h&m-comp-best.ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.7.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nimport seaborn as sns # data visualization\nfrom matplotlib import pyplot as plt\nimport os\n# import cudf","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2022-05-10T02:09:09.769623Z","iopub.execute_input":"2022-05-10T02:09:09.772673Z","iopub.status.idle":"2022-05-10T02:09:10.980763Z","shell.execute_reply.started":"2022-05-10T02:09:09.770168Z","shell.execute_reply":"2022-05-10T02:09:10.980064Z"},"trusted":true},"execution_count":1,"outputs":[]},{"cell_type":"markdown","source":"## 1. Load the Data","metadata":{}},{"cell_type":"code","source":"INPUT_PATH = \"/kaggle/input/h-and-m-personalized-fashion-recommendations\"\narticles_df = pd.read_csv(f\"{INPUT_PATH}/articles.csv\")\ncustomers_df = pd.read_csv(f\"{INPUT_PATH}/customers.csv\")\nsubmission = pd.read_csv(f\"{INPUT_PATH}/sample_submission.csv\")","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:09:13.159212Z","iopub.execute_input":"2022-05-10T02:09:13.160181Z","iopub.status.idle":"2022-05-10T02:09:25.861486Z","shell.execute_reply.started":"2022-05-10T02:09:13.160134Z","shell.execute_reply":"2022-05-10T02:09:25.860424Z"},"trusted":true},"execution_count":2,"outputs":[]},{"cell_type":"markdown","source":"**Reduce memory of train transactions data**","metadata":{}},{"cell_type":"code","source":"def read_csv_reduce_memory(csv_path, read_ratio=0.1):\n    nrows = 31788324\n    train = pd.read_csv(csv_path, nrows=int(nrows*read_ratio))\n    train['article_id'] = train['article_id'].astype('int32')\n    train['t_dat'] = pd.to_datetime(train['t_dat'])\n    # train['year'] = (train.t_dat.dt.year-2000).astype('int8')\n    # train['month'] = (train.t_dat.dt.month).astype('int8')\n    # train['day'] = (train.t_dat.dt.day).astype('int8')\n    train['price'] = train['price'].astype('float32')\n    train['sales_channel_id'] = train['sales_channel_id'].astype('int8')\n\n    return train\n\ntrain_data = read_csv_reduce_memory(f\"{INPUT_PATH}/transactions_train.csv\")\ntrain_data = pd.read_csv(f\"{INPUT_PATH}/transactions_train.csv\")","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:09:25.864611Z","iopub.execute_input":"2022-05-10T02:09:25.864929Z","iopub.status.idle":"2022-05-10T02:10:41.388404Z","shell.execute_reply.started":"2022-05-10T02:09:25.864896Z","shell.execute_reply":"2022-05-10T02:10:41.387320Z"},"trusted":true},"execution_count":3,"outputs":[]},{"cell_type":"markdown","source":"## 2a. Articles Dataset\n\n`article_id`: Unique identifier of an article  \n`product_code`, `prod_name`: Identifies the product and its name  \n`product_type_no`, `product_type_name`: id of the product_code's group and its name  \n`graphical_appearance_no`, `graphical_appearance_name`: Graphical apperance id of product_code and its name  \n`colour_group_code`, `colour_group_name`: Color id of product_code and its name  \n`department_no`, `department_name`:  Department id of product_code and its name  \n`index_code`, `index_name`: Index code of product_code and its name  \n`section_no`, `section_name`: Section id of product_code within index_code and its name  \n`garment_group_no`, `garment_group_name`: Garment group id of product_code and its name  \n`detail_desc`: description of article  ","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:01:10.897408Z","iopub.execute_input":"2022-05-10T02:01:10.897685Z","iopub.status.idle":"2022-05-10T02:01:10.911524Z","shell.execute_reply.started":"2022-05-10T02:01:10.897647Z","shell.execute_reply":"2022-05-10T02:01:10.910429Z"}}},{"cell_type":"code","source":"articles_df.head()","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:10:41.390122Z","iopub.execute_input":"2022-05-10T02:10:41.390448Z","iopub.status.idle":"2022-05-10T02:10:41.428878Z","shell.execute_reply.started":"2022-05-10T02:10:41.390405Z","shell.execute_reply":"2022-05-10T02:10:41.428098Z"},"trusted":true},"execution_count":4,"outputs":[{"execution_count":4,"output_type":"execute_result","data":{"text/plain":"   