6fdw7Oq+9d1ujPntc32m2j3Hp7NvNiyYlAAoiiUAAAEsAAIoijFRJlDVsyAKUlBQFIoVSUABSKJQAWD8p/P/05+as+vT7vru/q07nWdppeLevw5NdX6X9Hfmz9T6cferL80lEKAAAAAAAAAAARYARRqZKSUKpKCWFqgCgUAACkAABpfjX9n/heu3ly5WfZ2Oxr8dV+Ohpd9bn3P2F+S/1XbT3Mq/LjM4YqIoloiwAiiKCiSiKIogAEogNagAsoAUKoAKAACgCAABwfhL9w/iCu3X4Nnirt1uTk7fz63e71tzPr5PqHzVpo/XT4N9U6/L9LM8cNkWEURRAAAAAAAARRJRAJRrAlBlKJkJQFAKAAUlACLkYPKfHNPj9EfIPzhrdKr2Xl7xcvXjM+Tna8u20u1pu2Oz4Oy9e+Hew4PteDx8MvQzeh+kfGLl9/qzsvyN9T5N32Vhyci+KIsIolAAABKIoiwAijFYawKUmUoAUAAKpFFAlgPlV3n2v5/wDOcfezdC71t8ee4fV4THVaPptuizxmz7nPi7/B+m7Ph0eM+XXy7WDDh3ZZ51bs5GreXkNOcnIe6+4fmD0OCz9EtXa+Y2JUIogAAAAAAAIsIsNWqLKAAUAFAspQADrJeX+Db/H9Zi0sue6vGrdoa/PjDjmxxnBsWSuziLsceEKy4gckpMhNLsuA5dqZw9/9d/O31zi3+sHC0gAARRFEAAAAlEBq1QABQWUAFBQAB8k+t/nfo1eO3+n7X6bJyTl4TivPiY8/FTkmfFLl4tjA5uG88Mte9B4n0GPD1Z23L1HZS2c+LH3GOOGsc25o7h23ovI9/wCX3vLoe++N3hXIAAAAACUQACUatAUAAqUAqUoAAMvyp+jvyv28/F23n+17mftMuLmOOzOTNS4TmFw5THlxzhxfPfo6n15XvOfn9RHHlbF1OfA0OPPg8tvn6zel2Pdec7Q+t+7+WfVPmNcHPtLAAAAABKIogANYCgAAoCgoUAQHyz4V9C+YfRZcdjV4N1fp9/ou2v8AOxnjySw5cKYssTkmxBuddswxz10uTjuobt1ITW59WHVupzo9dr2HWbvuN/Yw5LI+hfavgX3/AOf04LORelEUQAAAAAEURYawKACgApQCgQFmB+T+m4eT6zF13W9x549n2PUdjp8dlz6fN6c3Djyyw2OPYMcuLlIC8WOBz62WRyZZcpoYdlreXyr0nVd/R7y5pz2ed7Pi5rI7f9D/AJh+482334+a1wBRJlCKEoiwAAAA1gALKAAWyiglgMS+b9H850efgOl23T/T450nddNE932XU9n6jc2NbO/zscvHyS5uXj5Bx8uJnx8tNfW5+OU5+Lkhy7eFNnU2eOHg9rtPPZ/XY9jpblkbLHO2L6rw+NXr9c8vSd38ftDxNSgEUECURYAAAaxCgWUAWUoKCAksHw/7h1+rx+VOk/Q3xfuZ+o876Do9Pjsux6HtInt+fX3tPjl2sLLm488DNjkcfJwYjPiheZmbPDtaB3OptWHlfFfRvlWSz2e55rVwaPQ+e58ePs673HmfRc+/9mYae5rwAAAAJRFEUSZQgANaAoKACylKAJYRRioflr9S/lnz78V1m5xbfU7Lr9zu4O53dTb309hlglyTHA5l1zf1OfUM+DY687HDLjO56ft9A3u06LvIdJ8I+zfHudd6D0nkfVfD/Q72Nvi7U7DX5/df6c9N4D3/AEOSHrwAAAAAAlhAAaqgQoKBZSgoAIokoflr9Tfk2u3yWvvaePfw9r0vf/X8Xs+fT2Oxl3eTX3JTLgzM+PPAuPJxjDlhlnwZncamzrnN2/T95Dwny/7N8nx2anr/ABXqvje33efDnh3cnNx8nqv7j9P+S/Wt/LCyoAAAAAACTKEBrJRKAKCqCUoAAEsOT8VfrT8c06N/Q3NHHu0vV+P9x9hw9bZ093s5dna1uWXPOPAuOxgZRxnNMYZ8vBTuePnxJ3fS95Dz/wA1+p+TrnzvW/a/jvyHV7Pn6rd5HT7Xk4d4+ifdPzz+htvNC6gCyiAAAAAAxWGrQAFAMkFSllEoAJcD4v8AFtxh6fTaW513mcfZeT9X9dxdbe0N7s5dnl4eeTj5BeXh2BxcvCFEjM7jPh3Dh7vrN6Gn0Xe+fh9N8t9P7/5Hf+Kux6TtuT1PQdl127Vd6L9L/mz9JbeaF1AFgAFgAAAIJRqgAWUAAVRZQAB8y+m/mrxZ8+7TU7Pndfz/AE/c9FZTn6Xz/e/YcPn3NTb6+bn5eDnlM+KmGxr7cuLPjzhlZymtal22zqb8OTm484aPX7GsfffX+f8AQfHbvw7hjnzel6Pf4drPr3f1D+S/1du5eYuoAAFIAAABLBKNSwUAFABbBQLiMoGf4n/W34+o1d52On2ODqeW816jyl+TuOy4ub7jhb2zr82+nncewXHHCWW3wWEyYy2OLe0YcmXHzHZbOjuHYau5ow0Nbbp+ld6ZfE9D8UavFz4el6vk4M8+ri/WH5i/Uu7mYwvzAAAAAAAJRFhqAWCgoCwWCoALYPmn5r+1fGcu3v8Af09rF0Oi8j6/oNOTuLNn9B+e5eQtjlmQmOeEt7huMCckrkzMcObI39zq+2h2Whu6MOv7bp97zP6Z1NzzPxO/8Y9303fYOt3OfDlVd7b9CfEft2/lWF1AAAAAAAAEBqJQCgoAFlEsABmfmDw+/rYul3WU1suvpppe86vN8n2PW9p9vxcs19xw7HDumvnr7Zza2xrGXNxJZ5ZcZlWZl33mfSw7PrOx62HXc+vzy/UniPY+K+G6P5L77pO95/X3xX7+t/Y/lf1To8kLKiwAAAAAAAiw00FspQVKAAAEFs6c/G3cdR3/AD+tu9N2HR1W9n9B9Z81+i4nk9vHm+owcuLD1HKvDLPY48zPg5cJTnxhneSGPNwcZseg876CHaaW3rw6jY4eQ/SvnO94/iuh+Ku/8/6Dmdjdjj8e/wBD+86fuOlxlj15AAApCwAAAEsNGylsoBQUCWAADxXtfmEevzr3vT9zzutqXR+32U/ZPyf+tPxj9Lxubm4uXv5tfm1e29NfgZmznjnDgmGHps83HsHJhOSGDX5DPv8AzXoDteKyHUZcHIfpHl09T4zf+PfR9T3nJ7WXPr+0R+j8DpcewAAAABSAQALA0aCwZAqUAAAEL8R+3fm/xZ4bmulz+rrfs38//orbzc/xd+t/yd9Dg2eHm4O5n22x1Jy7PDtQyxz1vTj5uDOWzyTlhx8mGBlhLLDuuh7KHo8tbOHVscj9EdN3vi/jeh8B7LV2eN2+P7B8f/RV+b3Q280AAAAAABKIBLDTABkgoAAAAM/x9+tPxTTo7vrt/Zyb/wBHevuPQ5Pgvz59N+UfTY+Xe4OXpVcGrxdp4nPLm4/UZaG1re15eLYOXPHlhhLxmFwylebHM7vPGeXVc3DyS/UPyb65+f8A4fp+Jzl5XZ4/1p+YP1nrwYDViAEKAAAAABLBA0wALBQCFuNLLCoPnv5g/Sn5sz6+69988+l06Pv1nBu5n5o8xz8f2OHZ0Obbt83ny1Tnxx5ZaeOd9MOfh2C3BDI4ZYZZ8Btcutsw7Xn67mho7GHKfp78t/f/AMsfD9nuczmdP236M+Cfe9nNE0ZaCAUAAABAgoIsNNBUpFgsFgMsaZJSKPkn56+z/Hs23f7Hq+0y7f1H1ngPSdLk/ljZ8jhvo91s/PV3n6XfmHH7j7Px/GuW3z9W3fku9f5+icngdi2Pd4ed39Hjs8cro88XBu8MuPk4uaG9lhwnLhrcMPcfNOHb+C+g3OfU5sWv6d99+ZfTd3LSy2gACoAFgqACAAA0gLBZYAIFSlyxpkZn5s+d9r1uPo83N1Mrv+tZbXzjXzvk6LaagxwywRYJpUJQKnHd1svUd12vkbr8e42fn90efpmv85w9x9F6vyPpcGnuux4e14nX4MsebzP6H9z81+ldHkQnuuoLZQCAAAgAAANIpAWBEhkBZRljkZam1kfiD0O7pYur9C+T/qf8zWU/WPzh9S+I3ZrY91JRjjnijGhQLBcpU2wZWVFSzKENvvOp7HLs9rta3Ll3cHJhZfc/qfzT6X0ORJZ7rAoKgEKgqAAAADSAAlhJYW45CwW40zY5Hifzp+w/ynTq7fg87p0bb8w+pfK9GDIltFkIiiKMbjmY2DLLDMoTkmSALLE99yzlydD0PH1uPizuZ1mx68fqP3vHya+chMALKEAAAAAAAGkAQsCY5QtxpUFBcsaZee78fmny/wCv/nOfb8h+U/X/AJB5t6RWrnRSJGJlOOGWXGOTHLAzz48zJBllh251de5T4vb9t83Pf9L5n1GbdzbWnt1++T6Ry/pm7PiS/KAAAAAAAAAAQf/EADYQAAIBBAIBAwMCBAYCAgMAAAECAwAEBREGEhMHISIUMVAQFSAjMkEkMDNCUWBAYRY0NnGB/9oACAEBAAEJAP8AzwN1jcri8zBLc4zIfoVYfcf9EnuLe0t57q6nuvUv0+zOAzkUXKuF+q9lw3iGIw0HHuO+uktnDPFyTFZn1b5vkL+eaxzvD+d8i/eHvMz6jz+vN8mfuXtcPY3NplbOC/xtx/0LnPMeOcUxd5BlRNIkkks3g79epClmb31v/wB9+39IQK5rH5XM4pXjxeY9KeZjlWASzvch/wBA5PyKTAWsMdljM7Llpc1lHzdyNdt15yfsvafZY0dt9xHshlFAESUV6kkHB5rKYDJw5HFXXBecWPOcZLOkf55QCSWfnPrEcxZX2J45Dr2WNAkDlxUVhCANrHZRnQ6xYeCSMqVveLnsskE13iri2YMtGMp/t9zvVcX5ZmOIZWLK417S7tcjZWWQs5Pzt9Zw5GxvrCdvVnjFtxvP262xtYBovSIAdgRj7e9tGNjRtYeq1dLpD7zHbEGr+2R49gtDJH7lgoDiQ16OxZbD4/IYTJWP571oguU5yTPfxAKgFLSaqz69l3VsF6bq+K6I1L9zUqhl96ni2xYK8aaavR/ieSwGNXLtkvz3rXZ65jY3AB0v3pHi9tyW0QkI6m2tgjL7RyRIh7yXkuOdHYZGfxHXWcISSpEtuAB73URVHc1w9JI+HcQST8/6z2Za84xd6uYpZpWVQLdoPvVpIFZT3hvCWXRvbgtrbzTmRh0pMY7qHNW9tdQSDcsigisht4XRVsLMY7HYywH5/wBTuScZyPXBwZG8uBECVoiW3uRDkYJ7S6sTAZosRGbuWNBWQtpXuDFGLbFZW8F09nBY3dxedILeGyvfKNx3BfsN1jER8zjXnh4l6kwcqzt7hriz/Or7MprlaGH1B5whieASkdhLLNfSWxvTmsrd5Zbf66XjqFLyJxVxG8Vy8gow5SOcz2TRwzFDEcMbUooV4itCaKCSK1nmwksuD5vxa9sKkUK7qPz3q9iBY8mjy6LGP/QU+1TKdhRWHtpy4KLcM6yMHGNS2nBQu9hOutTZKJRr4EKSBSW0dzZX3li4Dg1yXKLJypOySfz3q9YfU8cxt6CgCEg1F4/uxe9txNMzVheRDFyxXducnzCO5u7iSVLWKa5skyIRc3MsfV6lv/KxLGF+8wNQwT9XSKDiHG04thIrM/n85grTkuJusRdmcOpYOJJG69QWjYn7W+MEjo0pSyxZhtZ7e0N0BHpauZPIx6h9kKQcam5ATXpNx7GDBpmJ7T8+pKkMK9RsT+zcwygRJNqr6EVxO7BY7a0xuekO4kPHOUKglW+B5CrvFIXR4SPJJ33WLXSSOa9P4fDw3D+3/QPVbB/uOEt8qiTRlT70G6e4qy5TPYqEq45pdXcIt1RLwsOzVdN3YsTH7sKjnESfTrXFofBxbjkf/QcoAcPmwREI7y2V4iysCRUdmZm0Xgwo69hP9KIhomcAULhjJ44BbQ+GFhvDqFwmCA/Gj/xL0dsfk1qGeS2aGRDNNHMRMgiuQjDdLlAo0KmyC6J7S3Ut2/jirHWQiUMR02pFcbnW641x2YH+DX5y8OrDJGmHwSvkN6Lyyj+xmm/sI45ZyOxsLFEUHSKB9hquBc5ssVGmCzU40yqyt+fz062vHeQ3LtJ7aH6NSp2NWkQBBIjf+wpNGlUaq2sEvbuNpVx/NMphfHbwLgeW4XkGoYbhlZTph+bvbm0xts13kb3kvrVgrCKWHjcGb5dyLkU9kMvlJDtj+mtmok/9Qr7CokJ1sxR/+nY+6JSBrOJVZHkijQBWKxdg5GI5byLCBBY5ew9WLF2CZHDYrM4nOQmbF3/5ZVZzpRnudcU42XivsrnfWjOXXeLAY3I3N9lblrrJ3yoQU9rklXgkonda3SqKjUVAhbWhDF7gmol8gASgYIvCEQlk+PdtnfsR2VAKgK9HVh0IneQVC5tXhmhbjvqPewCGDkNQTQXcEVzbTfk8/wAqwPGEByd5yf1D5FyMSW8Li3VBpVNuw38TAR9wU+4AjtRNCeyvirtfeBYrSRjoqbGRahiJk6KbGynYqCnjtkG5KM7l1cs5XWlRV9wQotWlfrXTsNlz27fYksFNMexJ0qzSMI0j41d3WFmVo7qwv7XJW4ntn/IAEkADmHqGcfJNjMC8weaaaeeRIdkEoYCN7AtV7MGIs4yrDQx4LEARwGM27GovE+gaMtrKHeU6sC0ClEniEUQit1Msjwg0FkR20OkpBOj39jpPk6hUcxhnUIgjZlK0fIIEZVZShcCirHS9fILf2akleWTxq+Cyt/hrwXVubG+tslaQ3ls34/nfJDhsc9jZSiAK3XxqrLruqxFkhj7FY2ILSKG1vp4+wZ9qwROhLeJP7MijQaKT5MZAssCrIEeoolWOZ/GE/macgydOvkBby/JD28UXwjYK3Z4+xMUSgqrBF0ywOJY0ZvEegcMIgjOQWTvLJJJNFG5VIAkHXvHJJHxbLftV2YZCfx2ay0OCxV5kphPdzX9+bq7uNFwqufGpWMNUnRPOyghYzMqjq7GNGrQWOJt6T/cZJOzSHq3cnQjVpNFAgjm6H4rFIwhUlI1WG4KCNJH8QLEOFlAaRXZrglwXVGUKsk47sSSxWBVoqUbfn0/XQkZJPG5DBVkt43ZbMKoVgySMLiXYgl8bKCOKZgXtsLKZ/wAb6rZAquExSmEd3jUlVaLq7ugCSWkelZHcKKLMynSOrvs9zG2pu1AxrJAWeNXJgKsncLKzRRwiKeHVdAsXeuupmbfV0htERZJGVFK0iBTadhEAykOepFxGu1ErKpNRQ7cLpg5hmk7IoimuB3kLC1iouDLHKtRGK3kkVnh80sEysI5Fdo5WGMu3tJLe9hdHSVI5Yz+M9Sbk3PLb2PtCT5o9OEVUnWuyh7iWupAIC9XEy9TGjM8O2PzLANG4CztCzPoxeNmR+mwkocvd9ZdtKYFUTeU+ciRiBKx8oRWgkSGd8xDf5+THRXE81raJ1nvJ+S8cie83mLXNWksiGGyt5p3iH+C8PQQqQxl6vqQsYrl+8jv2RokUyAeK5JG4yZAYW8cgJpCYie54zc/U4a3G/wAWi9mUVyW9TI5/NXKUrdZQCQ2zogyE/UgIw0S1DsvlIBCIT2BEcRUGZpF6TlWl8Za5UoAVWJ2Ld2juNUN/VOYwOxS2BaJxu4dncy7jLStwOUMd5Oy4Rx6B0eQW1laWMEX0tu7SyJPJKrxkRKkTRmTcpM4DqLcyF3SfxIY5rwTTwRrXQ9po1qHtcxKUk7lJoWc2M3ltTCq8IuSwyVq34zk2R/aONZ7IBpiDLMu52ZUEiBpB2T3jIIBpQxKbohfAEKsyOZyU7lrjpUZ8hgMb7YlmKBJNWyoSX8MarS+xuGBADPGEngQSW5iJcfOYmlHheClBSOcbdH7xx9+zCMqx8Lp4NM5c210VEjblkXtLOsEcz+K0H08LwO5dDHC0tWx1cgB7liwbqcdL1uwyrwycRZwRH8Z6wXxt+MWFijyydbu3DGTZRwHt5TNDE5aCTsVU1F2KShmjRNxh6URIsJ7CVe7t2/lsyLsiVYS5QHv3LMUn1EpoIfEzNTlpY76QFj0T+oA+OQ9otLO5UTkHxyOdP1kKOxKm0U0kpe4MvSSWRYIAU+qIeTZWQzuiGpLs+QAESnblmSURRwMTExM0YlS16wt27cdkWDkWHmDfi/WW8M2dxVkrXocWdrd7upogYpYxZOElNqKhkZZFApCQui8bpGVkKJ4U8YZ9MIEctK25n+BEg6Dx7mEVwSHB73BFIvtAqOqQskTOevS4eYlE6G3aNV1EFm7O7u9rp5GkKSdnaWTzFoqDTmKb3dkDS6jndBAwFY+6cfXEPABIzClBbqoaft0SJagYLamRzF/LZCtYmdE/bp6f+pvxSjswFc9u2yPK87cBoVEtkEKoC9rNbSVbXBEkLVGQkxABf7VGU6r5KUuSVjJ/leNKll7M5K9laRdMWdmm6xyEu8u4mDL0BnjDu0EUlSSrNDcNt9O5Du/YRRhbjZaZWUovaK2Sra3USSPpUKw2nkUkySBCPErSyIoxUj/Sz7FkvSHy1aDbOVbqks8Yct2V3t+zoQoCT4stEY1esPcC7xOOn3+JknS1hnuXq98siPMTZfO3K1d7Fz5I1u1aGY6qCXyCGar0kEKWhUtC56lI4uxWLyII4kDO2luHLIHEwXa9SyOQZOyqunWRVu+zSFjLGJA5Yr/KI84VvcvL3+TGV+8rCrcOZ0DLC3W2aiC3hWIqY9wlg0RUJItSReO8v7YCNUCRppWZBFqrSPs8y0zQuIdx6kR3YxWbFdsr8Fyq3EN1YOfxPMbn6Ximdfd7HqGrNtrKBVyXXvJV2qSRbrDHvBZVeOzpAwFgGaBtUCg6FQpJKAl9P3K0sm3R2cP06MwQyHqnWZU3MohZJna5oqyCJPC47pLIy/M+/UmUIxaNPIsgCp/Tanyq8yMZRsdJGnOjAYwfG2TszBmrtytsGkVgASD5GWkVlhjYyHsfNAlK08oDPDFNbyulwh49fvjr2BkMM8dzDFPEfxHqhciPAWVopvPcOAbMde+6nDFSOu1HY6w3VFaMtdb07FLTfgh9lGkjXUCgsCAFDB2YGU/1EFyFuDGB1WdPHEHd2jjDR/Uqzyd9J57XxjcjOfi7MHbzRL0SBghk3DI602kgiAMMzTF/eNdJKEGmSL78kVPBHc7sZjHNN8oAm4dxI+iwJQzf1gPapLMfGs1xHaq8VzdQckkM9vDjMbx2yyNhiLWHJj8R6kXyXvIbGwR7iMG3Zmkt9pKQwc+5Krc/GQe+BYPNIpq+Ka0QA6WtuveJH+VRj2+yIVRCCHCohodPASojYPMQqRN1+mklqR28Q6lGWSWFY6BiK2u4TGjFVSYSxFULkOhMzpJE4KopZNG5kDK8JMdqy04ZJXDHkUffD37JVnOGKlqsnMjxxeRpRbIWmluuT2qO72FpPlMvck9ryCBFfesFcmwyGKvVqX/Uf8Tf4jEZQayGLynpXYTky4fL5XBXuAz15ib2ecOocE3ILd9rgewvbhDVw3kmWNXkLuz9QVMSSolQogaEqmkMcYqIfO10UIAtpVoOUFyWWeWKCXTo7MAiSAN5GY+IyaTVNJbowKCH3WxRGGyGclejwhFZECOjCkLuiANKoeaQVkFEtlfRCrCWGFXWSQ5SZw6WkX0s8z95WFm5FGzYVHbaYVbRjwOKxtx9Xi8Vdb/Fep8rjn2YlQi8WUjyLcnYDbwgU3d32azTy5W2UUu2LFozt5FBplRmYrUki9G2y+JXkaJioiEQSNZWDMhR3RkuXCv5NKAHljja82wigSZI9ICwLKYgVlti5hKDQ7CJdjssw0VANsxHcOwRTOoepx1F18XJeVpqsGDAVAFoKv8Aw0akH28QDVbAaI1wuf6jiWDbf4r1PGubZ41GKmUD+2IUifIdTjgDeXMhEB2d7i27uwodzoFuqdFUI5QNIqOv0/YAlXjLpomQkEFIfp2V4o6LwpHcxrXz8rs4MUetb03lbTBQEUMChdAWM3ZmRkktzKOugWSG4A1yi5NlYZJkVkCxhRWMf2WoT/TQNEg04qA16aT+Tjc8R/F+qH/5rm1qNfargVjGZJr5lrEdxDfvrYTQFQEgex2rBK8aBlDL0Ue4Df1ktRkaIRHyCNBHCwBiMhVRAdNE8XSUO08xVXDGGRdOQ7XjqURPJGqKsXjScGW6R45CCbYEugKuiiSFnXmMLTlU3Kuiwqwbq2qt22ordKaIq39mFelz/wCDz8X4v1Em8/OeSEVGvtV37VYtp76sUp/b09pG2wG0Oh70LgAN7d4wxRUDglyIyzjeo/5j+QMnkuNJ/JD3DyL1jYk2y7CoCbhFaZOiKzuoi1cGMrEC43UIIhcKX8jKx7WrFXY6ZdeB9Z