article_id  product_code          prod_name  product_type_no  \\\n0   108775015        108775          Strap top              253   \n1   108775044        108775          Strap top              253   \n2   108775051        108775      Strap top (1)              253   \n3   110065001        110065  OP T-shirt (Idro)              306   \n4   110065002        110065  OP T-shirt (Idro)              306   \n\n  product_type_name  product_group_name  graphical_appearance_no  \\\n0          Vest top  Garment Upper body                  1010016   \n1          Vest top  Garment Upper body                  1010016   \n2          Vest top  Garment Upper body                  1010017   \n3               Bra           Underwear                  1010016   \n4               Bra           Underwear                  1010016   \n\n  graphical_appearance_name  colour_group_code colour_group_name  ...  \\\n0                     Solid                  9             Black  ...   \n1                     Solid                 10             White  ...   \n2                    Stripe                 11         Off White  ...   \n3                     Solid                  9             Black  ...   \n4                     Solid                 10             White  ...   \n\n   department_name index_code        index_name index_group_no  \\\n0     Jersey Basic          A        Ladieswear              1   \n1     Jersey Basic          A        Ladieswear              1   \n2     Jersey Basic          A        Ladieswear              1   \n3   Clean Lingerie          B  Lingeries/Tights              1   \n4   Clean Lingerie          B  Lingeries/Tights              1   \n\n   index_group_name section_no            section_name garment_group_no  \\\n0        Ladieswear         16  Womens Everyday Basics             1002   \n1        Ladieswear         16  Womens Everyday Basics             1002   \n2        Ladieswear         16  Womens Everyday Basics             1002   \n3        Ladieswear         61         Womens Lingerie             1017   \n4        Ladieswear         61         Womens Lingerie             1017   \n\n   garment_group_name                                        detail_desc  \n0        Jersey Basic            Jersey top with narrow shoulder straps.  \n1        Jersey Basic            Jersey top with narrow shoulder straps.  \n2        Jersey Basic            Jersey top with narrow shoulder straps.  \n3   Under-, Nightwear  Microfibre T-shirt bra with underwired, moulde...  \n4   Under-, Nightwear  Microfibre T-shirt bra with underwired, moulde...  \n\n[5 rows x 25 columns]","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>article_id</th>\n      <th>product_code</th>\n      <th>prod_name</th>\n      <th>product_type_no</th>\n      <th>product_type_name</th>\n      <th>product_group_name</th>\n      <th>graphical_appearance_no</th>\n      <th>graphical_appearance_name</th>\n      <th>colour_group_code</th>\n      <th>colour_group_name</th>\n      <th>...</th>\n      <th>department_name</th>\n      <th>index_code</th>\n      <th>index_name</th>\n      <th>index_group_no</th>\n      <th>index_group_name</th>\n      <th>section_no</th>\n      <th>section_name</th>\n      <th>garment_group_no</th>\n      <th>garment_group_name</th>\n      <th>detail_desc</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>108775015</td>\n      <td>108775</td>\n      <td>Strap top</td>\n      <td>253</td>\n      <td>Vest top</td>\n      <td>Garment Upper body</td>\n      <td>1010016</td>\n      <td>Solid</td>\n      <td>9</td>\n      <td>Black</td>\n      <td>...</td>\n      <td>Jersey Basic</td>\n      <td>A</td>\n      <td>Ladieswear</td>\n      <td>1</td>\n      <td>Ladieswear</td>\n      <td>16</td>\n      <td>Womens Everyday Basics</td>\n      <td>1002</td>\n      <td>Jersey Basic</td>\n      <td>Jersey top with narrow shoulder straps.</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>108775044</td>\n      <td>108775</td>\n      <td>Strap top</td>\n      <td>253</td>\n      <td>Vest top</td>\n      <td>Garment Upper body</td>\n      <td>1010016</td>\n      <td>Solid</td>\n      <td>10</td>\n      <td>White</td>\n      <td>...</td>\n      <td>Jersey Basic</td>\n      <td>A</td>\n      <td>Ladieswear</td>\n      <td>1</td>\n      <td>Ladieswear</td>\n      <td>16</td>\n      <td>Womens Everyday Basics</td>\n      <td>1002</td>\n      <td>Jersey Basic</td>\n      <td>Jersey top with narrow shoulder straps.</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>108775051</td>\n      <td>108775</td>\n      <td>Strap top (1)</td>\n      <td>253</td>\n      <td>Vest top</td>\n      <td>Garment Upper body</td>\n      <td>1010017</td>\n      <td>Stripe</td>\n      <td>11</td>\n      <td>Off White</td>\n      <td>...</td>\n      <td>Jersey Basic</td>\n      <td>A</td>\n      <td>Ladieswear</td>\n      <td>1</td>\n      <td>Ladieswear</td>\n      <td>16</td>\n      <td>Womens Everyday Basics</td>\n      <td>1002</td>\n      <td>Jersey Basic</td>\n      <td>Jersey top with narrow shoulder straps.</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>110065001</td>\n      <td>110065</td>\n      <td>OP T-shirt (Idro)</td>\n      <td>306</td>\n      <td>Bra</td>\n      <td>Underwear</td>\n      <td>1010016</td>\n      <td>Solid</td>\n      <td>9</td>\n      <td>Black</td>\n      <td>...</td>\n      <td>Clean Lingerie</td>\n      <td>B</td>\n      <td>Lingeries/Tights</td>\n      <td>1</td>\n      <td>Ladieswear</td>\n      <td>61</td>\n      <td>Womens Lingerie</td>\n      <td>1017</td>\n      <td>Under-, Nightwear</td>\n      <td>Microfibre T-shirt bra with underwired, moulde...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>110065002</td>\n      <td>110065</td>\n      <td>OP T-shirt (Idro)</td>\n      <td>306</td>\n      <td>Bra</td>\n      <td>Underwear</td>\n      <td>1010016</td>\n      <td>Solid</td>\n      <td>10</td>\n      <td>White</td>\n      <td>...</td>\n      <td>Clean Lingerie</td>\n      <td>B</td>\n      <td>Lingeries/Tights</td>\n      <td>1</td>\n      <td>Ladieswear</td>\n      <td>61</td>\n      <td>Womens Lingerie</td>\n      <td>1017</td>\n      <td>Under-, Nightwear</td>\n      <td>Microfibre T-shirt bra with underwired, moulde...</td>\n    </tr>\n  </tbody>\n</table>\n<p>5 rows × 25 columns</p>\n</div>"},"metadata":{}}]},{"cell_type":"markdown","source":"## 2b. Customers Dataset\n\n`customer_id`: Unique id of customer  \n`FN`: Whether customer is subscribed to the newsletter (1.0 if so, otherwise missing)  \n`Active`: Whether customer is active for communication (1.0 if so, otherwise missing)  \n`club_member_status`: Club member status of customer (ACTIVE or PRE-CREATE)  \n`fashion_news_frequency:`:  Frequency of newsletter subscription (Regularly, Monthly, or NONE)  \n`age`: age of customer  \n`postal_code`: postal code of customer  ","metadata":{}},{"cell_type":"code","source":"customers_df.head()","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:10:41.443097Z","iopub.execute_input":"2022-05-10T02:10:41.443565Z","iopub.status.idle":"2022-05-10T02:10:41.459474Z","shell.execute_reply.started":"2022-05-10T02:10:41.443533Z","shell.execute_reply":"2022-05-10T02:10:41.458763Z"},"trusted":true},"execution_count":5,"outputs":[{"execution_count":5,"output_type":"execute_result","data":{"text/plain":"                                         customer_id   FN  Active  \\\n0  00000dbacae5abe5e23885899a1fa44253a17956c6d1c3...  NaN     NaN   \n1  0000423b00ade91418cceaf3b26c6af3dd342b51fd051e...  NaN     NaN   \n2  000058a12d5b43e67d225668fa1f8d618c13dc232df0ca...  NaN     NaN   \n3  00005ca1c9ed5f5146b52ac8639a40ca9d57aeff4d1bd2...  