2FZ3eMVfQ9HNQ/y5jVpJtRSnailpaQ9WHt6XTDz8gh/FRDtLGDWWvv3TN5m/2g9qvKtDp7w1ajpjMclMfc7pFUqu1jZAB2qNxIyoji40sWnS66DrpniHbUHfuFTTyoeu4xto4mDKHYy9JBoiJ+x0FnXZRRcPowq5VdCVZF93jt26um6kUukDFcrAXugBWWsyGk9uOYP95x3N3WGzl2ikGGTa0r+9Id0y60a9M7kpyEx/iuU5QYXjGfyW4Iykaiu2gKuzsVbE97sUymOHxqHPaVutKNjQpNRqfb4gz/ADQo5jCssh1sM4YGdfIWMYifv7O0n8vcjJFpvZkLAhCTBQGg6ii06lwxgllkijVQzM7tGojOhI4pkHRNVk9xtDIz5a07OshX0JsY3yHM3nTJYqbjuay2FmqCT2pG2RUH2rr2WvT6Tw8rw2/xXrJmALfD8eibxajpn1Vw2xVkvedgKB27DqjF5fYICso24jBQsoJ2pUhe6smikkX3JZ4NTeOeNyJLYrUZjYykmN+vjegIip7KsYcWybCMQiij2HdRUUjlQXgl66UtXxUS0CzLOFOSDJHaPVypuYhs+jdn9LZ8sm163YAwZHFclhSF9MBVu221UApNddVxV2TlHHtU3sSPxEs0FtDPc3MmXzE/JM9kcvOCnw1VyhX5CpztaxQD3iJTvoRhhCfnIdRBmbbMqR7lIYwKvZ2RGK6O1/8A3KzS/MlChMW6Z4h5AoUKhiZFiYxQxBUbyGRGr5+GLpXzPsUtn6KF2Sz9BUZPiiQC3btHOorKFhBbFm8S7JFenMCQ4C8kROT4KPk/Hcvh3qNnBXutptnBqL296i2WG64ugPKuMUfxHq1l/wBv4slgjWie9MNJVyNAVcqKw+xkVNO+/JUJKxlkXqfE7djENsRSqGfuUEUA8RKRpbdFJrSqsih0SP4kRSJBuQKCx766JMsMYUqfKZj1kLtFFbv3D7jBWKyYSsiAsItoZGWP2iIWxdvEwrIkyTw7UQ7aRWi4ND4uKYs7i/1I65SkI5VyhberIddVBHvVRxhWrF3Qscxgb2pAFdwPxHq1kxfctFijWiU6fDdXlTPsGsL/APYuXp/967QEhV0xPzVSqezEtCD3BpU+3sUCiP3CAbBjVQoXqNJIr/yZSC1wdrMhTxB9bmmYH2C2oejIpRO5hfv4kR1CL2ZI5EdolOrQ9E0AzC5+tkFOHafyb4tCYOM4CIrF/qR1lkK5rPbq2/tVovxBob7mphuB/eGX6i3tZ/w6MiHvIb2+ky2TyGRkNov2p1+NX33NXDa2KwekiuZCC+3RdROWf7PtpWCqpZ0JMgHx0SIyyxaR4x0TRSJiHLSFDGUO1Z2R+7mQSR3REryyhdJG8ihpSYgyK6bRiClmpqzZmldiAkSx/d0KWtzTu1na3OjFEDEqaeAKjOEx8X0+OxsFRf6kdcpRI+S8hVKtgParX2joH3Jq5Yi3fVW8RgtbOEj8Ny29/buJ8lutwDqFUVZr/TU3shq9X71ce8mqx6/T2IG49GU+8RZU7gOZNEVF5AD7IznyESqXfQ7Et49dlX+ySuJNaV99UUqHJKzFhqTudmQsGkVVWQAydjEoEkGntS3kmZgVlbTGu0oiUPVyB/LWuysoDVHAZ57a2jV9dmAEQ3KlZO6+tyuQuKth8hUTaUVGax1qb7I46zCyHcj/AIf1bvfpuIpahovvVl9hU3uAKvV+DminecCiECRRqye/sKLgKqCg7Ep8WLAAUDM7D3AbQ7Jt28aeT5lkHV/LG+hTs7ABYyGH+1yiylVJI6TvSdJBMRTxpsfG3Vu92C8bSmOfsobsoq6IGjRWORZNNgI43z2HDFv6mq/nNrjsndAxe7tVqPcUCQoqE6NcEi8/LMTv8R60Xe5+O2AMA3qrUaUU52TWQOoiKsIfLd04B2TVsFEkWwCJjbg0TssAsjMDcDyAPGwbrpysQEUcczbPRS/ZdIX28pKq5kMKowdtsAiyKJEc0rIUl0WKyJeOFMSNOg6xSAveN3IYicllHWNd1etoAbUKdM4xbyRXllKC/wDW9cynNtxDk8oMA9xVqu9UT8ajPyr0xhEmfvZiPw/qrd/U80yEW7dfcVB7LTtpWNX8m9jeLi6Wl9dGiSWI7WwGwdowA9grdlmOmmcs/WMGYskZKOWe3UVH28XYVExLozAOekjMsc2pk7kH+Tbdh3ETkKSY2C7LxRD6hVfcol2atfNGJlQ70Lj2BAC+94dlFp1jk6K7Q9kgkcSyaLk16knp6f8AKHqIe4q1H2o0h0339KU3dcil/Dxr2kRa5NefuHJuQXW7dfcUnsgq5fWlq5kLlqv8d+z8EwYljiAbVAohhXSDqZTttqko6xF2YlqhXaq7DbnroRtqL5qD1NuKG2i+KkRI4WSMSIFjlFCOaRl7KiIUhcJIrOtw3UgrPJ0KPOPJsJrtcKKDfEe9391NSb7MACrpYSkGJu8MD16n+3p1yqoR8hVt/ajSa3XpPFrH5+f8OJBAJJmpZDM5kNW6/wBIokKKupdhjXGMJNyLO2GPjHrLNAmb45jIFtlUdSQ4RmYaO+x2HkPmEbyxf0xaDShY9+RCC3cDbEhdB4zMgWt/4dWp33Oe8ah5EgRakIdgHkDRtFAgdwwhkUDyr9Q7bSUKt6RJGSTIut/fdXS7AFMn8re5CwtQCbBvJj8dJXP4PquC8ui1b+6xtVvQoHRNemkBh4lFJr8NyS6+i4zyS53Av9Iq3Hvup30uqun99V6R4D6DFXWbnj9TZTec/wAsgKdFjclQiNIzESOiLISE00rOzow0WVS42uigVXbdKykuxYt/iGHQaZI9sxHll9inaOACFVRdyLKI9/Thol7rGERA7JraK5aOd9WvVWlUFl02itwOyE6fooUM0zP0GmxB3hsKazyCbjvJISLE9re3aret+2qc9Uc1x2z/AG7juBtPw/qROIODZ2rcbbdQDSbqd/djXH8NPyTO2GMhqGGCFLe2t4+Uypc8s5PcgkLpFpQNJ7XDISo62y+49iwJZQUdNrtleNVKK6EEA6HYyyNRSN4yWkDl/JJ10VMKxt7tFIHZOsctu0hUP4YgKAi8gWu48M5NWZBeb5NoSMacfy1FP2+TKZXfTVgn8mBwb1l2CYXOOaxin6K0NQDQo1YWjZHI4yxUSaDsFH4b1eufFxWyt6tFLECmPSPVXMml1Xo9g/BY5HPzJD7Soa7ia5nkZo407Rlh8EANJ4XmiUlEReqijMO7aqMgUXRietM8YhKF3G3dpBuAsAWkQPJcnv0LyD3Up9NIqqXIkR4wAqKDMYmkLW3YLKhR0Elm22kpm+ZID+yE6Jb5BY2kIZR14wWPHMF3rm101lwrl06VYR6tbcVEugKavTiz+r5hYSFfw/rNdfLjlkDYR6XsauX1sVFbT5C9t7O3WxsIMTj7HGW9I6p2djZv2t0k2ihfepm0R720XjVtm6ZYxpqWJHRCQiID9h7sx3P7ClPVHDFGWPTKG1/iQ4IAkUyiRpSjRo0riOckqgZIZCSsod4VLd2G/jZOCzbA92apDuAHUhCJ30zOVJ6cS2eMYLZ9UJfD6d8pbdmmoIxX9I1X/wDfSO1+fIL0/iPVqcTctaLdtpIlq7k39q9I8L9dn7nLypXIpza8c5FODbQlYUGi4TWgkXk0pLAac6clmY1DHGU7dVWMKWQRpGV2S6M3QGvGAGGhEWK9KIAF11qQgSy96l8oT+sGctIFVpF6IrAeKRwoYqFHslo+jStog1Iw+n+0jnxSupJPYL14uhTjWBBr1llMXBJUqzX+UtPQFem1n9JxG2l1+GjXvIi1yrJHL8jy99uNtRrVw4DkmuA4f9k4jioHWvU2/wDo+JTwBgU6j22X+O7eNAqndyyjroomwKijAKsCQPvT6CgCumiN0E3rVIiCQMa2pSYq4jbvMsZkMioitTq3ebqzOmiOx05HYMgCroxDX9o22PYyHULCkKsSjK5/kjYxKhMPh0r1wuCuE4xZbtxqIUfepD1RzrD2Yx2Gw9lr8Nf3iY3HZK/cqzOQxpP9Na49iv33kmHxZDsGckCvWK/V77A40PsdVAFtEq7NTyBNEAfNhUUBVSrBlUn3GgGXVTDbUqkMpSkjb2IqECOVC8iklEPeEEPcbMy+MRKiuNm6Yrs76sF1HL2Ridqo2mgW1UHyTXaTfjk9vArns6iUOg02NO8ZjK9Zrgz8rwViHi9o6FYqz+vy2Ks9SHcj/h/VO/ey4ZdxIQfcUNdI69IrZZuV5C5b9ANkCudSi+5hnrmd4YvIzSMHkQIvWmV5tHcFiiBuxdeuwaAAApOiv2dZhRAIHsqa6/NJWRVApCQkBZovJ52oK3aEGKQKwuSUkdGFKR8l7IDpdUm/+bWXRX3lAZCal3PIqpHbxyS3GIiJwhMuLxvvzDKLm+Y8hvkKH4AUv3FcFtxPyzCA/iPV5QeHwOaC7NRbMaivSWeG25NkrV/0u7tMfZ3l69XFxfTzSy3AM0hCh7ZRF3D+BJYFHvGZrde6hTeWpYB5lnt5AVWaSJ2V3SpIJnCDxi3m+Rkoq6hSTCGLfGWMoDD0eDu76ZZS0kqpr5qpVYvZv5jkEKRqlYAfdX0ail1rZeTcbbCMqAsWsbmJ8lBCyvmxg+AZHLqbRDHHFGTrqimo/vuvTfR5RAfxHrJdiPA4OxpfvUOvtVtNeY26tMlYvg8za8hxNnlbUc3YJwvlMplXlN4Nea1XlNsdCWxHJcOWDBI+S4giNJLiDkWDDIxvocxhiS31vnxVwsfjuXjh69kT6dSWPU2xHf5GOaIo4nhW6UuxeBZjJErW8ZljjZvEs0PkTyXaxuCjFZE0ddAuhqvtSvsEbQLrYMm2DgGRVCt7YsifIRMTzzNqnHsDxxHQliKErEBdxv8A2r0xDzcn2g/D+schfK4WDcae9IGDA1G/ZdV6RXpWXkeKJ9WLtbP08z2yx9z+mhXtTD9OiH/aks0Q1HOmQyEewt8vIs4uj+4RcszKFOzJyyb381lFy+wfoLi2i5LipURPrYbyynZClyIoXVFEQilAcrOxn6kmCSVEeUMiGBiGMgltQjkXUk9sGJNy95ahNAY+eK1jN1PJd3M+Uvp7+eoYyAK6MDrUf3Ir0px0Vvg8jkNfgx/B6pXfm5TeJUI7VMwj61ayM4Dgel+xzXIqh9esn0s+L4dWP8B/yuq731hyGSt9eC/TkeVHcyUOXvIEE9ieR4icN3SLO4tT8L/9zxxjc/uX7nj18u7853DQ71dLfzXaFd26aSokNEUq6fdem+v/AIhZ/h417yItctvmyXJs3d9rYEbNZJ5LcJc6tZWCkK3o1bFsjy29r1oyX13PL2AN/Af8nVAUP4SqffqY1P8AaBfkVAtR8UFWcZdKSIKK2Wc7pD8h7elk/fA5SH8PkbmWyxeWu4VaTvFZybxqj4uF476czcosLTJZHJ31imMzecxcbeh8yvd8utCeR5I5jP53Jk1v9TR/yh/HYoGlkNWRDO9WChV0akPsa1oigfcV6TK37Nm3P4UfoCP7rzji7cTzL2ca4b5o5rgd7aXvE8M9rLzOJ7fnfMIq9P8AKHET86u+4HVEX+DVH+Af52LQs8xqzPS4kFW/+l2FM6nVEjdD7kV6TNvi1/8AhR+vqLhBnOJZDpFgJlW4KbxfNbrgd5NBFi+UWuM5ZzFcni8hyewHGLK9trScnZ/gJo/wD+/8A/gH8JrDpqKRqVSlxvVrdCIMrV9X3k3uS6A3om6JTso9MbJ7Lhto7/hh+iN0YNrnHHpeIclfwJc26Z+BSs3HlyRysdtDe+oMObNpb3OSx36b/XVarX6fY/wCh+o/hNYFDImjTkNdlVMr6XdICdapwykJSwzXM9vZxVBbR2Nta2UP4nk/GrDleMWwu5uQ8Bz/ABGEZSCaSwx2dga7jazNtND4Vl5DiHxGRkRU1Wq1Wv13W6JFA7H6ih/HBBcXVxBa2tvlbBcddNYi6umsMLYxlGsO0iGZxdaMapSgbFQ6kkUtXpbgbzJ8ksspPjiSSSfxSsVOxWZ9NMHeSzXuIHL8LfcV47Blb/F5fMTZY2/kXYr2r2rY/wCdj/ksP+S4/wCfvR9qU+9bH/OxQIoEVsVsVurfE3E+GymZJri/pzyzlfSa0sMzleOcXgucJwy66qF6gKjSuoqCRURQKJ7zRg1o+Kd1r0t43f3nI7PKXGOeR3O2b/yP/8QAMREAAgECAwYEBgIDAQAAAAAAAQIAAxEEITESMEBRYXEQEzIzICJBQlCBUpEFI2By/9oACAEDAQE/AP8AmBKdMNrFpp/EQ06ZyKyph1zK/wBQj8Ausp94i5RgAdYcxKiWJ/AL6hLU11OcoP8AQSta5uJT8kn5b3lUE5ARlZDZuPpeojnFohs7REG1kNJUQEZxUUaQJeplMZs+YAOPU2YGU6tli1HFxYEGDbtmRb6ACbdtBHqilmQTGYsSx1P4Cm11lPY+926QBD6Xb+4SFlZrj9/gaf1iFDa8BpgZQk1GsNJiBYL+BpfWWioWOZiqALCYrZWmzNoIlVW6HjmdV1MNUnQTDZp1vAIiypUp0E26jW5czMTiWxDclGi+C1HXQxawOTZcUzBRcxqrN0EvLylWekbr+wYuPp2zotfoY/8AkamlNAvU5x3eoxZ2JPM/DTqFDY6cRpHcub72i/2n9cPWNkPU7gAm8AJ0BliPhU3APMcNXPpG4VmXQwsx1J+KiboOGqZsZbLd28aB+Vh14Ym9zBDvaJs9ufCubIx6T6QQ67rUeKqxzUEwXIFxwjqXWwMNN1BuIIdd0uYlPB1amZGyImDpp9t+88vpwyC5I6RsEpF0YjoZVQ02KnUbqmbZyibqIYRKgs7jrwtEXeWssxfvN+t0pmDe9NYYZW91+Fw65kw6TF+8ew3Q1mDawgN4ZiR/s/XC4enspG0MxfunsN3ROyEMRo0xBu47cJSXadREFlj/AFmKN6p6AbtPSvYSjqIRlK4zB4TDDMmDSVspWO1VqHrvMOMhG0mJ1UcJhhlBMSbQm5J3Q1HhSFhDMQb1Ow4TDiyjwxj2Rt4NRKekOkrm9V+EpDKObCYt9q/e28X7ZT0hjnadjzJ4Nc2XvKekxFTZUyucl3i5hewlP0jsJVbZRjyHCUhdxNFmIa725Svqu8p/Z+ooyExRtTI5m3CYcfMTKh2VvCbkkyqbuem8oi7JBMYc0Hc8Jh/T+5iL+USOXgTck7zBLezclHhi/dH/AJ4SgPlEsCLESqgR2T6Q0qfKeQnMzyF5mGhyeeQ3MTyanKFHH2n4dZhUNOkqnXwxDbVVumXCUBYDtCxDWmIN6zbggHUQ0kP2zyE5meQv8jMNQQPta20iiwEMq+6/fhMKQSIKrtWqBjexIAlb3G623lEWVD1PgZVN6j9+Ew1TYYXjpTWqauzYMNZibebkRaw3ii1KnNsxnYxjck8IDaUMUF+WqLrMTXo1UZL3ZSNk7xavmG+zsqoyEuLXlWra6jlvf//EAEARAAEDAQUDCQUGAwkAAAAAAAEAAgMRBAUSITFAQVEQICIwMmFxkcETUmKBsSMzNEJQchQVQyRTYGOCkqHR4f/aAAgBAgEBPwD/AAwU95GiLncUHuGjkyU5A/oJ0Tk5yaUEx36A7QrpHQKRu8qMaUKd7QapnemkOFRt8miMhBTnZappKc4nVF1G5qz1LKnbyKghOZmi1uWaOHcDVYapkZfkCgAAAP0B4oSj3NCcfgCGaiFD8v0F+5ODhoqPKAwDNQmpd+gybuQuoiSdVY45J52RR6uPkN5KtNgns9XUxx+8PUbdFZJ5qFrKD3nZBRXbG2hkcXnhoFebcFpwgANwNwgaURKcVBZp7XJ7OBhcd53DvJV33dHYIyAcUru2/wBB3ck1is0+b4hXiMirRdUjKugdjHuntLQkHaYIJbQ8MjbU7zuCs93xQUJGN/E6DwCwFBitVhgtbQJAQR2XN1CfcE2LoWpmHvaQVDcEDSDPO6TuaMA9VDDFAwRwxtY3gObbbC21NxNoJRoePcUQWktcCCDQg7O1rnua1oqSQAO8qyWVtliDBm45uPE9betlqP4hgzGT/Q7PdbA+1tJ/I0u9PXqHyMZgDjm51GjiU6SNnbka3xICa9ruya94zHMIDgQRUEUIUsZilkjP5XEbNc7M5pPBo+p6iazQ2gN9qzFhrTMjXwUcEMP3cTW+Az5SVXkvNuG2SH3g0/8AFNmu9vs4I+J6XmsfSoh1JyCxEqqCvhtJoX8WEeR/92ZjMBawaNaB5J+RBTTVo6l2hTRnyDQK9Yi+zYxrG6vyOR2WyM9parO34wT4DNHthSBRnoIadSThdRNzPJLJCwYZZGNxZUcQK1UrWNke1jsTQ4gHiNksc7LNOJXsLgAQKbiUy2WaZ7MEmZPZdkVJmFH2Qt3UzENeCSp75ssGUf2ru7Jvmp72tc/9XA3gzJF++uezOJFCDQg1BUV8yt6MzA8e8MirJMy0RtkYCASddcuaOYVbG+0q1SghxCCBTDVjfDZZOyq5q6Pwkf7nfXqp2b1bo8FokHfXzzQ5Ivu27LMcgENVc/4Rn7ndVIKtKvNv27u4BUzQUPY+eyzPq5BXN+F/1u6oioKt4xWicd9PJOCCh7J2SQ0aU85piudtLG08XuPoh1VpOK0TnjI76qTegoTkRsk5yAR1Ueiu9mCx2YfBX/dnzhzAaZlElxJO81UuqCg0J2SY5o6qAFxoNSo2hjWMGjQB5cwc54JY8cWkLcpCgoOx89klOZW9XZHjtUA+MHyzTeYOYeQahUoKJ6CiFI27JIU0Zq54/wC0A8GE+ibynmnkGoUg6cg+Ip+qCYKNaO7Y3dkp+qhZVwV0t6c7uDWjzQ06g8swpNMOD3fVP1KjFXNHfskho0rVyhbRteKuhvRndxLR5IKtECSh1FpP21oPxu+qccyoBV/hskxyATRVyAoAFdseCysO95LvTkOZQ6mc0M37j9eSzDJx2SXtKH7wIAkgDU6KNns2MZ7rQ3yRCAp1V4dGedn+a/68lm+7P7tklOZVSCCFFIaMeDRwIPzCbeVsb/UB8WhC97UPyxH5H/tC+J98MZ8KhNvkfms/k5C+LMaVjlHyB9U29LGdZCPFp9Ey12WTs2iP5mn1Q0qNOYSGgkkADMk7grdKya1TSM7JcackDcMY789klzJQaMKhFI29Qx72GrHuae40TLwtjNJyf3AFNva0jVkZ+RC/nEv9y3zVtvKeeMxmjWnUDeiczyR/ds8NklCLGiNpA11UfYHWSZudyx9huyTMqE1ziwMrmFDXBnxPWHtvWEINagKADZNU+HezVRRyNcHUoCMx1hZhFK1JOZVFHHoT1v8A/8QASxAAAgEDAgQDBAYGBgcHBQAAAQIRAAMhEjEEQVFhEyJxBRBQgSAyQpGhsRQjUoKSwTNAYGJy0TBDRLKz4fAGJEVjc5OiNFNUg6P/2gAIAQEACj8A+AWOLspeay1yy2pQ6iSJ95/sKlmxZttcu3XMKiKJJNfoz3uA4m0A9p7V8F0KgoDvTcXxSXr93iXe8LKTdaViJJq/xrtxLvavcMyJotNtbgxMVf4PhFvv4FmyiWotT5de8tV/heFlbt6zftNxgvgNLIqRCVY4j2F4o8K06eDxgt8zIY0nE8LeQOl20Q4zyaNmHMf2DscVxfE8I/gey3TxfH1YHijYJVqyXdmFq0CEthjOlZJMCtj0qBUVivMK47g1La2Sxfe0paI1EKYJp+I9s8AjHizcw72y8I/9geL4/wBq8cl9PZ/DcNbFw+LbSddzoizQv+1PHZOLu6lfzpgqCuIWpNChneR7s1mgWXcU3D8Va2cZVl5o45o3MULHtHhdI47heSltnt9UPx9UUAlmYwqqMkk9BV/hE/SHtP7SF7z3rC8rYWCgeuZJrPKpPWsUexigOqsK1iaIrNA3APDv2mErfsyC1tq18NxdhL9lv7riYPcbH49cFniuHuWLpttpfRdXS2k8jVhE4uwGscLYtFEscNZAtJJO7tEufpD3Qax293EizeYe0/Z3EqDd4R7F4AFVu8m56Pj7X/E9n2blpICixaJMWh/ofKWqBSPwntngbdx+Da0yPbdGOh/j8+P7Fsffbd0qKFA0KVatav2aUz7vwrIBBFecexuFnM7pPx8w3B8VY+aOr0dIMChPWsj3bQaBNATnFF0/vbj0PuUHQVHUk1jheCsWP/bQL8fL+1vZnGF7toWm0QyQ6h4iVqJ7TV20G4U8TNy4LUJBI5NJaIAq7b8Wyl62LggtbuCVYEYIrJok5/CrlxeHtvcuMg8oCLqOcyfSr927+jvxDmxdnw0TcMHEGK8VeTadJ9CK5VcucPZ4q1dvLb3ZUbUVk4ExEmrFjiQj3bDWbjOrhN0Ytu3x7nQXRxl9x6XqJAM4puK8BAlvxoaFGyk8wKN17Yi3JJFobQvStmrMmiQZOm25tOs7gRiKuCQQZ0BTJnMVbTsu3ue1ZvuS7qskxyoNau8XZTTG4unw3FbMR8e8ntb2eFf/ANbhCEb71I+gSYJxvR35igW5A7iiV/ZJkUAYGPcCtu35G2h9QgCpseyQOKvE7eN9hPj4ng/aGlp5pxKFfzA9witmgCKUuAQJ5Ut1i0wigKDRsu76rSDmgxJ9azHvuW3ZgUurLF2Y4igeKukXeKcftnZZ6L8fZVvFHtuu6XbR1IwqLiMUcdGUwfdNaULZigECgEEbsu5obVDdPdyovxq8beSwzuzLaRAIKITE/wBgMgzUcPxbjibfpfyfuavMIpwflQTs1wKa4ZjzVb1w6fU6QtcGRsXaT+QE1qYE5iPdshP34qDcfiLv8Vwj+wMvwDaL3Xwbh39FavX3E7/jWhOxrPvBbD3D06Cv/D7bH1eX/sCCD7O4qQRIP6s159I1pzU/5e4LRJ9IHvltp5LUkgkk7k1t7N4X/hj+wO/BcSPvtmoZQKALfWXofdv7jHM1mtxW/s6wp9UGg/2B24LiD/8AzNcvdNRWPoLY4bLcHxT4S0WMm2/RTyNBlYSrKZBB5g/2ACqns3iRJMDU6FVFbAf6CbFq4pedmfdVP5mhfDuT4F0SgUblW5UtjjYl+DvMFufuzGqiPjnD8Jw673eIuC0n3tR9p8Zst+4hThEPzhnp71sa7wsKBbsK4GlYRf8AQZG7EYWmOvZNQlpEzTG6frvJMZ+oD0oBkGMVeeymW4fiW8e0fk8lavpsGu8K4ur66Gg1a4gL9dVMXE7Ohgj4uSaF3iV34XhB4975xhas+zrewv34v8R6gfUWuI4y+f8AWX3Nw/Kdq+1WACv3/TIBjzbkz+zS3GJYG3+z1LdT1NFnM/rTMBD9lf5mhFZP1u1TLQQOgoqDKmIJgVcs8UXLLetsUZee60OItEwOLtrF5R1dRhqS7YuCUuIZVh8U/XMJThbQ133+XIdzR9m8AwjwOHY+I4/825UDnHvJO4oFGBk9DRvpE+X66jowog9Dg+6SOQzQX/GYpiQ3mA2jaBT2rSjHN4GIrLy5nLfOtjOa8+T91HUxhTy6VgYGk8x1xS/51zgDtWAZOadZh7lsufCadxpoETDrMlD0PxJDxCHTf4wjUto80tdXp71+45e5duHU7k9SawCc0ANyPXajKiYoT4YO8+Y0rAbkVAYFziQ3TUKRWJJdzkFhygRApCVQGG+uT0HYRVg+TUzFSwk1DlgWEBIWSOcUQZWdPPqM0s8zGSRzonsOdETg7Z5VGtgMA8zQfcAnsd6Om1bkTsC3SjmdWR9Y8qg20AAxknlW2BWkxmagxhhkGjKwt6zPkuKam3cHzU81PcfEI9ocUpUEb2bbbt2Y8q0pbSSO/KtMAnI3ahJYljzIXJNKA7FwScALtRJuRBncE1pXxIA3MASZpIVInYtWw51nfYGJq6IEQrPAMU5AEsQTgnlE0khQqmMhiZzWhVtxjJJrHSVHmIoJ4dozDZJ5E0Q5UlzEZpsTpIrUWksiy2k/KtKO+mIyTNCDvA5AVvdKiR9gUSQ40SpgiMk0sCfNMLHaaCmCyllI2FJuVdI/GTzr/u7kLezt+zdHw8HwVHhoft3Gwi0Xu3bj3LrEbnlHajNy+WJiJUbAUZu3i2TJhKjwregchL8qA0IqRAEk70VgFuRgDaIrzZJPqazqHPHeaBJO8/kRX4/Kjlp2k4mhBIXVGQaWGYkiCGjpO1DzOFtyJ0ximMP5j0TlE1Oq4FUloYCQJxigfDACyNz0gbVGoqBGxisosAacnsKdWQ74MnrROZBmO9SADAMnlFJB1rqaQBnv2oB7c6wPtVPlAGNgKQB/IQ2QQdmqWtqUmTJG4DGh41lAyGZ1Wz06x8O+vr4q4PTyJTTpckgdNqkJYZh0AbqayLYQDu2aPnunWOy868+smSZkcqYrhZIjf0oEgBDIEj0PXNSpUsrEQYGAKILgvkRpC0G8mA2IJkz3qYta2z9r0pvGKHPJTy2FSLdiIJO550HcS5UkjV2k7UJBGrdgOu1DB1MYnV09KBLsX3g79KOhZeMQKEXXJgHJg8qcEEEPkhSKcG5diBAH40XVFMBswTnahrtgOcZ8xry3VhvWipGLZJJ+t0GZNMCcqxPmBFB2dOQEjpETXntGe0A5WpR0Dr6MJ+G44bh7Nlf4dX5tWmdXOJwakvbUH0JrKgCfQUJFgHtqajoRQrafSmIZzc0emZpjqcsUgET1nelOi1BxODyEVIFsCVIPqKAUEJM6aK6EVBEGfUHaikBUgiT6iKBVm0MZ05G0ChbFq2BJ79ZoXPFuIkg9MRuatWrPDWmLvqjxrqEL4aExgTJrhrTG6sm5dVSqirJYgKgRHfb0Fe0r67KU4VlWdhlyoq7ZCuXd2uqxM52tFquBtILaQCSAN1G/OjkHBDGfsjoKBH1DyULygUXUtKXAdvSgI8rHbBp3cyE5hZ7YEUdSnVJ+qCM0jICpBPVsyAak2XeyfRTI+G4ucY5U9UXAoGNJn1MQKBlUB/dNZLifmYoEeKACTMqo2rBUjbAB6RWnQmzM34AelKnkDFjuCcQDU6rqIAMSD0ipI0qgjB6kVpVruRJAAB3oDXfESJ8opSQ5KplVIVYMmiS11rnlIyJ2E1Gp5AaMACOVRIkC3kdMTSeAWbSTZLPHpNXeLAWXFwlM+iRVmzrJaEtoDj+8BNeITADQBBAj7RoG24RWiW1FfTam8ludgcEny8zS+eCVgagFEn61MulYacDpQHggC6ZoQx1SeQFA8o5GOpFDzL4bLiNXUTXnskLBO6DY1BAt3R+Kn4ZDWeCcWz/5lzyJ+JozpH5RNbPkdVqRp/nIoZuZJ7UWGh2KjkScGvNK6iMFoo/rdKyfyxQ8UcxBUaRORg/dSm4zFjvjmNQ3oDMk7DFFHNvWx1EwR0maGpWLkldQk9hUgWjEDTBanHhWMldJJPzpwRMviZbNEhLKqJHPrRPh2YwObc6bU6ATAFKCtvOYkHr3pAutXkrBnoZNaiQH0tMEgHG4GZrLXFRQMgAbxtULbtgk59YxXiLeSLfUu2AB60Wu3BquMZ+uNxNAsJRl3kegoKJ1gD7Q2IoiDqmQNqnxrYBGoUs3eEvIYxLKVf4ZD8d7QE90sLroHVYDUJnNTiPuxXMmpm0BHQGp8x1DcRmlRWvM3QeXnFGfNBClp7SPSjpFmSRBlulAAiRmCZHQUVCKW8rA88TI2q3C2gQAuwo6tShSJAxvviTTDxNKKMYFRAiYn8BRi5xAYwDstABj+yYhBgDpvToZiCV2BmiyqN2GcjoIG9K3hrIVcTIjO9FAlkiDBzNZGpyTnUYxHMZqcMW32GPx9KPg2POdIMS2AJ7UsIIHegdYk6RGabUMHGM43pSssFJA3I5U7XEfDTgBfTAoHxL+lj/jtlPhkLwnA6m/x8Q00NKEK9GGADH1oxp1p8t/c0FAABgUZUNnJPm9aWPDJiYg9K1I0aQJGnX1iaF0aYDEKRtOIIokACBJwCe1KDoA8zkETtEil8tlFEgNknnTJNxnJDSfL0D1Li7qZiFLaRmJFfYaDG5P88U1y4suVwcqPUUZZmGQNiZ61BKFsCRtShXuRAET8iYxUygHURvjlNFi0LqJkgjGwimWFEKMHUOsUCBbYzyOCagMQDgHlRmZMbxRkCcCI7UTJ1DUZNSpmJgGOmYolXtwFYxkd6E8NxVtjtsjAmufwrcxUp+ltaT/AA2B4f8AKtUNhSYHrXnQkjqBuK8yMPuOCKnSaCyhB+ZxQPlIUHrSghIiNMr9xppOlmgTkGOUUzb4IAO3QjFLcAZAXWGG/wAtqkNdSCSVMADYVA8UQ5AOkDcEDM0E0WjBEDXPZtpqUW2xXIOsEbn/AJUoCiFOJBPzIp1CWFMHT9k7nfBpjCHErgkRyFHStmIIYGfnSmWbXBgjn9mjCMVXrERkmTRMuWaBqJ6bxSwzTkQ252nMY6UJZPBnu451Gq5EntWQYOJ3oGTpH96aaBCsxMxPITQVVIMkNPm2AigUTJGJJnadqDBwVMCSQcVJNlVY908p+FeWzae63oilq8zsXY8yWMmjMxjuIrSICso51EiRRJe2rbDpNKvnBAHMETilYaNUlogfiCaCB7I+qIk/KhgoI9I9KYaTEAHMgbUoFvS7FmgaZ9KENxDamOPKo6ikZg7tBAdTOBSmERVK22WPQSaMQogF953lR+FPpJAYGdRPzkxTKSmZDVInmDtq7kVbdmSCRgbTiJNEQGOZ6AczUs2sjTnPyih5LOpivIDcGPymiv6qSWEAzsZ2rOpnHPJ8oqAOMugDp5ppRqBUeoqGtgzI1fMCjcLmQcbigeTgdT6iJmhGk7mfTIonw4KzkYMiKAYHxrY7H6w+FQX4cWB63mCUIZiemKiNiK1SD8iallAowqMpjJ8pIrdYaTzWiTmYMYopKeY5zt+1WAalQYYwDv1ginGVyoI27TFOdVtmAiQGY4HlGKtXIC/bLaATzihHjKqPpWABHlMnerLMAq/U7cpNJ4guTkLJ5gb0ms3gpwsRI9aUzG3qTiFpoGDJ6wPtEUytjUIAwxPRaGuFNokDVBiSJ1c6bzWgihhpPXZv8qJAtKvMHyyY60xIKkggdJ+dSrhL4/eEH8Vo48/SOZoFCRlzODXnRsspiA1F58y6mgxzOd6t+GDMuQDgchTOplGjA/Ggj2mUokzKnBUxUpcQMvofhJ18TxymB+xaUya+qmkL1PuJ/wCVQNUAUIt3XOo7Zhqh8N8mq4GKNISBqzymjAkST1oeo+Z3FDDwGLfhBEVrJGWOwnEyARTXF0KouSCPwiacrqlj58xOy06k3iTgwoOzAkZ9KtsmtgxdoJjnhaGghjIYmRGOtGS8D63XvFTyZABBx3MVA8WNAA2mAfKCaLQZkmZwT9o0zAopU7CJ+rjSKEOFLAHn6/8AOoDLLCBt+XOpLXGE9oHWoexcCNje05g/MGgYJABxuKPlMNjHrIryOukIiyVA2wDNXQybgjXCivFJEyPMdXQziatWg0ateD30jeuM47iXK27SougMzYjqZq0vFmXuWrT+Ilot9gNAn4TjgOH1OJg673mNMWZmcEwcetYgSNwKKgKTQIIn0Nd6ZoJSD1G1NOjaGxI65FGA2cDBOeUVJJiMSfvNEHUxET9kzO5FFtdwAwpDYyRim0HiCoWQfq9SwpkVVcqAQCoxgQaUIyFgQh1AdGIJmmlpLLDiNz1502lbOSsiD0rSHJJuEJpMicHMxSlYMlSOXZYojzqSXOAJJ2Y0HUMfMCCxxH2RmhqDAZORnvND62Nuo/xUB5I56p7yJNc8Rtisqsn7xROqDApgGMQBJB5Ak1btKo0zP2lzJp7jugVy40Wz3601pSIKWfJj13qTzJyaj9G4yzc+SuJrmT9/wnhuIMRrdBrA6BxBFXuCfMWuIT9Ks+gMh1FWG4pLSXpsFihRhIgsAZoHZehNRGKkfo34hxRlmG5kk0gAI2Hm+8RUpIgliTMbZpjoUse/Kc8hWlU33XLGsHVdAApmc6m0gtyzlcxTyEMjGCZOZAqdNkAZQZjuRRiBonSMEjaaiNJhASADJkwppxvuG/Z9R1ozH1lZRk4IyQeVDMsCPTcYiiFNxgSTAkx1NarUHSGA5jtA51H6wjSDiAZyJFDQTIXSNK42EUZBxzBE9akPZcHnkCgulhnnBFBQ0edhnHOKZ2AgFuXoOX0OVSb3BWHPqUE/C4a2OFVflZWiu8tvRya/2Y/eWrZy/wDCCaBHVuR+dACRIGBinQRhZkfIbChuCRlQNIxnNIxWyxY4O+DtHSgjGyCSqFSZ5ypFOdRUFoc5HLIM02DCg6g0HtvzpbvnAMkLpX+cUFMQBEbD1ilJFssRjsOQBijgazDEDc+lA6bbFTzAiOZpR9WYUEgGScKKgx5Soydo5k5ioKls7sccyZJ3oToPh6pB1DGJoy7kAegmhpFtz9ymoDNgdht9PKWWsn/9bkfC92sf8Ffftw8+vmFSBZgbbue/pRAya7Uh2w5gGBsTQBzqKiCSx7UwR1A8zEz/ABg9aXJkxpnM1Cq+sBsgNG8KaU6nkvGcnu0xilZcyN8NQUFo0qRssDlTE+HmDAAPPkKWdEbLO3z60RpsRgTgcqDQAJPZY+01NhYBnGJ9BRIC6YYBhvtC4ilJj7PeeQpSQIYiZ/D1oh2U27bd38v862H0xNjj7g+TqG+F87P/AA194/oAPvaKlRcVS08gJo7TkQaEmtOlYJ21HqdNHEGABMb048wAWG2xuQY5VmcwYX5cqXyLC6WUlQxAIyKB1PAErkAk8hRQkHZjG3U7U8tcnUQxjIEUWgAMvIgA4MAmvDkgFPMTOMgmCKEGwAD3+dWgdOdJjpvg5rzqoKEY1TPUucTQaCMsM8zzJo5IGqcL8j/IUGGKOhHZyP7xED3c/pbcTYf+JSPhe3FBP4EA9+9gf71QXuOWjEwYzFcvcJNsRAn50zsQAFOV29ZNCZ2XT3702AY3/kadcGMOZM7UfJkYbeJzUsAY5DlvJAoaXuxGJmY5kim8pCg7RjtFSTcCgnYCdsmlcKECkRz5+UUctzJ/an7RqSYAMyBEThcGaDEbw/mMjtvUSenrvHpzNDUUABHQijtM9hWJ+nlrFhx8mI+FY1CakcRx1+6PRnMe/Asif4hWQhEmPt5925r7CgT1+dHM5EwMdiKfUCxyrYnuQaEb53yaSCIiBA70xgHSsKd15ALWUgAgtsDERFMDdOkfXxk+lTpcqST0++lcs8a1MiBykAbVjxVUSeWOppk0hcIs5yc4xRgsBqYYkxiST+VSNQXcHIg41fyFEEAkTPJZxPryFMCFXnWqUaPQRWVzBrVf4H2KvG2D0Nm8Gf70BrcfRxf4C8h9Uhx8KhrXButr/wBW9+rT8TWwHv34dh+Io+RUiZGwise6CYjyzv3G1KQhWT5ftVkgk89vSrmw5P1FMdCAyC2d9s02ULCQxjA3k010CTGBEHc6iaJkzbgDlvp3qWN2NzMT0EUJYkp/dI55NHU10Env1wKQyIaV7eooGAAoQ8xMbRW9wmdyW6GIHLmaHlRpgxn6vKKiGqCTiBy6GgDctEEdxQez+h8PwzqdiLrPIo6uC4p7IJ+0gMo3zX6OHa9a/jtN8K891/03ieyJK2l+h9ZY/wDkKgFGz+PKoA9aJAmlOkiTuaChipIWV29JoSwJWDJ07HlSACBPlq3pAA3Hm5kUpbQTbA3YRuKRoXUNsGTnAqdtp5j0pBp87O0bST3NLp0NpZCdjzwBWDcMFhuc4MmYpMT9UxsAPsiisqByPIic/wCVAawxXVgnl9qfyr6xQbjm/f8AyrZxg+leYMSCvoRmvMrkCMHIio8bjeHT+BC1eTi0HB8Uel61lGPqv0d+Ps/iYNc/hASxYttduudlRBJNEHibson7FpcInyHu5+7J/lmiZYtOCINDoJ9x0giNROZ9aVACPsyOW1bSBvsfmaP2ebU0jeJ6dopoCwY1b4phKwCS2fvqW0eaIkTg86BWCRv67KIohtJK6cEAkd61eZzg5P3CiJMZ9Y50pjAa2DBEdgKbSZI5iJG0UoUXFEL1y2wrPilsjHQZHpRnVcJ0yexFEbEeXf7vSgov+0rj7R9RFWhrv2C1hj9i/b8yGirKxVlO4YYI+h/t6/gCfhMXvafEC138G1539/P3bW3NGQsj5Gh5mbt2rTiScUQAFNK5OTv0HYUIKzgL1ERShtI5CrYBAGBhhyG9AbapCmTI2zVkqRjILLiv9VoxOANuVGEXm3QdWoBhbBgRET3BM0RCGSZBMneSRFQC/wBfrvzAIo6onzdFA6zQhiIk8zBgTUnUxAacaRH86kwTmZMmedR4aSQZYS5xSyLhJEANnOD0FEm4brme7mvtCh4Q9scZojp4h+hizx9p29C2k1sT8Im37N4VLPbxbnnf6OFtfma+yGFHkKiPKREn1kTWYBgxTKYO+R+cU+e7ciO1EyASOm3aswNyQR+FEmRzI/lTo3Lz3DtjnFYe0FG/86dCoA1AKTjtJrAs4iYqP1XICRJHKjoBk+XPXdiKtnQsAKdUKCBBjaoZ7RVv+lk150XkYgACKklF2Aiee1GHkL6WzigXnYKR2zJqCOCQ/wAUtX2hWR7T4sH18U/Q2BIPcV/S2LT/AMSA/B4RAXY9FUSaluL4q7eP77SPo5LAfICjgRWNQoCJgbyMGoIAAG1JJIzicR6URCEmAM533oyUzvvI6GiTg51HaRH1q2Ktz7nrUywbABgjkJNAO4BiQIwPWlGk7kFpAxsQN6+sB4f1fKQOdNHhxOY1c+1apQkLJEiMxAJo/wBC46c6iOHBGZBfvyrVFkkTgGYosQdKsRB1HanONOGIkEVLMCNh0zM+mK/ouEsp/CgFfaFYPtfjSPneb6HIj76g2+Hsof3UA+DwV9n3UX/Fe/Vj862AH0YLHWKJ7mjnPOoQCYnUD8yKxg4FFWUgQ2JB33FSVG0gwBHavsgqxI2HSlZQBJLLP4Cp36E7dhRLi5OSR5em1MLhg2wNZAB3HIUNGmIjM0dILARq2C9qZdwNvN+eKk6STkkbRFAMbEMNiBmCIkzURaBxAgQYnekk22HQxqERHOKYqH1k/gJmi3nB2J260w18WoUgRHiEDn61tj7q5zX9JxN25/Gxb6E+NxtlPvcTXM/B4bjOPtp+7aBc/R51LRBWjLQBQx6VAKwfu3oE6V5A9KycnHpPOmb0LLnGcU2AdXlPlHb9qiNl5iYMdqMrO4JzpHVq8+dgIPTE0sz0H7Q70Skr51k7yT9QVhHCZ3IMZ81HypIgwJM9IFLlo2Uk5B70zfqYkiJ39BWkjhgqjrp2iOvrWCh/MCt5A5d6JIQsNJ2I6xSlW47hbhAWBvXOv6Dgr93+BCa5/Qxbe7dP7iE/CPqcPevt63HCD/d+hvWFk/dmj9b86MDOBJr6xZR/0a2B27elGEVTAiG+8UDtnEZ+6oOnoucjO9E+ijv3qWDYAiZpZJJkxI2FDVDQSVgGOUA0SfFwwLbTEYArQuCSxmIBO5pZa/sCJ3ifLUAoARER99Qot6jBYZmZoDxLQCTjIHeiwXhlZCQOfLpSjc9smaziBNCDqAzIx1HWtOh+H3jk4zXM1Eez3X+OF+jPg8C5+bkL8IkcLY4bh/mqaz+LfQ2FclRT3c1iZI9aImY3H1anTeiO5rZCfxoxpzg8qWWAjPz6UZyDtSgrceBIknsIMmo1M8jONI9KJPiFY1MMU2nYeUmO2BUG5fIztuROahkMSI6dINMXNwCCDETvmBUFDAURJx2zSnyFSSIhQcnJFa1a2BqVpAxONK1/siVG1c6ASZMRj1mpKvaGeUMT7s+FYHya8v0drFhPvafg+7AVi57S4gj0DwPobmove1vaT3pP/wBnhbZH4s9bACoI1A4qPOp/Cmgoen86wbTaSOg9KzAj0oHU0Aaj+PSoYXGAAnev9WTzo6iz4npSsCFw8EAk9yKcEyQYWDHSJmhpbOqZ37Y3pJN3GBqJBJjnRDa/2jAOMQYoKBb1QADmeneh5kOqV7dyKwOGt/j6UK6cj1qF1gEgn8MQamWMagVIC9Zre0h+9RX2OG/46fR+vxVm38kUn4Pi3be4f3FLVl2Zj+8Z92w92HuS5/ZRck1ps8F7KlEHIXXj8krnNSS3Q7E1kQD8sUknGmROe1D+jcDH7O+1MGj6oDfmKMl9IPOsqSTWdB5qdjB2oNbO5IB05+6mkJMATHQnnFFy1tioyBnviiq4AUBYlTjDTRGkySpOOcjTFEg3RDNHXnM5qVKiCJI77YoCTKiRJEbjeiR4NsHoJ9fdzjnFQTBgTv8A9Clyjttpya+twllvvQVt7NuXPnah63UfQg8Txt+78lhB8Hgp7M4iPV10CtgBWw98XuNPhWJ5WU3PzNY4e1wvD/daDVuvlxO1eY6WBjeay2Rg7gzypWaNQAxgVLBXLbYFAHVAOPSpuBwIxmgMHeK0wm2f5ilVEEMzNs3zFbWq1nwj5cTH7xoaWQxBB1E5IgUHTJCkAlgBzn+dDR4sggxGcjyivMF+tiVIG4LGjOsATPOJwsUTpS3JJkwT3rmy0OsEGgAG+VLOg86/2Dh/9wVOv2Txi/faNboPfsKzb4G2WHRrg1n8/g+bp4eyP37gPu392b93zt+xbXLt8hQSxaRLVpByRcCp8X2neA9EOitsRG1AFVA26UMCNhR/GhMTv0pGkbSMSaEEjmBsO1N65j5xUkLknp8zTL5tgzRt0FBR4cMGBJgesVDJaGwBhT2g4pHIWcKIB6g4FWkGk5ELqkAeaedT+sLNmYWd+VYufUC4EAdoiiP1kZx/vGhMIKWdYbbO0TQwI2pZUg9MTBzmhGkjet+Asf7tYX2bxRPytGvsD3z+k8XatfJmg0ABgAdBgfB88R7TT7rSFvfvXn4gnhuG/wACZdq2M/dU/rXJODLMZNAg5MigRGBS6tUEczQwGiOkyaIgdDTfcf8AKjE4BBH50uWWVIz2OTQLAZLCYHekmR5THX1qDoAYACWEbHc0QRaGxaIB2xFDUEUFgVJM0Lg8KNIYRQXyjQxgaiJxmTQAKnTBJnG52il1G7tIneO9baOtfWCNy5ZFDBPLrWokRAxuZ3qZMcxv3E1kcGgPyxWR7J4hB28QaK2Qe+V4Sze4g/JdK/CMrav32/fIQe/Vev3ks2x1ZzAoeFwlhbK9yu7fM15VRmPoBJqGcl/4jPuMMNxJoyHBBzyrkCc0PrwSQKG+8UZJmcUfXMb9qJLDeG5UGbUVjUZ/h6d6ILoNiWGO+1W2OhBA/A9ZohPs4OBiM4NYNvSAoGCaGbYjnBHYRWkqpBaVkntM0N8H5+ooeYKTtuI7mjLoSexn0rdZ6fnFGAVGI/a5VKj8/lUn9EWoLWrFr/3byrWy+/ZLPDqfUlz8Ixw/A8Na+ZBc1yrJNTZ9m2vJ34i7hfdBT2ZxMHuUIogBVBFZBxRwTiTX2TzI7UY5mTSEBhLVmcARtW8RR0sdtRPM8jTdJBIrnJknaa0gskRAwOQpogQFnGO1WykrpBUz23NLbhZZe3oZoKVUyTzOmdzRMjPKD8qG/wBbqKPyJog5FcooeXzH/lE0rOLevEHETiYNb8Fbb+LNf/Ue0+DtfcS/0Ibi+JvX29AdC/B92ArF3irhX/DML+ArlWFBNRf4oHjL/XVeyoPovuh+O4m1w/7gOt6ECsEzvWOoJNGD0msDG9TDDCkH8+lc+or8f8qO38u9EVoRSGGrOxr614Zx92KQKGgB1Nzl3NLOtciAsT3nNaQqiBJOrqKUHO8DlSmBAx3r7RwOW/Sj/njvQ3O0VsBSmRgt1pWi35ucEbb1twHDj/4Cv6f2q971Fm1H5v78wag2OCso3+LTLfB4XhuEvXfmqmKyc1yryX+KXxT0tW/O/wCAqByHQDYe4fqLD8Q473jC/gtZrfByKMbgGSKMhsdKIzJ5ZodiDvHubmTuaBIExEbCO1EyRGBkztWgpBIaAefSiZ4gnrjpipBuYx2oAG4JExROmCv/AEa0kyRqIWQaEjEgQRJyM1KgggR360APkIo7GKO3/W1KQCRO8GJoeY6cdTX+yWh9yxUpwvssXiOj8Q5/knvkX+OsWz6M4muZ+Dw3GcTZ4b9zLt7vs1nhfZd1k9bzqnvAt2L/AIK+lkBIoAHtyrsIontJxFCSBEE/yr5yfd5cgetGY7nl3rlyFEHHIUbhUYWdhH92iR4pLYJxPahpYu0ARS2xqEwTQwRpY/ymtpHlk7GoXVqAGMmmntv95o784mtycAzRPl3/AOYpHIiZ5wKGk8SJKkHaCZre3p+5iKm0OJ/RrHTw+GHhyPUgn3/UuvePpaQt8IyPatgD5q3u2FQ/GezGFnu1lw5X3eXh7D3T+4JFW3d7hd2LQSzmTRJXaINXy+voSu3anHchhVxR4gIkMYjMSaZdR3IIzSMf8VIxAlocCAfSakaOQ69SKMqAc9O00ygjfyjZa1KEltUb7YIpmuCZ1q+gHlFKQFnAjJNLEkSGmMcxUIXAmBuDApHklQAtD76P4Vz696O2aO0TtFIsAkt8tpogix4gMrBZDpJEdmrz8Lw99bP/AK1xylv8WqSBknmeZ9+3BcT+XwjPEe0HvH04dI/N/fo4nhbou2m7jkex2NaUvAi5b52rqYdKZPC4BrgYCZZCCq+hNW3xup0mroxuIb+Yq/bjabXMelPpDEzoYHPyoLDjkas84ll/nXBvG8PbarRBmNMD8qKiN1ZqvZWAByM4ORV0lWUeYLGQRSMCpEm2QASZnFDCmDbwdu9XACkagsxHpNW8A7gg1Cs8TjTv2pCFIhoEnnua5fl6e7lzMVuR0NHLCZ3oAwTnoBUrbLKY6PvUXH4+9xfEDn4dnCT6s3uwB7sWeA4h3PQNCj4Riz7PLfO/cP8AJPfuKxFrjEHQz4b1niX4bhk9Xuhj+C/TFXUHRXIFXgO51fnNE+XTlEOOm1cO8RvaAmOsUGn9i4Vg1cWF0/0Ycfg29W00qfrh7ZJI9KtcUhGVmYJ9aZSXBhSFGZ2irg0iCHRWAkd4Iq08thwWXc7QMVeWFEQyvJq6Np8lMCGEKwAmg3VhqzTQSYO5lhHrXEW7LXCn6nkVnDevKod4UKNkRRAX35oeNxfGG1PS1YAgfCMWEtWB+4nuEkTWA4BqVHsa9rjqLqVl7t/jbo7IPCT/AE4muITtrJH3GrN0tMs1sBs43WKYQwM226esVft7xKSduoqNwNR0iCec1w5aV0/rF2q1GoebXXiMQR+rGrBpksK58K3zjq3f6ORxfFa/XV8H3YCpD8ZcC+gPu8iGHHVWwaMGvqcNwthfV3LmpTgOE4fhB6x4jfi/9VHv2H0f6P2lI/fQfB5u2OB4i6g6slskVOu3JPVpzQaOR5xScJwPFrjhrC675UMVMs2EpinB+0eI4a2XMsUtOVUsa/8Awr3yUupqf0v2hxF5f8LOdP8AV9q2MfRwfaKAfJPg4YEEFTsQcEGmPs/iS3Eez3PJft2vVKHlzmlcWPE4e9H2bqOSRX/it258rsPWgJ/2T425++joErZR/Ws9Pfke1rvzm2vwjXxXARxvDRv+q+uvzWgVO07EGrXFezvaF7x7tuTadLygKzI3cUBY9qcNw124N7tm4g8J0ccnGirjp7SQcNea5BbRauC7AjqVH9ZwcH3Yn3bbmvNxnE3uK9FPkX4QDHI7Gj+gcVN/hG6ITm36oaZb9tfIPsivZtjieGYuF4+/4FtiOQeltIeKOm/YvJfsGRsroT/WDHbtUxLH94/8q5e7JyaMu6W1Uc3cgUPD4axbsp6W1C/Cblh7V3xeH4m2Az2niDg7qeYpOP4NM3rnDqyva7uh5VHEzJzUXVXRdtXQDqoCxdJezGwHNfl/Vbl/iLzhLVm0pd3Y7BQKt3uItAJxJtQ1pL3O2jgw+jYsMTQFxuGRIBEsw3Nea4dXoK+u4X3YmTV4cDwhfiGu3EKK1yPIqzuZ+F8oPQg0fZ3GtnShP6K57p9muFXj7vtlOGR/FLzYFlmbKHEmgq2gYGrVk94H9SCcLwl+xwwYifFv3zItr6KCx936LwG7+0OMm1YCDcpzem4vibqG17R/7QMIe6vOxwn7FvqwoAREUTG53gVgCBWEWT6micBRHVzFXU4Dgla94t1CqXLsQipO9E/1n//Z\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Recommendation 1:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "Recommendation 2:\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 150,
              "unconfined": true
            }
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vWQRYc34ntn8"
      },
      "source": [
        "### Vector Analogy"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "kNnaLfxXmxTV"
      },
      "source": [
        "_Playing with some analogies here_"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "G246yUs_pN6D"
      },
      "source": [
        "> Note: if women shoes goes with women tshirt, then men shoes goes with what?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "arkAwb24bZkg"
      },
      "source": [
        "# fill here with your product IDs to test for analogies\n",
        "PRODUCT1 = '01112'\n",
        "PRODUCT1_MATCH = '01117'\n",
        "PRODUCT2 = '01113'"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9vcm7i98YtuS"
      },
      "source": [
        "assert all(_ in product_embeddings.vocab for _ in [PRODUCT1, PRODUCT1_MATCH, PRODUCT2])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 417
        },
        "id": "TgyOLfqib0qm",
        "outputId": "a0671b5b-ca64-4921-c47b-9fea1646fa83"
      },
      "source": [
        "matches = solve_vector_analogy(product_embeddings, PRODUCT1, PRODUCT1_MATCH, PRODUCT2)\n",
        "# first show products\n",
        "for _ in [PRODUCT1, PRODUCT1_MATCH, PRODUCT2]:\n",
        "    display(Image(products[_]['image'], width=100, unconfined=True))\n",
        "# then display matches\n",
        "for m in matches[:1]:\n",
        "    if m[0] in products:\n",
        "        display(Image(products[m[0]]['image'], width=100, unconfined=True))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 100,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 100,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "/9j/4AAQSkZJRgABAQAAAQABAAD/4gJASUNDX1BST0ZJTEUAAQEAAAIwQURCRQIQAABtbnRyUkdCIFhZWiAH0AAIAAsAEwAzADthY3NwQVBQTAAAAABub25lAAAAAAAAAAAAAAAAAAAAAAAA9tYAAQAAAADTLUFEQkUAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAApjcHJ0AAAA/AAAADJkZXNjAAABMAAAAGt3dHB0AAABnAAAABRia3B0AAABsAAAABRyVFJDAAABxAAAAA5nVFJDAAAB1AAAAA5iVFJDAAAB5AAAAA5yWFlaAAAB9AAAABRnWFlaAAACCAAAABRiWFlaAAACHAAAABR0ZXh0AAAAAENvcHlyaWdodCAyMDAwIEFkb2JlIFN5c3RlbXMgSW5jb3Jwb3JhdGVkAAAAZGVzYwAAAAAAAAARQWRvYmUgUkdCICgxOTk4KQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAWFlaIAAAAAAAAPNRAAEAAAABFsxYWVogAAAAAAAAAAAAAAAAAAAAAGN1cnYAAAAAAAAAAQIzAABjdXJ2AAAAAAAAAAECMwAAY3VydgAAAAAAAAABAjMAAFhZWiAAAAAAAACcGAAAT6UAAAT8WFlaIAAAAAAAADSNAACgLAAAD5VYWVogAAAAAAAAJjEAABAvAAC+nP/bAIQABgYGBgYGBgcHBgkKCQoJDQwLCwwNFA4PDg8OFB8TFhMTFhMfGyEbGRshGzEmIiImMTgvLS84RD09RFZRVnBwlgEGBgYGBgYGBwcGCQoJCgkNDAsLDA0UDg8ODw4UHxMWExMWEx8bIRsZGyEbMSYiIiYxOC8tLzhEPT1EVlFWcHCW/8IAEQgCUAJQAwEiAAIRAQMRAf/EADQAAQAABwEBAAAAAAAAAAAAAAACAwQFBgcIAQkBAQEBAQEBAAAAAAAAAAAAAAABAgMEBf/aAAwDAQACEAMQAAAA6cAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAWzWpt1rrL6u0q1Yyme4zWanlm7w467FAAAABrI2a5eynvnfDDsp4aqAAAAAAAAAAAAAAAAAAAAAAADRZtqbwTfKy/WN1tJcsw1bPq7ZdgNqjqe0c7XGXb3WPJfWKRgAAFOYzx5can15yCiiyD6/DE9jx4F599fR612V8bsEAAAAAAAAAAAAAAAAAADWxslzJkuptjEuT4O2d0abttVx1cIqSoKG4VMopbPnNTLg93yWnMbseQTTDNl45i53Zmfzp3fHU7E8sAHNmX8u6VOwIX3OFxu14pvk9rVFXZl5dYh0TrTOaugsAAAAAAAAAAAAAAAAAWzSulO2eneT6G64tLZsnp4hT7cVNzt+wzH4LbhstRFMqyfOpFlym2iPavnUMRcJ9sm4tFZ8ousa5z+5WUy3F5FeYTmMqy9ZnWfVOG+rEMe1OUfB0teS494dn+c7Zll08KAAAAAAAAAAAAAAPNYG0LbzJqgzSz2Gpryuk1kUMyvkdFjvVRTdszbLOnZuOR36VzW+ugj6yV5f8k9eMA2Vl+GeXeNUuY2P6XKkin0u0FuzrL+N0fT9CWP5vXAM8tmIcLt/Wc6lMRrMazfbaex8AzvLEuYdwajIva+sqw1UvyO97zgWewAAAAAAAAAAISJZMVNitE4vXT2BaZwrNy3St9tupMkT59kFVOoFrJtunk32upitrsbg75zfzAHpzmfmL3jvm602UZT1mus9v2KfM6XSz76snk3gVBLl+jM+hv+Xcdatl3bFc2upJNo3L7J0rMwvdhyvHKlZdis/0Z2nddJQ87eLdHBizY6apIfJsk251rxT2tkAAAAAAAAYXzx1nXOI6Xsf1eO79Q2z31YoZ8yd0k64Wmd0zbLDmvvDet6XadB8/prDzPcc8e6ZJn8Nw3+zU/bORUuzMr9mOcKDojnn53WZOoKWLpkuBZobUk2LL8rtXVmpI3lJx7DDZ1m1XK0q6SdgnqzsGxYbSd5X2CsoPPYa6TI8+o0PpD5USo8q6euKSPyZUM6CaJU2WV3dHBPcmWUgAAAAAAYbmHHRhVBIqKq7tYfPRnZWX6Y6h9WLVN2phvm3o6TdNZfR5Z1b8Hn+fWQ3/AAHeJiM7JdPEcEmi+d1u1pi6J1NbVGxbBuYFgFxoeWr9iNVSnm+9O5p7cZdYbRTerE/IsHs3DWxotTy/HvOMRt3lV9PKc6exxNp5sBHLj8jx74T5E6SRVdFUEqLIcqNaTs7tpjK80JT9wcSdzRfgAAAAAAY3wv3fwmU1TTzKm+Q+kdNN8JU7yIm1dvl1cp9k2B3zm0/X+Kkp7S8NZZYJVQZheteSPqctiY9YaDz6rrb5RePftVRVJdaK3wblVSeRYsCZ6QJswlRXapiweZn6YXOy+3VjE2uooi88CGKEjpYszLXlmLWYz+yYvOq5yaWKJ8MuIi2lqr0+gFRpTdcAAAAAAOE+7NeHGnk8SPfY6lIoCJCIpSSTIU2ovaWoIab2AuMCQVKmnFVRSPRF5HEM6TEQeQ3It8WSemPXOdaytgkekiTXSyhTJZMmyIy9XTFruTZlwuZYam9yy0YrmdoMcXWzk2fHEIoYjx6IIZ1WZj2Rw52RF7AAAAAA89Gg+c/oRy6aqt2WYkRzafyofIvCXBVxFHHX+Fqm3CUUcNdOLXJyuYYhOyuEw2qv0BR1lNTl9uuITTNbDjMorZMM09SKYr5duF1it1aKar8KGOppxUS46qZ9umalyl0lfLRS4YIl08yfEcUc3UlTpd16Sln11XVpnXO7ZWjr3kfr/jq7AAAAAAAAg5c6mpD5/wDu/cMNe0kyMkQxqglzfCot+Xzqw33YeGFt898j32GaS/b5b6pGWXGMB9vF1MRnTfCm8mUhDLnwkEyrFJ7AK6ZRVJNmS4i3V3kNVk6XBqR7M1ZU5t9wW6SYI4STWQSy4z7P7tefLREXKlp5+VN3VzP1xl6AAAAAA8lk1QUJfWJW0z9q+3mAan6ItprzOpkg5/rb7hhc6KvuXWWKrhocWKfSZRuWOnzTEjJ8a8953ZeN455Zd52MU0tdKp5ZOqIdqlw3NqGCOn/eXITdeiKqE0t7tHCyzzKXyq7ylFZc7AL5bqQVftH4Vak9KqKijKv20VJcIrdGXGfRX42N1BpbZsXxTzz0AAAHlNVDGsO2sOf7T0sOXafqkcpQdYeHJvnWY5L861HJNJ2D4cbVvXnhx7au1/Dia8dg+HH9B2f4cMT+4Fcg+dfNTjWxdzsXiC4dnjjx2GOPPewhx67CHHzsH04+dgjiyj7f9OGXcw4Yd0DhqPuL04zo+2BxRM7UHF1x7A9OQo+uvTlK4dO+nOl03uNU5blIoqz0AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAf/xAA9EAABBAEDAgQDBgQEBQUAAAACAQMEBQYAERIHExQhMVMVIlQQFjAyQVIzQFBRI0JDYhcgJDRhNURgoLD/2gAIAQEAAQkA/wDw13X2GE5PPK62i7KeWZTGxWrannHoOrcq3t4kd+p/+A21xW0cQpljLd6z4s26opGhdVsHmmDaW9beVdwDjkCSl1FelPxYzrkHKZr8g5GTK5xTXV6S0uO1oEeGM93J8Xjfh5x1KrsURyKwLWXdSrIAnLdU/VW1hbN5DWQM/wAOsmkcYv40uLMaR6LI/qE/I6GrPtzriHe01g2bsO1z3KXsouXDA3nnxcUzagz3GHWn4z9x1PzQ4JRlso+X2jTjpsWlX1Py2vNFC+pusFTNQWruu6oT4Eutxx2DYdM20k5/ARPws1yljE6V+YuoCLf3Lsqeb0t149ndGvrxckR4s0FSSz0rhQ4NttEv23G3gBxs/wCndRepSw1eqKN+VPJ+QadyDMiwIE1koRvoomUdZFfeE0iSmRSVDLkba2BbbIrk5JrJMux1F1r9GZShummpYp58ek1rWQcplzLKe2426AONn/zy5UeDGflSXc+y97KbV6aS0UBY0JnREJIrYrRYva5I25LZN/pwDrK/D8mmxrfH5rISmMZyQpR0co/6ZaZHWVKqDzj3UVpVUYlblnVS2fCXVssCx4pSJX2Y0FhUVqPbNR7JiO5InRYgQH3CZV5TRw3XURkJKKm7+OWQqpRmodDLAVV4BrED1GTCpQBFnOWFZ4NxTgSWZSJvrEOo11i5A007jGdUOUgIxZH/ADdXc0WY+uO171PXLazwJRQBISThRVb2RW0WAGs0uZ95fRMJxo/+GV5Vikilypc7tqtpKjO8cXJcZlHhFDiDAEhihJ/SMz6jwMXNYbDUbrdkAvor1Rd9XGpNCydVGshtLYjees6ynq4shpyZqXUWqPuAAQwSOLMZvSeLebNRPsNMpIdJwXhYQi4u5c7HivRYzTRxHI8R5RCSA+Wu+zsm+nHxXdBG3hRrTw/cJpllkQBlJtTGloR7SIkyB5utxLBxogMHMV6x2kBGo9olDm2O5CIpDn/YqoibrrP+pUetYfraR8W5dpKKJFSLAYgwwYjA6ncLbVxPZ6Z4i43wwTH0xylfs7M5fUmmhSVYMK61q8hgI+wdVSU1Y+UmDVQpjPIIhO/0V66qWCUHbLM+odfSQSZrn7J2dPN4+5BgE3ElvOyXrp4/4Kq8oApOuC6b3FGwbZV0o5Gq1hRapJ7rVZTWluw5KitSLqppBNuqalz5U14zM47CNqhroHSTbzSSiErauA8+YKqRxWSqIWuRLyEzAXhbUxbF1tDQCVVB9FTUvGmXt3IjkmkvYIdx6BHslbVNjqOpOS1QCDFv/wAasqQdu/ddSshtgJqXbx0n2ndcjBUxArW1GLLJ9DQDc1iVOsJDyq21RxZPULLZOQzx6oZGbZtVUd3ZS89Ynk8nGLJJApAlxpkZiTHfK0GPl+LxS/om6bb66hdTiYdeqqU2LrKJjivLbT7mQsqQTpQt3YjTqiEhWSU2ddkEdlI1GGHxVHeHihF9RVUmAqp51GcQKSkdg+EyXN7bInQWU+quPKpkbMUkVREG2EI20RwjYbE0JhJYIjYjo5YqobxxsHW+atVyXViCAIsFcC4vOTTDaV5g4yhiDL6/4JR35CIroqzYkeyK4YU8gFWdXO02KOKShBGgxf8AzMtwcbifNFp7CeDaFLBXJQMvSXGtDS4niMdmwyK1tLq/6pT0r4LcWFX0FUzBiJe2B2drOmOEKbJoh31gGarjry1tgcjJYU/qRjL0OT/Qbe0i0tdLnyjyTKcgyKQ8+9NO7tqWuZMbqsnnJNxwmn5JmpKalXi86Eh5iXIN5hmO9WtuQwIm3ZokB+cZyQ2oEolKn1sxXlcBompAeRsJGkl/oMwQc/M4sdI3dQiN4eRIDao+4iIbjcdxDZZSPUdM574o9fzbimSluLCp8QMYN1FRCOCbiiDEYLkKNuVjBr20bcpzVeLCm7MZIUfaentcCVHajF81vRE66igdKbd00CyyF7A8EqmGltrzLqMsYkwTj2JTh7jSo9VQbHIrRitiahdMIiSfFZBcxmI0CMESFGz+zStxmdwVPNUVC+wg1DkHFebdbXHrULukrbEF/nVVBRVVbjq/hNS4+yE+6653M4ybq2LbKbOycU5ViEh3tAhlGc8QBxjjNjABGmGgZKEicgfM57LDitxwi3Eoi7MBKS6eFd5D2O2exm5NILWsNRGx8UTiGr0Ap8UU2OEc1oOXbrHZcp5eLrseDzUUA2qxlskQ0o8cnXs9IFe1YXOJdNIxfC0xqxu8gzkAulydgDyq9EERtERRTSqKNrx0sttdm0KtxbLLhsPBY+10xs+2BXGQni3TqtFz4jZnm3TfGHgdpaGx6xWE8DAKuwz3MLJUZ+LUXS3JsjUJVmfUO7oHlocdx4+55IhJhNwxQPS561OM5kF9Y/DrKoYI1D506uWRuzaytRQTzJdq6LHlTY7Up6zOqKQ6NbH381RdAg9xORdNpVU7iVSzXu/zmU9ScWxJ0Y06ZlvWQMigP1kKuiRWEDkao6LW3yIw0huqxGbZRDaNEEWQQOaF4N1NlZCI06Y7SHa99lSdTRTbCMSI6+1eTEZUX4MO8oVNCsmW51NK2WEy7IBlEQ2nlZcQSNPCsoXHsIwzyVtG5LQNugrb9TBk3MuNWVyZTkMTAaKPR06YV08cguhkOTjfvOY71ciWkp28wmZk034lWWDfTPim9rmDtd0qpUA5s13q1itKrgY/SLlHVXJ9nKzGm8Czyz/9Yy2fh3TPGV55Nf1uSvzpKwem3Tpvpa9CR2fmV+5n+F4m0UfDMfvcryfJyL4vcgPEdkT/AMCsVoGIwcww+bUQLGZe2cuz6qMND26yBa2kq2dZmTiDbiiooGqKXmir/c0/XSKW5Ci9ILB1nK4kcnv5YzBsVIzk5Lj0NC8Rdz+qWIQUJAmzut1cEdg4kQ+quWWan4BYXUC9bhuKkmyKDePLKmtP1NdDaEvDuNoKmIAAAKHpF1FYJ9xAHToPvIy8MdHEJZCFXNlGLn2pxtWAtm+LCWJDxTcjYNARWTaaIlVp9yMRCKmyMmbERQWUNpMZ7jj0SG+UplSiBCxLMJyALGOxunV2QGtpa19VQ4lFsZEKd00p1yW4m51at7ivmusix2myqrOstmD6YZNGBpmB1CDpPKeUStc+idMsAgbE7XRQrqZkzra2/wCq+OQSMGZASuqGbkjURpnEMBxtd7+7yXPphtFWY4DqK+4pvONMk6YgGpmOxYkFl4LtR23RdDsi77LLcJVXnzMlHdePkioMolR9wELSJuq+aJrbXmiLrpxBKfmVKgv/AMmRgAEZnMzmgiuK0EiTnhE2qwIORsz8omlMsIH3Wq2yXnjQY1WofMcXj17cQwWLUpLmnsC2k+trJcnxrs2VVWLSqcJJjzrQLGlRBaCSSq0psEPHYl5tqqEkGxdhOg8ybT7bboKw+rkVwmDs4Eb4HJZ7jMT4DXmrcivlPxbpkAKZD8NWvbCxLLvsOmxIaF3bQS0AlLtOPMgivRgwW+co8jfnJZzuqFIRGq2GO51U3V9XVR4/1FcKBg2QjGTBlai4TirbeimIn6rYKmlnOF5Dq4yCro2u5cWkvqdPs31iYnj54Nkt+A2Ga5M1a4Bi3IKGotMyye5DsPTyRGx4N6kNkrhoii36b6EkiiijoTNXN93SXlpEJG3V4om/roR1GBDkxhXRbmRrsiJtoRFdIP2H5LvqjtJdLaRLCGdBZHcUtXYuN/yGW5xUYmwqPutZhJvpTr06ZLuficcYsmSVY3/o2q1k0F3SSMSwJNlfRmSii2U9ISqaqc9uNH/xN5ptwAEAMEfBtVJuOj7qIiqZynVFOayaylmeUmtfxRkvlhWsmDNrnWxmMSYwxHHmHSbjCY7prw7ja8hV5xIpsvCVDg9rcQYdutixheORBRyfkNkXSmIkppah093EZUjVE0TqppT80UhxjD7jJIr8mG8x0yaRQ8VcUnT6vrp0eYxV5Lj1reUNjBWBgGZwqyCWM30h60p2m+87e2fUvE4G4RHQtepmYGTFTBaw7DMffN/Jb93qEUQVjY3ST7OVaPnJtJ5TWgFERHZ5l5Isd9zd+QRIu6ab2RFcLRkpkpLoPzJpxNyXTaL4WVpNJqAieMjqqiKoibt+Wh2RE1v9haaT501hCouJY/t/IZzlTeJ0b0xNSrCTPkvy5rxtoaKumLOwjKotyGcmc32kx46P2BAkKLEwTJpKcnmovTUA4JJvpPTmGrP/AENtPZmVsl6FPaB4kTcUN5toANx168rmOaq67kgMu/LEBxNlRDFwW1VstXKlEqpAWUewfdkyJCkba9vy3ckgDZGaKSFsTj0e0zi1aaaq4LXTrOLUtpTGVYsWEMxRlWbbfBERXYcKhmV805kg2yFVQVbQjVRaWi6j1GP10KtpMNd6p5tJQW66E9lWdTV3cyhi6yivmDPj3cy7wbLRV3JoP3X6PR1CSWUrlGB0SmGP4pa53k1spC9PWQiIm+nJqEiigk+ZKm67r5out9/NdPJ2o7DKInmojs6u6oCfYH5k0W3It9Nf9pNTf7IiojylpSHffbkug3/uirpF0umU+ZNYSChiVAi/jmYtgZmvUrMjyexjiDYqpea6HS+e26ACGQpqlvrjGJayKyTiHUHHMoBsCMtmQU3DtOoOD1guBKyLKeqFRfxzZhYy7bTXHWwRxEkvjuWkaYacRJD4eJlPBVtk090vx6KwEmY/1diU0d12lxibdWttNObZSuCtiS7m4arsiVcWJaWMZqZY1+V9LccZ8JUU0rq44YqFfhErqnm74LIixredLyG1lW1i9wRNPuqooO4bl8qLWwHX5HI2v8JGxJNSJTYNibxOT/m2c0/bRiESUnp7HI1FSkoqr8ivOqiDtzXy/wATZPP5di/TXBV9V46BpXTBtNPmjrzpjoPJVXSBpU0nkul9T1H/ACS01vpF1HJEeRF1Aw/JbNd2q4OnkSMHOyyE6XF4n5GnGqcV2CC4zA8tm3GWfQDBswJF1iwdvG6Ifx8wlLDxa/fRZ26vIm7fppNLpN00q8k0TSKmytk0pJsqC0g7bKINbChojqbouimA2QptUVN5kc06mmg1/R5inaV/K8vmz+k+MwJTtPXypUyfLcmySir2DFA1cPUkmTxp2FbGQ+LYrU1bUMNyFRBRJsGTktA0LiO3sgZ0g64DdAQHgKvOoo7Ii7F6q3JaFhWnGZNqTEh4GWzuZhuC7zWbLXmguqZLsqn+g62Xz2FEL+6Np5b6QBTWya2T7NtMpsrh623VEVBHz0uttcdKOoSIr4N77+W66rcemzR77yxpdPjycayLJy2Y9uhPOXDxb/OssyUlQu6q7rvy3/VV1zVNYR1LscedYhznIkuPOjMSozv4uTVaXdBbVynLjm0bRrpEVNIut9b6Vdb631vrlto3iRNLuKKBJSybGmrHGG7F6RAEXpT6TJYz3Tfcc3aEVUdOvJ+UdQZDMVHXlV2UkYIjTEYsjtXBVsYx3dw6LfKexY2LffsXrdh1wAUfEPP7+SKRb6XbyRVZ8iV50VH5d911tv662T7ETXkmm2ZDykjMdnHrZ5sHj0mPx2l3lXqVuFMrs9erEwLdUSx+C4g8pJHygKeM9FUoN6dPaRyTlD5iiqmuY65JrmmlNERdRSkPyGhiMsV0arHxktuwun5CmiGcgjLkukMtCS6RfsTWy6XQrsuujF4Uurm1Lp/jZXXvwLy2Zda3RfVNb630q63+1dcvm00+4BE8GnHn3V+Z19Aa4xhcaRdty068pLsmlLUFsXXF7i81kuHJe0PbRQVQQgAdkbsSVHuz21c2DZNb6BN11uvntp4RBhgE1v8AYia200y6+qCy0lS40iFMkI9Wxf4UY7KWYmAqRvEnm4XLfdVX1+xN9/JRNOW+otlIjEPbfenfEU3mRkrap7zRPu/XH+S6HF6/yJ7JBqsShCpOhIuuyx2YEN5wjMlVVBV3Xbh5aFtf10IeW+kBNdvZdcdLpU0nrrp5d/Acmr3zL8bqHiUPIaSY+LLjaAgEKtNE+YNthsK6RrdPVWiTW32kui3QeOuOyIm7CCjqGQjuq+auLsPFNba280TTh9mEDQCh65ajqhPt8kMuRbrrZV1x16JpoFdcAE0+YuPumOlIU8t4tRazFTsVzOOKIi5OsSdx2CSJHiu3Vg4iAwaiplyNUBNKKa46INOBrbQrsqaUUQl00qJ5lqOyLybIbNWDhKimFG04RCrpUYqLaipUUXtnxfkVYjzEdO16ARbiQAIESochSd2QEUE5aRzb0XuD6KquCu3nzTW6r+iNul6AzFeVf4LMd0fItY3IKXQVD5H+KqIqKipnHStgI8y0oQJUbdUWztm4znYsYpoukLTfaNVFzSiK6UV/RVbdUVVGkZdVRJGlRU/Mg7I06ugTch3TZNcR0LSvOg03qS74iQZ77fY0TYtSTU0MfXdpl93+FHCnuHERQqfuxdkIK4xFxhBlC0/chQYxAJxHnnbmkggg1tPOymdJRAUpEyRLIVdcQNba8tKaJojQkVEUHOQoul2XRDvoh0qa3RRTQISr5aTuJ/qNyJDX5Hws7RrdG7Ebq1BUUbFy2nufxJhS5TnJFmG5IX1N0yQdlXjttoWjLz37YJ+qCK+gC2q+jYsu/sRh/dNjdbeaHfvx21MgU2r+Mw1RQXW2cYhOV+PU8R38dU3RUXWd9LLGqmP2lBFu8JyOhisS51av2bqioujFQLZfsR0h9NJJcRd0UZp/5tOyIznBDiI7XL5FX9usXjsQMVylsc0WKYF38S3DxltARTE8TAhVK4Z+PMr8tHKyCNs2ESh+81mHbVlSya+IlNbYrGcaCJTu+ZFyU2SRDA00b6eiacc38tJ56BUQyRdKu2jdRNG+v6aVwl0hlptziWy6Qt9b6VEXyVPDEf8ADVxpxpUQ2kXiIgKCD3tqqoqoum9z8kVWj/vBx+9tGHJUCpkxJMRw25LAuJxJUINyXddI0eyrwQkJQb5PsuIqcBbYfXZdmmTQRRSBktCwW3kKwVdTircKtN0hQRzeTGiw6+nbfoZPjKSok7/yDjbbzZtuh1WwiLQyYtnUxOX9/sZAJXGOZryEybcBftFDcfFoEmwJUARN1zuJotwXiQCQqSeZESLtrfXLS/MKFpU9N9JwT/MhCm22uYqpbL5qqLyLkukDQhtp75djTRHunkpbrrjpB0ippRRU2024vouhLW+ttI/KbbNtuS6ji7KGmYE1wBcHQ1svbbXwqRw59xyofICFHrm1uslrq6mp42TU8ijgR25yhy3UtkN3yLnzcL1NxFNETdqzktIgvsjdRx/NAS/ip6RVyJpE+WKmTup+WM5kdqabtiVvdPb8rRiG2BK738SYcjYxQMuJ/IzoMSziPQ5rGU9F3wV2Vjz8TpPlsuqdmpEdAm3HGnAefffUFec2+wk8tQHIgSXDkSptybsUWZk3Da2Idk44urWt3hWKS6aPWkJPrLj+ZCCrrbSDrbZFVNQYVZ3as3pjkZlZckGSh1kHx4tmmWVtUjMd+tZhRi8C4ZQsgqYcSurbGMxx0unPNFTQeiiulTW6IvnpmDNk7eHgv1lnG3V+vQlT1Qk3+YdNnpF0i/YQKioYKCmBmbcruZEyPLw539pwAXYKT7x4tmoQxsmbQJLkp99X1A35IgpLyNVTSJpQQk12F33QuBDrYV9QRsP2o3/bXhuX6N15GqawfExtLmEwYCKCKCKfymY9JLmPKlz6dI9JayLWNUjEa6S5w9NSKsHI+jUKBizj1cbfy7tnqFKZhq73K52JRSeRA6lajLoOx7D4bbOn/gHFmX9S+6BSkBERETTMVyQfBoZER6KfB0CBeKqhffRoK2PDSrXJZQu9xuBBy/GkiRpBS8mzt2eDMWEmPZMCP9i5eyLI41t2Yde+q6VdEqJquq51vKFiDGidDL1zwCybLHemWJY6AKEAAaaFBbAkAxUTS+6bYhfoZvVmRdFL+tVx6nfkxZMGQbMmOJaFdb6RdKKFpk5EUuUaS3e3Te+0wri4JFTxziG+XN99BFNKqfbvtrfW+vLSKmkIU/VJDIepBPIPNtnpHYTXcqQHT/FUhT1Un2R9XDs4Lf5pDuR1LX5pbuaUbXrKd6h0bfo671OqQ/KjnVWGn5GHOrCf5IzvU9TeB/wBdVbFfRleqVsvoGTuMWsg57MNTJswQiBKZV87pKiK4CEzZzK4oqES6R5f1EpQgochWvaajrHWVNjRJ0ZiPEB8HI6kjzTU8Wx49hZrThC2kTtdzzSsKI8XpAfjK3/GZF1ttNgaWQq/orhlqNXPyFQiSiymXjsBuDXgvUa+91eol97y9Qr/AN9eoF/9QufX/wBSueX/ANTaX8i6DhYsHTx1d5Ab69h91sdJI/uCSR/Z4of2+KH9viv9vif9vif9viv9niS/RtH3f2dx9fRVN39TAJTiEraDJNT4LrY/38f/ACjf/gK6Y9xVo6vH4qvg5PuMQsMSoG1GAxHvoUhE4m3Jac9CRUX8FdOtvEnylLgWTqL235+OZI+q9uZIwrKTVd5LuBZKvqR4DkXtlgeQfTrgt/8ASrhF/wDSLhV99H9y776P7mX30f3Nvvo1w69+icwSzc/PWL0+m/rVJil0KIiQX8RuXmyBYS4FcfTLgNoSh3Iw41bAKCMD7vW/0U/FLiU2CJDXCr9PSBDw67bktPOQ/g957Pwe69iZjltLaMSi/cu9Vf8Aswwq08lcYZxSWzsqQPgdp9H8DtPo/gdp9H8Dtfo/gVr9H8Ctfo/gNr9F8Btvo/gFt9H8Atvon8KnvmR+ELA7n/IyuCZD+kX7iZH9F9xMl+h+4eTfQJgOTfQp0+yVfWIzgdkAAJ15dPL9TLhGTp1kPsp05v8A2k6cXi+rUTAruLy2jPdPrWSSE5Xh01tPom+mlqv/ALJrpjbey10usi25Gx0qd/1H4XTaPHVFJ2JjEaMibE1CbaRNtIKJ+Psmtk/txH+3Ef28R/bwH9vAP28A/bwD9vbD9nbb/Z2m/b7LXt9ln2+wz7Xh2fa8Ox7XhmPZ8NH9nwsf2fCx/Z8JG9nwkb2PCRvY8JG9jwkb2fCRvZ8JG9jwkb2PCRvY8JG9jwkb2fCxvZ8LH9nwsf2fDR/Z8Mx7Ph2Pa8Ox7XYZ9rsM+12Gfb7LXt9lr2+017fab/Z22/2dtv8AZwD9nAP28B/bxH+2yf2/+lX/AP/EAC0RAAEDAQYEBAcAAAAAAAAAAAEAAhESAxAhQEFREyAwMSJQUpFCYGFxgIGQ/9oACAEDAQE/AP4gT0KmzE+SPMNUEmAEK7OJBjLErx7j26jnVGUwUi7TJku+ED9rHCeeQpHKRUITWanLktHcqtu6kKQi4BVnRBu6JxNwq0QnW+YQIIyUI8pY06LhjQotIUO2TWx97qGoAC6Re4TCAgR1XGkTCNqT2wVb/Wpd6j7lFNe5sQULXcIPaexuIBQ4g6BeF4nICMkWNPcLhNXDZ6UbIzgRCFkNSnWYwpQs4gk3SBqqhve50GApcUKljuoGZc4NTSTpfQ4nEprIM3wNvlKeWSpKxQzsKOSFGQq+h9rigVgsN/JYG34ff//EADoRAAIBAgQCBwQHCQEAAAAAAAECEQADBBIhMRNBBRAwMlFhoRQgQHEiQlBSYpGSI0NTcoCBkLHB0f/aAAgBAgEBPwD/AAfwYmKKkbiOw9mv5M/CaPjAzAb0ZJkmaHvYO1xLuYiQutNdW0C7MFA5mnOExwbIy8TxiDTAqzKw1Bg/CqsneB40PZhoUuN55gvpBrSTAMTpNHXsVUsQAJJqzaFm0FnXdjWKvca5p3F0Hn51zBO/IisxZiSST4n4NFs73LjfJRNHLmMAxyn3t9BQt3D9Q1wrg+oaMjf3LV1rLh1iaxOMDKUt8xq1DqA1+Et27tyQiEx4V7JiP4fqKOHvDdK4dz7hpMNcbeBS4S2gLOdtydqfEx9G0AB4xVsEohPNRUDwq4+HiHZavGx+6zf8rWgdaVGdgqiSauW3ttlcQa5fAAE7CiaERPuDQ6GlxN9drpPz1/3S9I3NmRT8tKt4q3dOWCD50b1lO9cUVir/ABTlU/QHqeoYu8FChgIEbU1x37zk9XDuBc2UgeJqPHqw91LJZmBJiBV1zdcuR2ti0b9wIHVZ5mrXRiKf2rFvIaUuEwy7Wl/uJoWLI2sp+kUFjugD0q7hLF3MWQSeY0NP0awnh3B8jVzD3rWrIY8R1Wb7WZAVSp3BFXTgHAPdJGy0csnKSR5+7FBSxgAk+VJhHOrsFHrS+y2OYJ8dzV+8LpAAMDsD2VvFYi13brR4HUetP0nfKjKiA8zvXt2LJnjH8hFL0pb4YL22z8wNqfpK6e4ir6mrHSDqzG8SyxpAEzV3pAurKluJEST1C2791Ca9mvQSUgRMnrw2HS4hd9p0rg4ZN1UfOrxwjASdfw1NkbIzfzH/AMrjuBCwo8hRdm3Yn4ixh3vk5YAG5NX7NuzCi5mfnGw6hEidqGMsIoVUaB5VfxfFXKoIHPrFy4FChyAOQPxsmIkx9qATWQ+IogLuwFEdSrmrIvnWVPCjlHIUd+2EQSaEZSZ61EmKBkgUTBjsAxFZ/MUxBIMiRtQmAD1AxWf8NZ/wiiT2/BbkyfrWtRIJq2oMs3dXfX0p1AhlnKfQ+FCRQRpkIfyogg6gg9WnMfYM1xH++351P9Hv/8QAShAAAgECBAIGBgcFBQQLAAAAAQIDABEEEiExUWETIkFxgZEFECAyQlIUIzBicqGxQFBTksEzVIKi0QY0Q0QVJGBkg6CjsMLT4f/aAAgBAQAKPwD/ANjXRBxYgUKfEdJiBAsaaG5Uvf8AKoYcLPMkKlXLOGkNh/2CSCEaXbcngB2mseyfOI1oxSvoI5InDV0yo2VrIy2P+ICklkha0oXXIeBr6FC0z9FFgcLGswivpmme5zEUe871GL+ktM+39k1aNjVmI5RAn7NMRjgLspNo4r/OawmAils0MBgzO4/BY0uIi/veC372iNYUXNskrdE9+56jmjOzxsGHmP3jhIH+R5VDVhZkUEsUlVrAURg4CUw6chu/e1OkZYhCwsGpo5Ua6uNwaCQMpVugiSK3dl2rFQl/fMEpjLd+QrUs6fw8YonQ+La0+Ck/j4cmaHxX31rDTxS4yQq6MHU2SurhsLK/mn2QbEPdMNH8z/6CpJRG/TPsxkmc31U7iiSzFDGWKlvuBH7PnYGvdXNdFKghdC4RtQvYgFKzagyIbOh7Rn5Ega1jVcYd+nwE6K6Yjg0Lrbrr2rQZGF1YG4I/d/1ouuIxSfB9xKZmveWU6kngP6mkOInyEYuVn6WOMalUXYBu1t6EkgFkS17t2aVZFyjrOpIJ1Ggo247itKhvpZlQKR5USK7xQvmuOF+VRwdLhXVHlIVc5K0rowurKbgg9oI+wWOGFC8jtsFWmXDx3jwsW+ROPeazyOc7qqq5DsL27GDWrOr9Vo4xdSFPuASbIPiaoMPhEf8A3zEZjFJKNCYk4R7KAailnWyKk8LQppsoe7Hn1r3NTYLFwMHgcjLlK7NGy3BAuNiRQXD+mkmRk2EPpHD6uF4LMozfuwtIN0TUjvJsBQkPBsTGpqPB6ZZZEbM+u6g1kW1wQLseNKX+ci7U2FxEKCJXPXjkS5IBFwQwqXEYlEujOuRBftVL3PeaOZ2LNYaZu00CDXSJ2cRXWJ2HZRPeKWO/ukEhvyr6TBa4Yb2NWI7K+kYS93w0huOZWhFirdbDSkB/Dj7f1ERvi3B0dxslMcLhyGkIu12Gy6eZrpVUXtYTAKON7Oq8T205hcB8S4e4XDjc2OqDsUUcNhsKgEro1gq24imOKXXI4ZFYjndhQxGGc26TLoeaf6rUyW/2iixsiMHbJ8+r/ukYrH2uY72SP8ZrAzwlrEI7RGpsPjMRmBEuUmECsUzMdS8hAPgKedQ4z6ldOQBqOSK5KTh1ysnYSdxRlkJIJQdUEnW52AoKyyAGOIZzbm50HhSxq5BuWJf+dqZSLB8wIa4Ftb0vX3e2o7qVcQ0KHEZQFHSdpA9RrtqSCSJWXOgDBlJvqptrRyIoUZjqbDc1km7GWjkvo4HVNMrA3DAm45gihjoPnLWmUUqzdsE31cg8D7Cy4xgVedTdIe7i9MzuS0spBIVb3LMausOkjlM5z73fLZw3BbaUJDmyjM30giRiCBmSzGQ9i7Cg/pr0lrISbuGIuFLDUhKC43GJ087ynVE3C1inAazMqg2qLFYRzYq6hgDwdW2NYXDysLZ4kCmgJzGZQnaUvbMP3Nh1YGxXpBcUk+PlT6tRtGD8bUz4iZmLyu3Hck0QYIRJ0SrmzBSAddNhQAbRVCirdpJNqyq6krLJ1VPcNzTS6NnD3RNfuUHwsGKjw+ckKqSzKWUBe5aSDAJ/a4/EnocKluDHV25Lc0J8SN/SmIS8hP8A3eI3WIczdqdmdizEklmJ3LGrsPy9QD2Byk2NjUxINmQqVZeZDWqONGsFeSRVFz2MNxUAaMAyp0mtuKcuZtSzZtYuhkV8w53IsaIbmpsCOwk0HW+o3FdDJ8m6GnaL+IlpFoAg7AWIqcoNo5SJR5NWE7+hqURHdEtGniEqZMPHrPiApZEU7nQVBMGkzSO4MUt+w3BPhWg6sbZulIuNRFKmoPFmp4oMFGxwayAJmbLqwJAJRePbTn0bgn+pRvjfdRRDAZ5bVfW5ue2mkwsthi4eI4j7y0kkEyB4pF2ZTX+8YTGJm56MB/l/cqsEus+I7O5a6LCdjiJUXTsXjTGYuekckamrZxemd0YD6sZiCeNtqw8M6jeGMM/9FU1lcxlSzHO9zcafCKJ6Pq5mN2Pbv6sPild+mMU8SSI0vYSHvqtB0i/skAskY4KK72NG+W4LaKe9qfP2oiZ0OvzVEGJBaKSYv5ZL02XsSNAhU/dcm/jandwTlaads3+ULesCGPYcPe/MZzWAAXUL9GUVgXe9w6q0LDmLVjcKXF2D2xSeRIsKVolF7YN+jcniYstCcE2EUaFJQRxVzXRuNGjZgrqflYbg1hpjxZAG8xU8V+yPENb871jToN8T/wDlQFuxpbykfzk19HnjXqvFdQBwKCyml6UqhfJcKxfQWU7UmMxD5bYaMCOEG2bI6pq9qeD0ZG/1sxXKMnA/0WsmHw0bMxJ96wuzHma/tJmy8hewHq7xRPoyZywb+7SH4vwntpZYMLKkPSJqpEmhPm37iASGNmtexYjZRzNPGjNaKIOVijXko3r0hLDi+khaMTkIpTcWoZUsBdQesaZmJ7Bm/SsP0gAC9K+YMtu1FvcjnU6iHEtJ08YBLnb3G7KstwRFOrJr25tGrOFYOAkqy5g24IBB0qMBWXPnUxtZ9soOhoEuWzZHDi67E99Sg2BtlNHkWIAp5OsBaJC4U8GqKI390npsw/CNqJBX/jG6eCjSjkA0X3FHgNLVK7ubRxRxEtITsAoBqL0LAwAWKRekxL98VwEFJOMNIqiVLKHVlDKWU/oKGa+q65vFW0J5CjmJsVBEb35q11Y0pe/uqpV7/hfquaOe+iaI9+SP7x5g1mYE2Wxzg8g2/eDSz5D1S4JdSPlfRwRSliTrOy50P3GVax08RsBNJF9GitxzylQa9G4N8tzDBmxs/iEsK9Ohn06a0MCKeaEEihjvRmPiMuBxii2cJo6MB8S0yMrBxlF727DUS4iZ2dXYjKiprcstNjZCbukRIBPNjUUEC7Ii2q0mIKwIfxamrj2CHVlZTtYg3vQ+vgVm5Psw/bgABck0+MmjuGXDJmXNwzmwqHAQ/O1pZaxGJlLe87lhc/KK1A1r6ThpmGdL2sw2dXHusONTxRBzkQKpa/32BN6hiLuR0VjCpYcSwufOmEA0C4ZRKxPIR3t3moItLgzy9Iw+6Uj1Br0aARc2h6Rh4ML2GxNYMxi3XGH6o7yFug76LRNsA6vGwPb8QtyNQSEWZmRWga69pKdW9Ytevmt05IB76iXrZrysZBc9ovYUyIbXCgIPyoAEiwLBCSeIOhY8KIYnRG+rcnlnujmhGQvSTzyXiECW997B0elm9IyDJJiSPr5eOTsRaxUGTCTz/Rchut0DJdWon69bAEMQuQa5X7OYrRRqAeqBzjbrIOY1oFANQql0VeatrGKzK1gFuZYnHct2jWsSYWGhxKBMM4+7I5BRawGEit1oSWxbjkGGQAVj8fZAfrZUwyAcmUA16L+kWAzCNsTMpH35LilnDm4fGSFwO5EsKkiB0EGDTIPJaOCwZXM2KxT5mtXS4D0HDLEZ/hmmksGZaupNyARfz1tUeMxDKscHSu6xxgakkJqawGGlmWR8NisGmWzDVgasw31vrXVihMzcM8h9RigLXlddWCDe3OmjgBsmc3YjifUqgAk35C9qW8cI6ePNd1kPvFv21pMURcYaAZ3A4tWKwsEqsry9KodgazAaBdhSWIy5SoKkcCKihdV+HUm/C50rUL1r661mym4vxpVNiAU6tr69mlMq22YBhm+bspLuxzvC2Qg75wN9TU6iPR1nUMLdjDOCbGsPOFIBMTmM27GXNevSPSJYrLmJULzMds/iKwrgE9RWBzX+IouU38KMFiOs65V8HFrHmwtStmuVLtvzWVNPFqIJBKrZYJO9S3UI5nWhn3KL9TLa+5ifqEfeGpqK8jWEYjNgvNZNQPvW1pGlxJAXI4eAKBq0kT+7GOVXmawUxpb6TMT1n7gdFWvpPpmW0kcEvXTCd/GSmhwvpnBCL6QdAHydE3kyivRKLKqjEQ4kkFJk0dgLVgIFGqxwIZ2UjtV5CtY30nLEwOefEZAx5rCEuKwUErbNBAodjzIBaseIiNJp16CLvzS2rA4JWNzFAWxcv+Wy1j/SGI/gPMEdv/Chu1YOF/728CzSjmWe4WsH6OS/SyqrrqWNyKGMxFrNj8YpWPvCnVqnmiJuMOp6OFeQRasK3op1Rq90vfg6305UiLg4mjgVrGSSZ/w+9YUXb+LN1F8FrNNMvSE2y7nqgcrAVvXZ69HQqe40QJIZYmBPv2W4/Zwqjck2FYJLbgzLenxDgbQof1a1RB2DiRZ5bZCDptuDSBO18PhmkVF7yGoYqdiwkxbuXAX5QgsFqCaYqFMiqEbTjktUEsStosrMj69mZdxSxgN7g7AdRatzcHW/qA7SSbAAakk0vRSglA7aAe7utr3Ip7hgWeF9ARycbeNBGuCBOpUDldMwrpsOq5ulhImGgub9He1qIzIdD2qajurEhwAG13F+FPGFF41kHTWa97a20NQspPVsTmRhwVOw1iYAerJFKfq0bmjhtKjkjBJllwwMIvxOmQ+VSzYdjYhYTicrcJItTm4EaVjFiYZlMwAw/lOUZTWCwCspBAnbEyjxXL5E2rEY/EyQv9Ix5iVOhgjGZxdaLRQzGD0TC+oDJvJzyVemeLPniljIWaCT54zUZw0TExLicK4dP5M1TuPkweDyeTOwrFekZB8eOxTEfyRZK9HejYUF2lhgjgCjnIdan9L4nhC31XjK1D0RgZBvFeLqc5D16GOxwbPJBAS/g1N6HwJFpOh6k8vJmHuinkfcs7Fz5sTWpNRS4hgC8Cr7nI1a1agXvRuOGlyasLaXq7D5q0QBB3IMo/T1dvr1FCLLP0t+PRDNb9kCqBcsTYCjO4NvqxpcczUUrleqHmCa1inzKqiKDGFYgBwSpbH58WxHH5hWG1UaSYgv46sa9DwtxAUuCO5WpFjvp0Qd7W5C1YiPEBiXlgKRZx97c+Jp8dH2xTR3ZhxEiAVPhLzs7B4s41WwU316vKo5bu6jo3AIvsxU2NbpcKwysSDY6GjoxW/MVlkQ3BqaBbNmCsXUljfMVJ1NJOouC8T9CLjjEDlN/CsLJGLbRWZebBLMP0qfByWBVkcsmv3wQawnpCIG4nQBZdd7SoASNO29TYNg9+jxYzgg9nSr/VRRSRTYjceDbEcx6o2JUrdlDaHvpoJlW/1d3Eh++HzaUuARsMVxMoDPGNrAItemMYGFsscKwrfiGc1JGJyyRYqbEdNIHtcXFBekSOJu55ADXU+gK/8AjdizesmocLcXWNjmlb8Ma3Y1JM5NhicWP0QV9Gw+fRJnCx/4F0Wv+lMT7pmkBCA8s2rUYIb6QYW0Sfh4tV2J6qr1iWPAPuedA2Ow9VnOx9fyjXmfWLGaMG+1s1Kbm+vn7WWaCTOhIuPEcDQR8Rh0kZRsCw/YRLi2H1eHU68i3AVIcQzEnDFyI14ZBwqeGFEAyQnNGSO0otRG+wfqHX8QFQEW3DKfzVqgPdKP0N9ahUfeXL3aggVYLpm6PPt2BlBNStpooyuDfmNRUkuXS8zk+GpWo47j3gv/AMiP61t23sPPWsyMNn2/zXH51AG3DKOjbzW16ljvmVY8VZ010sGaxpIgzIkMvvQ51/zC9JDPBOY3jN2z5tVsx5eqxvcUy4lHLRsJLX4ix0tXo7C4TEXcMxZZl8F0asbiGZtWwyLh45PxnUGvRzzupQNNPJipWJFuoUMmU0jDDgxKbXYAEkDN2j1630YHVahiw6SZJJJ2tmy7AAU7ltAIUVQeQLE1i58Rh2Do0mdrNxGy1kSeHVpJFDAjrAhaGDlwkzrh55Pc1NzG5r0Yke4c4uOx/Op/Sc/YmGXJH4u9D0Thit7oCJSp4yPrQx2My5pMNGTPKW5hTWGwMCC0csn1s4XiUNglS4qa9yZGLle5GtkHIVcHQX2sOKnUDkKGUDYC404Zr2FFjHHlQl9Vd9FsBwqxH5+zp0kX6N6xZS7ai/uIWpD4i/f3+3/yafsAbESHo8Mh7XNNLLMxZyxuTer7kcRpRYfJKA9hb71KF7XS4tflUmJLi6/Rk6YkeFmrDYRLXJxEt2A5oA9PyGHgKfmzVOMQb5fpMalGI7GMYUrRiniBDK5uLHtUsCCDxo25D/67/pSIGNg+47rpYg94p2K7lBlsAfmFgfEVNkOpcfV6HXgVatGsRl6gYPZh1TddeFAdmQWUtyy6qfDSpYIZYQQssLx3A0DBWuCBxFSB8VMJ3RuuixqCsYuQMrdth6nZVtmCC51NqUqZC0SyEbX/AEFvGsaIREsSDCYUxplX7z0kdv73is5Hcq5qSb0niYy0eFw0djDFsZGLUW5nSsRhsRHGWhcOMsjfKFCmukG4IGpHMVlLX04CjK8WHAebEygSO27M6WuFr0fggqgssMJdu+z2KrWKtuwitHbvVQGUVjWlD5izO8yNyYHZT94VifR2NtdsXhEEieBXUA1j8QnbGkDZ/wAlBFHEt/ecYb2P4BdqaCAg/wDV8MVgi8bAMavrex62vG7a3oECx6xLkHjrTEeVC/Kifypev9cxA1JfRb9y+rQevto6PCfzYevaGY/+maO59v8A5NPt7KqlieAFZMNh0PQrxDn3j672objeniuAskOYlXHaGtUeC9KBMr4adwM/OJ296o40AuWd1VR4k1hpn/g4S+Jc/wAmlTzFAVgxkz5JYg3BIr37iayWFzkKq45ELrTFmbVrlNOd7msrSWCiIWcheJ18TREuJmj6vSAwbe8S2oA3JrFemcbDGqPkZxEzLwC2XLWDwULJkgIyK7vzCA3Ap8Vi5WZ4pTmKIQLdYaCwvoKuzMWdjuWO5pieA1JpcBhGIGJxMp0jj3Nl7WPCsfj5xELSrgxnk/xyBKmcJ7hxWIETd5jAuFr0NgMMgJOJRfpQIX4Q1YmTFO1i0umiiwAHYBWlaCjdevbstQAX3ysRlHiqamkMStYkD6VEGHM2kVv0pAmfTppPpEQN9lkFmz8joKOZbMBNpl5pOLhm4X0rOxcAh2+uT72eO6yeNOR8L2WN78SRcMauw7TcnX8r1ZeGwHgKG/YKJ17aAon1aswXzo2voDrYDQDyrs09n/gg+TD13LRygAbk5DtTQpe5kxP1S5TyOprvTDxgDwZ6xUv3ppypNzwULSjkZH/1ojuc0aBrbAwfmg+3sVwMoHey2q9o0A7rezrprWnugDYDjrTMALWZtDzoABbCwuR4ms9je7UB3C1EljT46dwJC1x1E7Hd9kWvR2AuueVInBfzkqT036RRT0DzK7w5x8WuhUU0ks5BcEKgS/BRsvACsyhrsXN71JEUiVmu5KSaDP1TexB4UUKjMGzDIeIbuoNPJqgLBZAD/DSTqSX3JqNlvd1WEzKDxaHe/wB5dKURh7K5dmjzcFmtnVuCnQVL1hmnxFlztk2VnUAEA7VoorQVpQnQvmUglQo4UqsAFd1bKQ51axGo4UOkUBQwBLWp1DbhLRg94S1C/GmNAeq9va91D5nq/wBa7fa/tY3j1+8LeoYXCgjNLINTyVaz4i13xUxvJ3LpZaNtbAbUdu80dTff2XxHo64UodXiHFKWSGVA6ONiD9sV6fDsoI4jUVpKtxqDYqcpH2YcyEBVJ3b+gFPh45znmaMmAS8mmjuSoGwNCRkKnpHGaR2OigTLcMbagHYVIMSL5Ff3Tb5gNrdlasbu3E+pTKRljzglQedQvNIBip1AuNyVW3P3iKwqxS7xkM8TEcnay1EpXQFUXpVHJrEhO41ITlaGIpdRKSNWFgNFFM4LZrnT12HGgRGDITfVuwC1+01dt7kbk739RPsySFVLHKuwG5NYfDxsLh5pANO4VCqgi/RoTpyvWOk5pGgr0mvBmC1Mh7OlgrDvI9i0E4MDALz1Bp2QD3o7SKb8Ctajcdo9l5nD3GUUkkpfqRZwQpbXXuFHLfQA1r9jdsK4ki/A/wBuYpI8ZKejA6uViSD9nZrELoNBwF6kc8CxFZliuWYbPI3vH+nq09S9FGjSycQqam3AnYUc0xLug6oUnYLbYAUGI0a/xjgQKBsbrcZiOQv2cqjtCoWQAKAZTq5sOelan1g6G4tc1YysZW1+EaID+fsvIT8ovUWHBFwrHM/LQV9Jk060ui+AFCNH3CDLpta4p7cMxtV/ZlUWt1GtpwqCax98xgWuNBcVPEbXbK9wP5r06nX3olPdexoqPlEAzWrE41xc2llCoLcQKTCxEfBoaNz9nlgmYwTcMr/bomNw8Zljmy6sEFyjGtDvyNZnYgKOJP2G5tR2tQIQFyCLg22HnV7nX2Mr4olnNz/YodB/iYeu6qS7DkgzUbk9b2LM7AeeldW4VPwqLCtanYH4ihVfzqGCMgGyDpHNS4uQfHPYp4LoKGHjAsFjAWiTxP2SjYkHao2UmwcdXbt61qcg6raxvxtbsp0yA9pGo+Jsw/SiyjU2Zc4sdQACb99A+6UZsw8DcLYVbc2y5du+jcamxFCxO99q7vaNXqJe8gGs7NhI8zcSBYn7a4NOXzmSXB7qU7clWZbMOK0mTFaTQCwaCce+LfI26n12GU2PO2nqvTEDewvTZRfW3aaIrchf61Yga+vrOwUeOlAqoWJLfJGMq+sAhAq6jdj/AKCh51M/4Y2b9BWMKnY9EwBtzNRQKxsGmlRawiF5DEDGc5F93qXGBQAmZzEGPctYaDQZzlztfkWrqD3Rcga91q2GgAygDgAPs+sPUBVtLUyXGuXS9MLHSw5Wq+pPu2pjm3FE11jXb6iaNClFIP8ADTDutUjKe1nNImX0iitlQAnOp/0rrx4SMN32/YHxmEnb63CqLzRX+TitSJBIitnWzBC3wvbY+xoQGU8QfYv361EwBvqtKOrYlHZTWKTUXtID+orGW4hUr0h0mUgWKAXPctYphsSzDTwFTv8Aie1RSfiY/nXo2JQQSWjMrkj8VYaLKNo8NGL1iQ3FWy940rEs1jmvISCDrRJI3JN6RbHTjqPY7Bb2tD7AY8L2byNOt/mUirsat36VqKFDzrFT4dL3nVLR6cGawp4pFOqOK8Kuaaw302rQkCrD1CjRvV++ibWFgKDYo4tJ5kXXowgIUNzJar9Lg4XPeUH7CrowsysLgg9hBposJPcTKuqJL6wj3+pkOwJ3U8jTI6mzIwsQfYGY9pIAFROhPvICQO/QUPOmU8DpRHsag0BQoC1AihoPY29vUexKqMtmQOcpB5VqKBU6ghrilFANcClBIIo4aDD4dUeGDaQIAoJ4LSkuyJhsnuIsa5nVFuTcs2rGt9qa9t70T3mtQb0mI4G+V6kHiKfzFHxek82NRpzEd/1qVA2hAkyfklGWc3I0soJ7ddSas6YCEH+X9iSaCVcro2xFdIv91lNmHJWoRSpKyjCS9SR1HxCrMpKsDxG4ouUQIpbUhRsL+wsJ6CQLnj6RHZtApFxa/GoIyITCEaF0JBHLQ1FjjHh80SpIgKvf37ORWMOILO2FnLA9GSbqtwbFaxFrKsQjHxP8TclrUj2IXjkUnECV7FTwGWm6EYkJG24KM29+VRyQmNhnIBCNfQm9waw6OsoRo8OgUsjDc20uKzS9I2sikqq9guBapIZGyrikETpA2bZ0L+xsfXPL+CNm/SsTEPvxMv6j1aj2bMNjUkErG5I1Rj94UuJQ6Bo1v+lWVdhlIA8BTAns6M/1p8PluyksI9RrpU+Lny6vMxYITuBVz9gwo+q8YbPKfuJVgBYD9lGLgklZxCukqBqdMXNKIljkUoQx43qOJM1mnZ1KVPiPSkCiR2J0lA95VWusOw1h8SGFlMg66X3ykgjXmKxWEbsjeMSxjuKEEeVYcshupDlD5OFoyEm4WKZT+StWJhYoyPHIDfUW+LY1sKuasatao0kjjs7m7B34qBpUQIbMu9LDjZEJlzhjkeo50vnlkaMgEjYLQhwa3dAkTEM57Gy1NJhVbPIZUtd12C8APYeaVtAqisLEJVY4nQsYeCr8xpcXiO3EYq0jUqgbBQAKVgdwdRSQTn/j4a0TUnpCEahNEmqSCZTZ45FKkeB9qWI/cYilJ4tEhorfcoiqT42qSVuLsW+yFX7taueLGwolHwkoKqLKNj9uopB40nnS1eiaPqiMqAhXKgsO40KWlhmckyiPQMT8VqIXMM3aQO21Pyvh2WlNxcF4msagkAt1lv299fnVgWteg7ysOneE6Ki/Arc6jwzQACIMxysNiGY31NWs5W4NwSvCsO+lgZIrsPFSKwuZtgqSXPd1q/lWYfq1MP8AA/8AU0VsNiApo0B6sqcTSoiksW+Jmbcn2TRqHEcC6gsO406pfVN/I1cKxAvRo01NRo0aNfnS+dIPCh5UWyi5AtenBvYg0aPqjRSNS29Nl7VRCfzJq0rizzPq7fa2q1GmPjTGiaNGmpqamp6eiab86cADhUnlUnlUpVTcrbenAA0AFSeVSAqeFSVKvRm4tvesR5tWI82qUsSDcgk6UafwU1ITxIqTyqTyqTyqTyqTyqTyqTyqTyqTyqTyqZWbewp/FPUfYtTuwGrEUQvZcGv8po/yU/8AJUjE8RRzD4hoaalpBQHqNH9jFChQoUKFChS+VL5UvlS+VL5UvlSeVJ5UnlSeVJ5UnlSeVJ5UnlSeVJ5UnlSeVJ5UnlSeVJ5UnlSeVJ5UvlS+VL5UvlS+VL5UvlS+VL5UKFChQof+St//2Q==\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 100,
              "unconfined": true
            }
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {
            "tags": [],
            "image/png": {
              "width": 100,
              "unconfined": true
            }
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "cd06UqOkpYMz"
      },
      "source": [
        "> Warning: Analogy test failed."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LuoewrktmxTX"