NaN     NaN   \n4  00006413d8573cd20ed7128e53b7b13819fe5cfc2d801f...  1.0     1.0   \n\n  club_member_status fashion_news_frequency   age  \\\n0             ACTIVE                   NONE  49.0   \n1             ACTIVE                   NONE  25.0   \n2             ACTIVE                   NONE  24.0   \n3             ACTIVE                   NONE  54.0   \n4             ACTIVE              Regularly  52.0   \n\n                                         postal_code  \n0  52043ee2162cf5aa7ee79974281641c6f11a68d276429a...  \n1  2973abc54daa8a5f8ccfe9362140c63247c5eee03f1d93...  \n2  64f17e6a330a85798e4998f62d0930d14db8db1c054af6...  \n3  5d36574f52495e81f019b680c843c443bd343d5ca5b1c2...  \n4  25fa5ddee9aac01b35208d01736e57942317d756b32ddd...  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>customer_id</th>\n      <th>FN</th>\n      <th>Active</th>\n      <th>club_member_status</th>\n      <th>fashion_news_frequency</th>\n      <th>age</th>\n      <th>postal_code</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>00000dbacae5abe5e23885899a1fa44253a17956c6d1c3...</td>\n      <td>NaN</td>\n      <td>NaN</td>\n      <td>ACTIVE</td>\n      <td>NONE</td>\n      <td>49.0</td>\n      <td>52043ee2162cf5aa7ee79974281641c6f11a68d276429a...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0000423b00ade91418cceaf3b26c6af3dd342b51fd051e...</td>\n      <td>NaN</td>\n      <td>NaN</td>\n      <td>ACTIVE</td>\n      <td>NONE</td>\n      <td>25.0</td>\n      <td>2973abc54daa8a5f8ccfe9362140c63247c5eee03f1d93...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>000058a12d5b43e67d225668fa1f8d618c13dc232df0ca...</td>\n      <td>NaN</td>\n      <td>NaN</td>\n      <td>ACTIVE</td>\n      <td>NONE</td>\n      <td>24.0</td>\n      <td>64f17e6a330a85798e4998f62d0930d14db8db1c054af6...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>00005ca1c9ed5f5146b52ac8639a40ca9d57aeff4d1bd2...</td>\n      <td>NaN</td>\n      <td>NaN</td>\n      <td>ACTIVE</td>\n      <td>NONE</td>\n      <td>54.0</td>\n      <td>5d36574f52495e81f019b680c843c443bd343d5ca5b1c2...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>00006413d8573cd20ed7128e53b7b13819fe5cfc2d801f...</td>\n      <td>1.0</td>\n      <td>1.0</td>\n      <td>ACTIVE</td>\n      <td>Regularly</td>\n      <td>52.0</td>\n      <td>25fa5ddee9aac01b35208d01736e57942317d756b32ddd...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"markdown","source":"### Age distribution of customers\nMost common age is around 21-23","metadata":{}},{"cell_type":"code","source":"sns.set_style(\"darkgrid\")\nf, ax = plt.subplots(figsize=(10,5))\nax = sns.histplot(data=customers_df, x='age', bins=50, color='green')\nax.set_xlabel('Age of customer')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:10:41.460662Z","iopub.execute_input":"2022-05-10T02:10:41.461926Z","iopub.status.idle":"2022-05-10T02:10:42.042648Z","shell.execute_reply.started":"2022-05-10T02:10:41.461848Z","shell.execute_reply":"2022-05-10T02:10:42.041927Z"},"trusted":true},"execution_count":6,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 720x360 with 1 Axes>","image/png":"iVBORw0KGgoAAAANSUhEUgAAAncAAAE9CAYAAABp4UT1AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAAqYElEQVR4nO3de3BUZZ7/8U/TTUuY3Azb6QZMsYposQhhdhRIkQlr3E6QwJhIss7spQYUUXTFgMuskRGQq1OLSlFUqZnMqtTO7CCXhDIRCQSlkxkYRgQj6o6DTqyEIh2NuRAgdNI5vz/42WUGEhpIupPT71cVVfDk6ed8Tz8c6sNzbhbDMAwBAADAFIaEuwAAAAD0HcIdAACAiRDuAAAATIRwBwAAYCKEOwAAABMh3AEAAJiILdwFDBRdXV3y+wf/U2GsVosp9gO9Y54jA/McGZjnyNGXcz10qLXHnxHu/j+/31Bz87lwl3Hd4uOHm2I/0DvmOTIwz5GBeY4cfTnXDkdMjz/jtCwAAICJEO4AAABMhHAHAABgIoQ7AAAAEyHcAQAAmAjhDgAAwET6LdwVFBQoJSVFs2fPDrT94he/0MyZMzVnzhw9/vjjam1tDfzs1VdfldvtVmZmpiorKwPtHo9HmZmZcrvdKiwsDLTX1tYqLy9Pbrdb+fn58vl8kiSfz6f8/Hy53W7l5eWprq6uv3YRAABgwOm3cHf//ferqKioW9v06dNVWlqqt956S3/7t3+rV199VZJ08uRJlZWVqaysTEVFRXruuefk9/vl9/u1evVqFRUVqaysTKWlpTp58qQkaePGjZo3b5727dun2NhY7dixQ5K0fft2xcbGat++fZo3b542btzYX7sIAAAw4PRbuLvrrrsUFxfXrS01NVU228XnJk+ePFn19fWSpIqKCmVlZclutyspKUljxoxRdXW1qqurNWbMGCUlJclutysrK0sVFRUyDEOHDx9WZmamJCknJ0cVFRWSpAMHDignJ0eSlJmZqUOHDskwePI3AACIDGG75m7nzp1KS0uTJHm9XrlcrsDPnE6nvF5vj+1NTU2KjY0NBEWXyyWv1xsYa+TIkZIkm82mmJgYNTU1hWq3AAAAwiosrx97+eWXZbVa9aMf/Sgcm78sq9Wi+Pjh4S7julmtQ0yxH+gd8xwZmOfIwDxHjlDNdcjD3a5du/Tee+/p9ddfl8VikXRxRe7bU7TSxdU3p9MpSZdtv/HGG9Xa2qrOzk7ZbDbV19cH+judTp0+fVoul0udnZ06c+aMbrzxxivWZfZ3y0bHDVWUfdgVP3/e1662lo7+KA19iHdRRgbmOTIwz5EjVO+WDWm483g8Kioq0v/8z/8oKioq0J6enq6nnnpK8+fPl9frVU1NjSZNmiTDMFRTU6Pa2lo5nU6VlZXphRdekMVi0dSpU7V3715lZWWpuLhY6enpgbGKi4v1/e9/X3v37tW0adMCITKSRdmHKfWXaVfsV/WwR20i3AEAMFj1W7hbunSpjhw5oqamJqWlpemJJ55QYWGhfD6f5s+fL0lKTk7W6tWrNW7cON17772aNWuWrFarVqxYIavVKklasWKFFixYIL/fr7lz52rcuHGSpGXLlmnJkiXatGmTxo8fr7y8PElSbm6uli1bJrfbrbi4OL300kv9tYsAAAADjsXgVlJJUkeH3xTL4j0t+TocMUGv3H311Zn+KA19iNM4kYF5jgzMc+QI1WlZ3lABAABgIoQ7AAAAEyHcAQAAmAjhDgAAwEQIdwAAACZCuAMAADARwh0AAICJEO4AAABMhHAHAABgIoQ7AAAAEyHcAQAAmAjhDgAAwEQIdwAAACZCuAMAADARW7gLwMDi8/vkcMRcsd95X7vaWjpCUBEAALgahDt0Y7falfrLtCv2q3rYozYR7gAAGGg4LQsAAGAihDsAAAATIdwBAACYCOEOAADARAh3AAAAJkK4AwAAMBHCHQAAgIkQ7gAAAEyEcAcAAGAihDsAAAATIdwBAACYCOEOAADARAh3AAAAJkK4AwAAMBHCHQAAgIkQ7gAAAEzEFu4CYG7RcUMVZR92xX7nfe1qa+kIQUUAAJgb4Q79Kso+TKm/TLtiv6qHPWoT4Q4AgOvVb6dlCwoKlJKSotmzZwfampubNX/+fGVkZGj+/PlqaWmRJBmGobVr18rtdmvOnDn6+OOPA58pLi5WRkaGMjIyVFxcHGg/ceKE5syZI7fbrbVr18owjF63AQAAEAn6Ldzdf//9Kioq6tZWWFiolJQUlZeXKyUlRYWFhZIkj8ejmpoalZeXa82aNVq1apWki0Fty5YtevPNN7V9+3Zt2bIlENZWrVqlNWvWqLy8XDU1NfJ4PL1uAwAAIBL0W7i76667FBcX162toqJC2dnZkqTs7Gzt37+/W7vFYtHkyZPV2tqqhoYGVVVVafr06YqPj1dcXJymT5+uyspKNTQ0qK2tTZMnT5bFYlF2drYqKip63QYAAEAkCOndso2NjUpMTJQkORwONTY2SpK8Xq9cLlegn8vlktfrvaTd6XRetv3b/r1tAwAAIBKE7YYKi8Uii8UyYLZhtVoUHz+8X+sJBat1SI/7YbMFl+WD7Rfs99XX46H3eYZ5MM+RgXmOHKGa65CGuxEjRqihoUGJiYlqaGhQQkKCpIsrcvX19YF+9fX1cjqdcjqdOnLkSKDd6/VqypQpPfbvbRtX4vcbam4+1xe7GVbx8cMvux8OR4w6O7uCGiPYfsF8X1ezXTN8/6HS0zzDXJjnyMA8R46+nGuHI6bHn4X0tGx6erpKSkokSSUlJbrnnnu6tRuGoePHjysmJkaJiYlKTU1VVVWVWlpa1NLSoqqqKqWmpioxMVHR0dE6fvy4DMO47Fh/vQ0AAIBI0G8rd0uXLtWRI0fU1NSktLQ0PfHEE1q4cKHy8/O1Y8cOjRo1Sps2bZIkzZgxQwcPHpTb7VZUVJTWr18vSYqPj9djjz2m3NxcSdLjjz+u+Ph4SdLKlStVUFCg9vZ2paWlKS3t4rPUetoGAABAJLAY3z4gLsJ1dPhNsSze22nZYB8mHGy/r746c8V+V7PdYMbDRZzGiQzMc2RgniOHKU/LAgAAoH8R7gAAAEyEcAcAAGAihDsAAAATIdwBAACYCOEOAADARAh3AAAAJkK4AwAAMBHCHQAAgIkQ7gAAAEyEcAcAAGAihDsAAAATIdwBAACYCOEOAADARAh3AAAAJkK4AwAAMBHCHQAAgIkQ7gAAAEyEcAcAAGAihDsAAAATIdwBAACYCOEOAADARAh3AAAAJkK4AwAAMBHCHQAAgIkQ7gAAAEyEcAcAAGAihDsAAAATsYW7AECSfH6fHI6YK/Y772tXW0tHCCoCAGBwItxhQLBb7Ur9ZdoV+1U97FGbCHcAAPSE07IAAAAmQrgDAAAwEcIdAACAiRDuAAAATIRwBwAAYCJhCXevv/66srKyNHv2bC1dulQXLlxQbW2t8vLy5Ha7lZ+fL5/PJ0ny+XzKz8+X2+1WXl6e6urqAuO8+uqrcrvdyszMVGVlZaDd4/EoMzNTbrdbhYWFId8/AACAcAl5uPN6vdq6dat27typ0tJS+f1+lZWVaePGjZo3b5727dun2NhY7dixQ5K0fft2xcbGat++fZo3b542btwoSTp58qTKyspUVlamoqIiPffcc/L7/fL7/Vq9erWKiopUVlam0tJSnTx5MtS7CQAAEBZhWbnz+/1qb29XZ2en2tvb5XA4dPjwYWVmZkqScnJyVFFRIUk6cOCAcnJyJEmZmZk6dOiQDMNQRUWFsrKyZLfblZSUpDFjxqi6ulrV1dUaM2aMkpKSZLfblZWVFRgLAADA7EL+EGOn06kHH3xQd999t2644QZNnz5dEyZMUGxsrGy2i+W4XC55vV5JF1f6Ro4cebFYm00xMTFqamqS1+tVcnJyt3G//YzL5erWXl1dfcW6rFaL4uOH99l+hovVOqTH/bDZgsvywfYL9vsK13bNrLd5hnkwz5GBeY4coZrrkIe7lpYWVVRUqKKiQjExMXryySe7XS8XLn6/oebmc+Eu47rFxw+/7H44HDHq7OwKaoxg+wXzfYVru2bX0zzDXJjnyMA8R46+nOveXtkZ8tOyv//973XTTTcpISFBQ4cOVUZGhj744AO1traqs7NTklRfXy+n0ynp4srb6dOnJUmdnZ06c+aMbrzxRjmdTtXX1wfG9Xq9cjqdPbYDAABEgpCHu1GjRunDDz/U+fPnZRiGDh06pFtvvVVTp07V3r17JUnFxcVKT0+XJKWnp6u4uFiStHfvXk2bNk0Wi0Xp6ekqKyuTz+dTbW2tampqNGnSJE2cOFE1NTWqra2Vz+dTWVlZYCwAAACzC/lp2eTkZGVmZionJ0c2m03jx4/XAw88oH/4h3/QkiVLtGnTJo0fP155eXmSpNzcXC1btkxut1txcXF66aWXJEnjxo3Tvffeq1mzZslqtWrFihWyWq2SpBUrVmjBggXy+/2aO3euxo0bF+rdBAAACIuQhztJWrx4sRYvXtytLSkpKfD4k++64YYbtHnz5suOs2jRIi1atOiS9hkzZmjGjBl9UywAAMAgwhsqAAAATIRwBwAAYCKEOwAAABMh3AEAAJgI4Q4AAMBECHcAAAAmQrgDAAAwEcIdAACAiRDuAAAATIRwBwAAYCKEOwAAABMh3AEAAJgI4Q4AAMBECHcAAAAmQrgDAAAwEcIdAACAiRDuAAAATIRwBwAAYCKEOwAAABMJKtwdPXo0qDYAAACEV1Dhbu3atUG1AQAAILxsvf3w2LFjOnbsmL755hu99tprgfa2tjb5/f5+Lw4AAABXp9dw19HRoXPnzsnv9+vs2bOB9ujoaG3evLnfiwMAAMDV6TXcTZkyRVOmTFFOTo5Gjx4dqpoAAABwjXoNd9/y+Xx69tlnderUKXV2dgbat27d2m+FAQAA4OoFFe6efPJJ/fjHP1ZeXp6GDOHpKQAAAANVUOHOZrPpn//5n/u7FgAAAFynoJbh7r77bv36179WQ0ODmpubA78AAAAwsAS1cldcXCxJ+tWvfhVos1gsqqio6J+qgD4QHTdUUfZhV+x33teutpaOEFQEAED/CyrcHThwoL/rAPpclH2YUn+ZdsV+VQ971CbCHQDAHIIKdyUlJZdtz87O7sNSAAAAcL2CCncfffRR4PcXLlzQoUOHNGHCBMIdAADAABNUuHv22We7/bm1tVVLlizpl4IAAABw7YIKd38tKipKdXV1fV0L/r9gbwSQuBkAAAB0F1S4e/TRRwO/7+rq0ueff6577733mjfa2tqqn//85/rss89ksVi0fv163XzzzVqyZIlOnTql0aNHa9OmTYqLi5NhGFq3bp0OHjyoYcOG6fnnn9eECRMkXbyL9+WXX5YkLVq0SDk5OZKkEydOqKCgQO3t7ZoxY4aWL18ui8VyzfWGWrA3AkjcDAAAALoLKtw9+OCDgd9brVaNHj1aLpfrmje6bt06/fCHP9TmzZvl8/nU3t6uV155RSkpKVq4cKEKCwtVWFioZcuWyePxqKamRuXl5frwww+1atUqbd++Xc3NzdqyZYt27twpi8Wi+++/X+np6YqLi9OqVau0Zs0aJScn6+GHH5bH49GMGTOuuV4AAIDBIqiHGE+ZMkW33HKLzp49q9bWVg0dOvSaN3jmzBn98Y9/VG5uriTJbrcrNjZWFRUVgRs0srOztX//fkkKtFssFk2ePFmtra1qaGhQVVWVpk+frvj4eMXFxWn69OmqrKxUQ0OD2traNHnyZFksFmVnZ/M8PgAAEDGCCndvv/228vLy9M4772jPnj2B31+Luro6JSQkqKCgQNnZ2Vq+fLnOnTunxsZGJSYmSpIcDocaGxslSV6vt9sqocvlktfrvaTd6XRetv3b/gAAAJEgqNOyr7zyinbs2KERI0ZIkr755hvNmzdPM2fOvOoNdnZ26pNPPtGzzz6r5ORkrV27VoWFhd36WCyWkF8jZ7VaFB8/PKTb7I3NFlTulqRudVutQ3rcj2DHDLZfsN9XuLbbX2MOBL3NM8yDeY4MzHPkCNVcBxXuDMMIBDtJio+Pl2EY17RBl8sll8ul5ORkSdLMmTNVWFioESNGqKGhQYmJiWpoaFBCQoKkiyty9fX1gc/X19fL6XTK6XTqyJEjgXav16spU6b02P9K/H5Dzc3nrmmf+prDEaPOzq6g+3+37vj44Zfdj6sZM9h+wXxf4dru1W57oMx9sHqaZ5gL8xwZmOfI0Zdz7XDE9PizoJY1UlNT9dBDD2nXrl3atWuXFi5cqLS04O7mvLQYh1wul7744gtJ0qFDhzR27Filp6cH3oRRUlKie+65R5IC7YZh6Pjx44qJiVFiYqJSU1NVVVWllpYWtbS0qKqqSqmpqUpMTFR0dLSOHz8uwzC6jQUAAGB2va7cffnll/r666/1n//5nyovL9fRo0clSZMnT9aPfvSja97os88+q//4j/9QR0eHkpKStGHDBnV1dSk/P187duzQqFGjtGnTJknSjBkzdPDgQbndbkVFRWn9+vWSLq4ePvbYY4EbMx5//HHFx8dLklauXBl4FEpaWto1B1EAAIDBptdwt379ei1dulSSlJGRoYyMDEnSn/70J61fv16vvPLKNW10/Pjx2rVr1yXtb7zxxiVtFotFK1euvOw4ubm5gXD3XRMnTlRpaek11QYAADCY9Rruvv76a91+++2XtN9+++06depUvxUFAMHgbS4AcKlew92ZM2d6/Fl7e3ufFwMAV4O3uQDApXq9oeKOO+7Qm2++eUn79u3bA68AAwAAwMDR68rdM888o3//93/XW2+9FQhzJ06cUEdHh7Zs2RKSAgEAABC8XsPd3/zN3+i3v/2tDh8+rD//+c+SLt69mpKSEpLiAAAAcHWCeojxtGnTNG3atP6uBdfA5/dd8iDD3h5sCAAAzC2ocIeBy261d7ug3GYbctm3MlQ97OnT7V4uVAIAgPAj3OGa/HWo7Elfh0oAANC74N9ODwAAgAGPlTsAIRPsQ4d54DAAXDvCHYCQCfahwzxwGACuHadlAQAATIRwBwAAYCKEOwAAABMh3AEAAJgI4Q4AAMBECHcAAAAmQrgDAAAwEcIdAACAifAQYwARwef3yeGIuWI/3o4BYLAj3AGICHarnbdjAIgInJYFAAAwEcIdAACAiRDuAAAATIRr7hDxuNAeAGAmhDtEPC60BwCYCadlAQAATISVOwDXLTpuqKLsw8JdBgBAhDsAfSDKPizoU9sAgP7FaVkAAAATIdwBAACYCOEOAADARAh3AAAAJkK4AwAAMJGwhTu/36/s7Gw98sgjkqTa2lrl5eXJ7XYrPz9fPp9PkuTz+ZSfny+32628vDzV1dUFxnj11VfldruVmZmpysrKQLvH41FmZqbcbrcKCwtDu2MAAABhFLZwt3XrVo0dOzbw540bN2revHnat2+fYmNjtWPHDknS9u3bFRsbq3379mnevHnauHGjJOnkyZMqKytTWVmZioqK9Nxzz8nv98vv92v16tUqKipSWVmZSktLdfLkybDsIwAAQKiFJdzV19frvffeU25uriTJMAwdPnxYmZmZkqScnBxVVFRIkg4cOKCcnBxJUmZmpg4dOiTDMFRRUaGsrCzZ7XYlJSVpzJgxqq6uVnV1tcaMGaOkpCTZ7XZlZWUFxgIAADC7sIS79evXa9myZRoy5OLmm5qaFBsbK5vt4jOVXS6XvF6vJMnr9WrkyJGSJJvNppiYGDU1Ncnr9crlcgXGdDqd8nq9PbYDAABEgpC/oeLdd99VQkKC7rjjDv3hD38I9eZ7ZLVaFB8/PNxlBNhswefu7/a1WCw9fjbYMQd6v6uZp3Buuz9ZrUMGTC3fCtd3fa3HSl9su78NxHlG32OeI0eo5jrk4e6DDz7QgQMH5PF4dOHCBbW1tWndunVqbW1VZ2enbDab6uvr5XQ6JV1ceTt9+rRcLpc6Ozt15swZ3XjjjXI6naqvrw+M6/V6A5/pqb03fr+h5uZzfby318bhiFFnZ1fQ/b/b12Yb0uNngx1zoPcLdp6u5nvs6233t/j44QOmFil83/X1HCvXu+1QGGjzjP7BPEeOvpxrhyOmx5+F/LTsU089JY/HowMHDujFF1/UtGnT9MILL2jq1Knau3evJKm4uFjp6emSpPT0dBUXF0uS9u7dq2nTpslisSg9PV1lZWXy+Xyqra1VTU2NJk2apIkTJ6qmpka1tbXy+XwqKysLjIXBz+f3yeGICeoXAACRKOQrdz1ZtmyZlixZok2bNmn8+PHKy8uTJOXm5mrZsmVyu92Ki4vTSy+9JEkaN26c7r33Xs2aNUtWq1UrVqyQ1WqVJK1YsUILFiyQ3+/X3LlzNW7cuLDtF/qW3WoP6gX1Ei+pR/+KjhuqKPuwK/Y772tXW0tHCCoCgIvCGu6mTp2qqVOnSpKSkpICjz/5rhtuuEGbN2++7OcXLVqkRYsWXdI+Y8YMzZgxo2+LBYDviLIPC+o/GlUPe9Qmwh2A0OENFQAAACZCuAMAADARwh0AAICJEO4AAABMZMDcLQsgdLjTEwDMi3AHRCDu9AQA8+K0LAAAgIkQ7gAAAEyEcAcAAGAiXHMH9DFuVgAAhBPhDuhj4bpZIdhQCQAwN8IdYBLBhkrpYrAEAJgT19wBAACYCOEOAADARAh3AAAAJkK4AwAAMBFuqADCxOf3yeGIuWI/HpkCALgahDsgTOxWO+93jQCEeAChRrgDgH5EiAcQaoQ7AANOsKtdAIBLEe4ADDhXs9oFAOiOu2UBAABMhHAHAABgIoQ7AAAAEyHcAQAAmAjhDgAAwEQIdwAAACZCuAMAADARwh0AAICJEO4AAABMhHAHAABgIoQ7AAAAEyHcAQAAmAjhDgAAwERs4S4AAAYSn98nhyMm3GUAwDULebg7ffq0fvazn6mxsVEWi0X/9E//pJ/+