      },
      "source": [
        "_Finally, we add the vectors to our product dictionary_"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ubvBCRYub5EY",
        "outputId": "b7775c32-5055-478c-95b7-33ae09da8f39"
      },
      "source": [
        "# add vector to products\n",
        "for sku, p in products.items():\n",
        "    p['vector'] = product_embeddings[p['sku']].tolist() if p['sku'] in product_embeddings else None\n",
        "    p['popularity'] = random.randint(0, 1000)  # add a popularity field to fake popularity data for later retrieval\n",
        "# remove products without vectors for simplicity\n",
        "products = {k: v for k,v in products.items() if v['vector'] is not None}\n",
        "len(products)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "10"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 241
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "igNBEsAPmxTZ"
      },
      "source": [
        "### Personalizing search"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b0mXwT4vfHrj"
      },
      "source": [
        "def re_rank_results(session_vector, skus):\n",
        "    results_vectors = retrieve_vectors_from_redis(skus)\n",
        "    distance_matrix = cosine_similarity(session_vector.reshape(1, -1), results_vectors)[0]\n",
        "    so = np.argsort(distance_matrix)\n",
        "    return list(reversed(list(np.array(skus)[so])))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-aCOJjyLqbm9"
      },
      "source": [
        "> Tip: dense_vector data type comes with premium version of elasticsearch (line #23 in the below cell). So we used text field instead."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "W3oLmO5WfL2D"
      },
      "source": [
        "def upload_docs_to_es(index_name, docs):\n",
        "    \"\"\"\n",
        "    index_name is a string \n",
        "    docs is a map doc id -> doc as a Python dictionary (in our case SKU -> product)\n",
        "    \"\"\"\n",
        "    # first we delete an index with the same name if any \n",
        "    # ATTENTION: IF YOU USE THIS CODE IN THE REAL WORLD THIS LINE WILL DELETE THE INDEX\n",
        "    if es_client.indices.exists(index_name):\n",
        "        print(\"Deleting {}\".format(index_name))\n",
        "        es_client.indices.delete(index=index_name)    \n",
        "    # next we define our index\n",
        "    body = {\n",
        "        'settings': {\n",
        "            \"number_of_shards\" : 1,\n",
        "            \"number_of_replicas\" : 0\n",
        "        },\n",
        "        \"mappings\": {\n",
        "          \"properties\": {\n",
        "                \"name\": { \"type\": \"text\"},\n",
        "                \"target\": { \"type\": \"text\" },\n",
        "                \"image\": { \"type\": \"text\" } ,\n",
        "                \"vector\": {\n",
        "                      # \"type\": \"dense_vector\",\n",
        "                      \"type\": \"text\",\n",
        "                      \"dims\": EMBEDDING_DIMS\n",
        "                    }\n",
        "            }\n",
        "        }\n",
        "    }\n",
        "    # create index\n",
        "    res = es_client.indices.create(index=index_name, body=body)\n",
        "    # finally, we bulk upload the documents\n",
        "    actions = [{\n",
        "                   \"_index\": index_name,\n",
        "                   \"_id\": sku,\n",
        "                   \"_source\": doc\n",
        "               } for sku, doc in docs.items()\n",
        "            ]\n",
        "    # bulk upload\n",
        "    res = helpers.bulk(es_client, actions)\n",
        "    \n",
        "    return res\n",
        "\n",
        "def query_with_es(index_name, search_query, n=5):\n",
        "    search_query = {\n",
        "        \"from\": 0,\n",
        "        \"size\": n,\n",
        "        \"query\" : {\n",
        "            \"script_score\" : {\n",
        "                \"query\": {\n",
        "                        \"match\" : {\n",
        "                            \"name\" : {\n",
        "                                \"query\" : search_query\n",
        "                            }\n",
        "                        }\n",
        "                    },\n",
        "                \"script\": {\n",
        "                  \"source\" : \"doc['popularity'].value / 10\"\n",
        "                }\n",
        "            }\n",
        "         }\n",
        "    }\n",
        "    res = es_client.search(index=index_name, body=search_query)\n",
        "    print(\"Total hits: {}, returned {}\\n\".format(res['hits']['total']['value'], len(res['hits']['hits'])))\n",
        "    return [(hit[\"_source\"]['sku'], hit[\"_source\"]['image']) for hit in res['hits']['hits']]\n",
        "\n",
        "def query_and_display_results_with_es(index_name, search_query, n=5):\n",
        "    res = query_with_es(index_name, search_query, n=n)\n",
        "    return display_image(res)\n",
        "\n",
        "def display_image(skus, n=5):\n",
        "    for (s, image) in skus[:n]:\n",
        "        print('{} - {}\\n'.format(s, image))\n",
        "        display(Image(image, width=150, unconfined=True))\n",
        "            \n",
        "def query_and_rerank_and_display_results_with_es(index_name, search_query, n, session_vector):\n",
        "    res = query_with_es(index_name, search_query, n=n)\n",
        "    skus = [r[0] for r in res]\n",
        "    re_ranked_sku = re_rank_results(session_vector, skus)\n",
        "\n",
        "    return display_image([(sku, res[skus.index(sku)][1]) for sku in re_ranked_sku])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FYaklITifT9t",
        "outputId": "3837c5c7-f3ed-4dd5-f918-ccfbe81a491e"
      },
      "source": [
        "upload_result = upload_docs_to_es(INDEX_NAME, products)\n",
        "upload_result"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(10, [])"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 263
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "mscKXiQYfWjm",
        "outputId": "69089427-3f5a-4fdf-a67a-a89ebf9fcc35"
      },
      "source": [
        "es_client.indices.refresh(INDEX_NAME)\n",
        "resp = es_client.get(index=INDEX_NAME, id=PRODUCT1)\n",
        "print(resp)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "{'_index': 'catalog', '_type': '_doc', '_id': '01112', '_version': 1, '_seq_no': 1, '_primary_term': 1, 'found': True, '_source': {'sku': '01112', 'name': 'shoesB', 'target': 'women', 'image': 'https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/b98cdc08-0691-48a8-99f7-bd4856ab5628/air-force-1-07-shoe-AKTdww3y.png', 'vector': [-0.09506063908338547, 0.01327930111438036, 0.02451828122138977, 0.09916432201862335, 0.07408668100833893], 'popularity': 270}}\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wRAlr04amxTk"
      },
      "source": [
        "### Load data into Redis\n",
        "\n",
        "> Note: To simulate a real-time use case"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "OmxQeSuelVVV"
      },
      "source": [
        "def redis_upload(redis_client, rows):\n",
        "    with redis_client.pipeline() as pipe:\n",
        "        for r in rows:\n",
        "            pipe.hset(REDIS_HASH_FORMAT, r['sku'], json.dumps(r))\n",
        "        res = pipe.execute()\n",
        "    \n",
        "    return\n",
        "\n",
        "def load_vectors_to_cache(products, batch_size):\n",
        "    # first we flush the cache\n",
        "    # ATTENTION: IF YOU USE THIS CODE IN THE REAL WORLD THIS LINE WILL DELETE ALL DATA\n",
        "    redis_client.flushall()\n",
        "    # upload data in bulk with pipeline\n",
        "    rows = list(products.values())\n",
        "    for i in range(0, len(rows), batch_size):\n",
        "        print(\"Uploading {} rows {} at {}...\".format(len(rows), i, datetime.utcnow()))\n",
        "        redis_upload(redis_client, rows[i: i + batch_size])\n",
        "    # do some test\n",
        "    print(redis_client.hmget(REDIS_HASH_FORMAT, [r['sku'] for r in rows[:1]]))\n",
        "    #return total number of rows uploaded\n",
        "    return len(rows)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zDgw57CBljAv",
        "outputId": "72e4d549-1c8a-411e-830a-0b430b3ce25c"
      },
      "source": [
        "load_vectors_to_cache(products, batch_size=5)"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Uploading 10 rows 0 at 2021-07-18 07:48:46.480836...\n",
            "Uploading 10 rows 5 at 2021-07-18 07:48:46.481941...\n",
            "[b'{\"sku\": \"01111\", \"name\": \"shoesA\", \"target\": \"men\", \"image\": \"https://static.nike.com/a/images/c_limit,w_592,f_auto/t_product_v1/e706a692-be9e-4d5a-9537-3d64bd2d0e34/blazer-mid-77-shoe-9SvVxB.png\", \"vector\": [-0.027305342257022858, -0.020312489941716194, -0.0527806393802166, 0.04385489597916603, 0.09405443072319031], \"popularity\": 330}']\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "10"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 266
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "D9w9dkpgmOGW"
      },
      "source": [
        "QUERY1 = '\"match_all\":{}' # put here the first query to test\n",
        "QUERY2 = '' # put here the second query to test\n",
        "TOP_N = 5 # top N results to re-rank"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Czd1DyAKmxTM"
      },
      "source": [
        "def get_products_from_catalogue(catalog_file):\n",
        "    \"\"\"\n",
        "    parse catalogue file into a map SKU -> properties (sku, name, target, image url)\n",
        "    \"\"\"\n",
        "    products = {}\n",
        "    with open(catalog_file) as csvfile:\n",
        "        reader = csv.DictReader(csvfile)\n",
        "        for row in reader:\n",
        "            if row['sku'] and row['image'].endswith('.jpg'):\n",
        "                products[row['sku']] = row\n",
        "    \n",
        "    return products"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EBb4mDdrmxTN"
      },
      "source": [
        "products = get_products_from_catalogue(CATALOGUE_FILE)\n",
        "print('{} products in catalog!'.format(len(products)))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FphkKUqvmxTm"
      },
      "source": [
        "### Query ES"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5SSQH5JFmxTm"
      },
      "source": [
        "_First, we query ES for a \"vanilla\" search_"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-f0JhV__mxTn"
      },
      "source": [
        "query_and_display_results_with_es(INDEX_NAME, QUERY1, TOP_N)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "l64aRiasmxTo"
      },
      "source": [
        "_Now, we retrieve from Redis the vectors for products in the session_"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "CylOQKEdmxTo"
      },
      "source": [
        "def build_average_vector(vectors, v_shape):\n",
        "    \"\"\"\n",
        "    not exactly fancy, but... \n",
        "    see for example https://districtdatalabs.silvrback.com/modern-methods-for-sentiment-analysis\n",
        "    \"\"\"\n",
        "    category_vec = np.zeros(v_shape[0]).reshape(v_shape)\n",
        "    for v in vectors:\n",
        "        assert v.shape == category_vec.shape\n",
        "        category_vec += v\n",
        "    \n",
        "    return category_vec / len(vectors)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GjD-XTiimxTp"
      },
      "source": [
        "def retrieve_vectors_from_redis(skus):\n",
        "    session_products = redis_client.hmget(REDIS_HASH_FORMAT, skus)\n",
        "    return [np.array(json.loads(s)[\"vector\"]) for s in session_products if s]\n",
        "\n",
        "def retrieve_session_vector_from_redis(products_in_session):\n",
        "    session_vectors = retrieve_vectors_from_redis(products_in_session)\n",
        "    \n",
        "    return build_average_vector(session_vectors, session_vectors[0].shape)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "81meOQBBmxTq"
      },
      "source": [
        "session_vector = retrieve_session_vector_from_redis(PRODUCTS_IN_SESSION)\n",
        "# debug\n",
        "print(len(session_vector), session_vector[:10])"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "753ON5cXmxTq"
      },
      "source": [
        "_Finally use the session vector to query_"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8d6LX8NAmxTr"
      },
      "source": [
        "query_and_rerank_and_display_results_with_es(INDEX_NAME, QUERY1, TOP_N, session_vector)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JiYYrqLLmxTr"
      },
      "source": [
        "# uncomment here if you like to use Coveo index instead\n",
        "# query_and_rerank_and_display_results_with_coveo(QUERY1, TOP_N, session_vector)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ujyjCHSbmxTs"
      },
      "source": [
        "_Try some other query_"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mow2FVHJmxTt"
      },
      "source": [
        "# vanilla query\n",
        "query_and_display_results_with_es(INDEX_NAME, QUERY2, TOP_N)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "pXK_Rio-mxTu"
      },
      "source": [
        "# now personalized\n",
        "query_and_rerank_and_display_results_with_es(INDEX_NAME, QUERY2, TOP_N, session_vector)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1tmClLFcmxTv"
      },
      "source": [
        "### Appendix 1: word embeddings, where it all started"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8cpFXxIomxTv"
      },
      "source": [
        "def get_sentences_from_corpus(corpus_file, max_sentences=None):\n",
        "    \"\"\"\n",
        "        Read the text file and process it as a list of lists, where each list is \n",
        "        the tokens in a sentence. Don't care too much about pre-processing,\n",
        "        just get stuff done.\n",
        "    \"\"\"\n",
        "    sentences = []\n",
        "    with open(corpus_file) as c_file:\n",
        "        for line in c_file:\n",
        "            # remove punctuation, strip lines, lower case it and normalize spaces\n",
        "            cleaned_line = ' '.join(line.translate(str.maketrans('', '', string.punctuation)).strip().lower().split())\n",
        "            if not cleaned_line:\n",
        "                continue\n",
        "            sentences.append(cleaned_line.split())\n",
        "            # check if we reached a max number of sentences for training\n",
        "            if max_sentences and len(sentences) == max_sentences:\n",
        "                return sentences\n",
        "    \n",
        "    return sentences"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XMiZs625mxTw"
      },
      "source": [
        "# texts from 1BN words dataset - if you're using the docker-compose setup, change the path to tmp folder\n",
        "TEXT_FILE = os.path.join(DATA_FOLDER, 'corpus.txt')  \n",
        "MAX_SENTENCES = 1000000 # how many sentences to load for the word2vec example: keep it small if you don't like to wait!"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SuEfzKsMmxTx"
      },
      "source": [
        "training_sentences_data = get_sentences_from_corpus(TEXT_FILE, max_sentences=MAX_SENTENCES)\n",
        "print('Total sentences: {}, first is: {}'.format(len(training_sentences_data), training_sentences_data[0]))\n",
        "word_embeddings = train_embedding_model(training_sentences_data)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QDBAFjoymxTx"
      },
      "source": [
        "_Now test similarities and play with analogies_"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vhvXuhFfmxTy"
      },
      "source": [
        "for _ in ['paris', 'france']:\n",
        "    print('###{}\\n{}\\n'.format(_, word_embeddings.most_similar_cosmul(positive=[_])))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RNNqcsqrmxTy"
      },
      "source": [
        "print(\"BOY : KING = WOMAN : {}\\n\".format(solve_vector_analogy(word_embeddings, 'boy', 'king', 'girl')[0][0]))\n",
        "print(\"PARIS : FRANCE = BERLIN : {}\\n\".format(solve_vector_analogy(word_embeddings, 'paris', 'france', 'berlin')[0][0]))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "uNab76iymxTz"
      },
      "source": [
        "### Appendix 2: how to visualize vectors and impress friends"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hreWrrhWmxTz"
      },
      "source": [
        "def visualize_word_embeddings_tsne(word_embeddings):\n",
        "    # colors\n",
        "    colors = ['red', 'green', 'blue', 'purple', 'yellow', 'black']\n",
        "    interesting_word_groups = [\n",
        "        (['he', 'she', 'it', 'they', 'i', 'you', 'we'], 'pronouns'),\n",
        "        (['london', 'paris', 'berlin', 'budapest', 'amsterdam', 'prague', 'rome'], 'cities'),\n",
        "        (['italy', 'germany', 'spain', 'romania', 'finland', 'poland', 'norway', 'sweden', 'austria', 'brazil'], 'countries'),\n",
        "        (['pasta', 'pizza', 'steak', 'pie', 'fries', 'burger', 'salmon'], 'food'),\n",
        "        (['john', 'mark', 'jane', 'jessica', 'donald', 'simon'], 'names'),\n",
        "        ([random.choice(list(word_embeddings.vocab)) for _ in range(0, 100)], 'other')\n",
        "    ]\n",
        "    all_words = []\n",
        "    for words, group in interesting_word_groups:\n",
        "        for w in words:\n",
        "            all_words.append(w)\n",
        "    all_keys = [w for w in list(word_embeddings.vocab) if w in all_words]\n",
        "    all_vectors = [word_embeddings[e] for e in all_keys]\n",
        "    # get projection\n",
        "    X_embedded = TSNE(n_components=2).fit_transform(all_vectors)\n",
        "    word_2_emb = {k: e for k, e in zip(all_keys, X_embedded)}\n",
        "    # divide groups\n",
        "    data = []\n",
        "    groups = []\n",
        "    for words, group in interesting_word_groups:\n",
        "        groups.append(group)\n",
        "        data.append([word_2_emb[w] for w in words])\n",
        "    print(groups, data[0])\n",
        "    # create plot\n",
        "    fig = plt.figure(figsize=(8, 6))\n",
        "    ax = fig.add_subplot(1, 1, 1)\n",
        "    # add groups\n",
        "    for d, color, group in zip(data, colors, groups):\n",
        "        x = [_[0] for _ in d]\n",
        "        y = [_[1] for _ in d]\n",
        "        ax.scatter(x, y, alpha=0.8, c=color, edgecolors='none', s=30, label=group)\n",
        "    # show plot\n",
        "    plt.title('Plot color-coded embeddings')\n",
        "    plt.legend(loc=2)\n",
        "    plt.show()\n",
        "    \n",
        "    return\n",
        "\n",
        "visualize_word_embeddings_tsne(word_embeddings)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HJii5BWcmxT0"
      },
      "source": [
        "def export_vectors_for_projector_visualization(product_2_vectors,\n",
        "                                               product_2_label,\n",
        "                                               target_folder):\n",
        "    # map dictionary to list to preserve order when exporting\n",
        "    all_p = [p for p in list(product_2_vectors.vocab) if (not product_2_label or p in product_2_label)]\n",
        "    all_v = [product_2_vectors[p] for p in all_p]\n",
        "    # write vectors\n",
        "    with open(os.path.join(target_folder, 'vectors.tsv'), 'w') as v_f:\n",
        "        for v in all_v:\n",
        "            v_f.write('{}\\n'.format('\\t'.join(['{:.5f}'.format(_) for _ in v])))\n",
        "    # if avalaible, labels can be paired with SKUs for visualization purposes\n",
        "    # if a mapping is specified, we produce a \"meta\" file, otherwise we just return\n",
        "    if not product_2_label:\n",
        "        return\n",
        "    # write meta if mapping is available\n",
        "    with open(os.path.join(target_folder, 'meta.tsv', 'w')) as m_f:\n",
        "        # header\n",
        "        m_f.write('sku\\tlabel\\n')\n",
        "        for sku in all_p:\n",
        "            m_f.write('{}\\t{}\\n'.format(sku, product_2_label[sku]))\n",
        "\n",
        "    return"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}