9Kdqbm7WkiVLdOrUKY0ePVqbNm1SXFycDMPQunXrdPDgQQ0bNkzPP/+8JkyYIEkqLi7Wyy+/LElatGiRcnJyJEknTpxQQUGB2tvbNWPGDC1fvlwWiyXUuwpgELJb7Ur9ZdoV+1U97AlBNQBw9UJ+WtZqterpp5/W22+/rW3btuk3v/mNTp48qcLCQqWkpKi8vFwpKSkqLCyUJHk8HtXU1Ki8vFxr1qzRqlWrJEnNzc3asmWL3nzzTW3fvl1btmxRS0uLJGnVqlVas2aNysvLVVNTI4+Hf4QBAEBkCHm4S0xMDKy8RUdH65ZbbpHX61VFRYWys7MlSdnZ2dq/f78kBdotFosmT56s1tZWNTQ0qKqqStOnT1d8fLzi4uI0ffp0VVZWqqGhQW1tbZo8ebIsFouys7NVUVER6t0EAAAIi7Bec1dXV6dPP/1UycnJamxsVGJioiTJ4XCosbFRkuT1euVyuQKfcblc8nq9l7Q7nc7Ltn/b/0qsVovi44f31a5dN5st+Nz93b4Wi6XHzwY7pln69ceYwf4d6e/tWq1DLltLf3w34drnwfD3IRzz3GXp1A22G6441oXOCxpicFn1YNDT8QzzCdVch+3IP3v2rBYvXqxnnnlG0dHR3X5msVhCfo2c32+ouflcSLfZE4cjRp2dXUH3/25fm21Ij58Ndkyz9OvrMX1+n+xD7SHfrqRL/m7Gxw+/pO16/t70JJz7PND/PvRHv2DnOdhrAr/66kxQ20V4XW6eYU59Ode93fgVlnDX0dGhxYsXa86cOcrIyJAkjRgxQg0NDUpMTFRDQ4MSEhIkXVyRq6+vD3y2vr5eTqdTTqdTR44cCbR7vV5NmTKlx/7A9YrEC+0jcZ8BYLAL+TV3hmFo+fLluuWWWzR//vxAe3p6ukpKSiRJJSUluueee7q1G4ah48ePKyYmRomJiUpNTVVVVZVaWlrU0tKiqqoqpaamKjExUdHR0Tp+/LgMw+g2FgAAgNmFfOXu6NGj2r17t2677Tbdd999kqSlS5dq4cKFys/P144dOzRq1Cht2rRJkjRjxgwdPHhQbrdbUVFRWr9+vSQpPj5ejz32mHJzcyVJjz/+uOLj4yVJK1euDDwKJS0tTWlpV155AAAAMIOQh7s777xTf/rTny77szfeeOOSNovFopUrV162f25ubiDcfdfEiRNVWlp6fYUCAAAMQrx+DAAAwEQIdwAAACZCuAMAADARnnAJDHA9vciel9sDAC6HcAcMcJd71tzlHlbNs+YAABLhDgBMqacV38s572tXW0tHP1cEIFQIdwBgQsG+XUS6uOrbJsIdYBbcUAEAAGAihDsAAAATIdwBAACYCOEOAADARLihAgAGAJ5nCKCvEO4AYADgeYYA+gqnZQEAAEyEcAcAAGAihDsAAAAT4Zo7AEBQouOGKso+7Ir9eJ0ZEF6EuxAK9h9GABiIouzDgnqlGa8zA8KLcBdCV/MPIwAAwLXgmjsAAAATYeUOANCnenog81/j2jygfxDuAAB96nIPZL4crs0D+genZQEAAEyEcAcAAGAihDsAAAAT4Zo7AIhwwd4AAWBwINwBQIS7mhsgwoE3YwBXh3AHABjQeDMGcHW45g4AAMBEWLkDAJgCD08GLiLcAQBMgYcnAxcR7gAAYcFdukD/INwBAMJioN+lCwxW3FABAABgIqZdufN4PFq3bp26urqUl5enhQsXhrskAMAAcDWngy90XtANthuu2I+bNDCQmDLc+f1+rV69Wq+99pqcTqdyc3OVnp6uW2+9NdylAQDCLNjTwdLFU8LB9D3w4H7u1MWAYcpwV11drTFjxigpKUmSlJWVpYqKCsIdAKBfcKcuBhJThjuv1yuXyxX4s9PpVHV1dRgrAgCg51PCf90W7OlgThvjciyGYRjhLqKvvfPOO6qsrNS6deskSSUlJaqurtaKFSvCXBkAAED/MuXdsk6nU/X19YE/e71eOZ3OMFYEAAAQGqYMdxMnTlRNTY1qa2vl8/lUVlam9PT0cJcFAADQ70x5zZ3NZtOKFSu0YMEC+f1+zZ07V+PGjQt3WQAAAP3OlNfcAQAARCpTnpYFAACIVIQ7AAAAEyHcDVKnT5/Wv/3bv2nWrFnKysrSG2+8IUlqbm7W/PnzlZGRofnz56ulpSXMlaIv+P1+ZWdn65FHHpEk1dbWKi8vT263W/n5+fL5fGGuEH2htbVVixcv1syZM3Xvvffq2LFjHNMm9PrrrysrK0uzZ8/W0qVLdeHCBY5pEygoKFBKSopmz54daOvp+DUMQ2vXrpXb7dacOXP08ccf92kthLtBymq16umnn9bbb7+tbdu26Te/+Y1OnjypwsJCpaSkqLy8XCkpKSosLAx3qegDW7du1dixYwN/3rhxo+bNm6d9+/YpNjZWO3bsCGN16Cvr1q3TD3/4Q73zzjvavXu3xo4dyzFtMl6vV1u3btXOnTtVWloqv9+vsrIyjmkTuP/++1VUVNStrafj1+PxqKamRuXl5VqzZo1WrVrVp7UQ7gapxMRETZgwQZIUHR2tW265RV6vVxUVFcrOzpYkZWdna//+/WGsEn2hvr5e7733nnJzcyVd/B/f4cOHlZmZKUnKyclRRUVFOEtEHzhz5oz++Mc/BubZbrcrNjaWY9qE/H6/2tvb1dnZqfb2djkcDo5pE7jrrrsUFxfXra2n4/fbdovFosmTJ6u1tVUNDQ19VgvhzgTq6ur06aefKjk5WY2NjUpMTJQkORwONTY2hrk6XK/169dr2bJlGjLk4uHa1NSk2NhY2WwXn2Tkcrnk9XrDWSL6QF1dnRISElRQUKDs7GwtX75c586d45g2GafTqQcffFB33323UlNTFR0drQkTJnBMm1RPx+9fvya1r+eccDfInT17VosXL9Yzzzyj6Ojobj+zWCyyWCxhqgx94d1331VCQoLuuOOOcJeCftbZ2alPPvlEP/nJT1RSUqKoqKhLTsFyTA9+LS0tqqioUEVFhSorK3X+/HlVVlaGuyyEQCiPX1M+xDhSdHR0aPHixZozZ44yMjIkSSNGjFBDQ4MSExPV0NCghISEMFeJ6/HBBx/owIED8ng8unDhgtra2rRu3Tq1traqs7NTNptN9fX1vF7PBFwul1wul5KTkyVJM2fOVGFhIce0yfz+97/XTTfdFJjHjIwMffDBBxzTJtXT8fvXr0nt6zln5W6QMgxDy5cv1y233KL58+cH2tPT01VSUiJJKikp0T333BOmCtEXnnrqKXk8Hh04cEAvvviipk2bphdeeEFTp07V3r17JUnFxcW8Xs8EHA6HXC6XvvjiC0nSoUOHNHbsWI5pkxk1apQ+/PBDnT9/XoZh6NChQ7r11ls5pk2qp+P323bDMHT8+HHFxMQETt/2Bd5QMUi9//77+pd/+RfddtttgWuxli5dqkmTJik/P1+nT5/WqFGjtGnTJsXHx4e3WPSJP/zhD/rv//5vvfrqq6qtrdWSJUvU0tKi8ePHa+PGjbLb7eEuEdfp008/1fLly9XR0aGkpCRt2LBBXV1dHNMms3nzZr399tuy2WwaP3681q1bJ6/XyzE9yC1dulRHjhxRU1OTRowYoSeeeEL/+I//eNnj1zAMrV69WpWVlYqKitL69es1ceLEPquFcAcAAGAinJYFAAAwEcIdAACAiRDuAAAATIRwBwAAYCKEOwAAABMh3AEY9Pbv36/bb79dn3/+eb9v65tvvlFeXp6ys7P1/vvv9/n4u3bt4tVTAK4L4Q7AoFdaWqof/OAHKisr6/dtHTp0SLfddptKSkp055139vn4xcXFffoC8SsxDENdXV0h2x6A/sdz7gAMamfPntXMmTO1detWPfroo4Gn/Hd1dWn16tU6fPiwRo4cKZvNprlz52rmzJk6ceKEnn/+eZ07d0433nijNmzYcMnT4evq6vTMM8+oqalJCQkJ2rBhg1paWrRo0SK1t7fL6XRq27ZtGjZsWOAz1dXVWr9+vc6dOye73a7XX39d5eXlOnHihFasWCFJeuSRR/Tggw/qzjvv1PLly3XixAlZLBbNnTtXLpdLBQUFSkxM1LBhw7Rt2zYdO3ZMv/jFL+T3+3XHHXfoueeek91uV3p6urKysuTxeGS1WrVmzRq9+OKL+vLLL/XQQw/pJz/5iSSpqKhIe/bskc/nk9vt1uLFi1VXV6eHHnpIycnJ+vjjj1VYWKjRo0eHaMYA9DsDAAax3bt3GwUFBYZhGMYDDzxgfPTRR4ZhGMaePXuMBQsWGH6/32hoaDDuvPNOY8+ePYbP5zMeeOABo7Gx0TAMwygrKzOefvrpS8Z95JFHjF27dhmGYRjbt283Fi1aZBiGYezcudN47rnnLul/4cIFIz093fjwww8NwzCMM2fOGB0dHZf0X7hwoXH48GHjo48+MubNmxdob2lpMQzDMP71X//VqK6uNgzDMNrb2420tDTjiy++MAzDMJYtW2a89tprhmEYxt133238+te/NgzDMNatW2fMnj3bOHPmjNHY2GikpKQYhmEYlZWVxs9//nOjq6vL8Pv9xsKFC40jR44YtbW1xu23324cO3bsar9uAIMAp2UBDGplZWXKysqSJM2aNStwavbo0aOaOXOmhgwZIofDoalTp0qS/vKXv+izzz7T/Pnzdd999+nll1++7DVux44d0+zZsyVJ9913n44ePdprHX/5y1/kcDg0adIkSVJ0dLRsNluP/ZOSklRbW6s1a9bI4/EoOjr6smPedNNNuvnmmyVJOTk53a7z+/Y9lbfddpuSk5MVHR2thIQE2e12tba26ne/+51+97vfKTs7Wzk5Ofriiy9UU1Mj6eI7TidPntzrPgEYnHr+lwcABrjm5mYdPnxYn332mSwWi/x+vywWi372s5/1+BnDMDRu3Dht27YtJDVardZu17RduHBBkhQXF6fdu3erqqpKv/3tb7Vnzx5t2LDhqsYeOnSoJGnIkCHd3kM6ZMgQdXZ2yjAMLVy4UD/+8Y+7fa6urk7Dhw+/1l0CMMCxcgdg0Nq7d6/uu+8+vfvuuzpw4IAOHjyom266Se+//77+/u//XuXl5erq6tLXX3+tI0eOSJJuvvlmffPNNzp27JgkqaOjQ3/+858vGfv73/9+YBXwrbfeuuLNEzfffLO++uorVVdXS5La2trU2dmp0aNH6//+7//U1dWl06dPB37+zTffyDAMZWZmKj8/X5988okk6Xvf+57Onj0bGPPUqVP68ssvJUm7d+/WXXfdFfT3k5qaqp07dwbG83q9amxsDPrzAAYnVu4ADFqlpaV6+OGHu7VlZGSotLRUK1eu1KFDhzRr1iyNHDlSf/d3f6eYmBjZ7XZt3rxZa9eu1ZkzZ+T3+/XTn/5U48aN6zbOs88+q4KCAv3qV78K3FDRG7vdrpdeeklr165Ve3u7hg0bptdee00/+MEPNHr0aM2aNUtjx47VhAkTJEkNDQ0qKCgIrOotXbpU0sVTrytXrgzcULFhwwY9+eSTgRsqvr1RIhipqan6/PPPAyt3w4cP13/9139pyBD+Xw+YGXfLAjCts2fP6nvf+56ampqUl5en//3f/5XD4Qh3WQDQr1i5A2Bajz76qFpbW9XR0aHHHnuMYAcgIrByBwAAYCJceAEAAGAihDsAAAATIdwBAACYCOEOAADARAh3AAAAJkK4AwAAMJH/B2RCWB4zig42AAAAAElFTkSuQmCC\n"},"metadata":{}}]},{"cell_type":"markdown","source":"### Customers' club status\nLarge majority are active members, with some beginning to active it. A very small part of customers left the club.","metadata":{}},{"cell_type":"code","source":"sns.set_style(\"darkgrid\")\nf, ax = plt.subplots(figsize=(10,5))\nax = sns.histplot(data=customers_df, x='club_member_status', color='green')\nax.set_xlabel('Club member status')\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:10:42.044117Z","iopub.execute_input":"2022-05-10T02:10:42.044619Z","iopub.status.idle":"2022-05-10T02:10:44.147159Z","shell.execute_reply.started":"2022-05-10T02:10:42.044576Z","shell.execute_reply":"2022-05-10T02:10:44.146287Z"},"trusted":true},"execution_count":7,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 720x360 with 1 Axes>","image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","source":"### Fashion newsletter engagement\nAbout 2/3 of customers do not receive fashion news","metadata":{}},{"cell_type":"code","source":"f, ax = plt.subplots(figsize=(10,5))\ncolors = sns.color_palette('pastel')\npie_data = customers_df[['customer_id', 'fashion_news_frequency']].groupby('fashion_news_frequency').count()\n\nplt.pie(pie_data.customer_id, labels=pie_data.index, colors=colors)\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:10:44.148780Z","iopub.execute_input":"2022-05-10T02:10:44.149117Z","iopub.status.idle":"2022-05-10T02:10:44.694230Z","shell.execute_reply.started":"2022-05-10T02:10:44.149072Z","shell.execute_reply":"2022-05-10T02:10:44.693278Z"},"trusted":true},"execution_count":8,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 720x360 with 1 Axes>","image/png":"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\n"},"metadata":{}}]},{"cell_type":"markdown","source":"## 2c. Transactions\n\n`t_dat`: transaction date  \n`customer_id`: unique customer id (customers_df)  \n`article_id`: unique article id (articles_df)  \n`price`: price of purchase  \n`sales_channel_id`: 1 - online, 2 - in-store  ","metadata":{}},{"cell_type":"code","source":"train_data.head()","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:10:44.695628Z","iopub.execute_input":"2022-05-10T02:10:44.696777Z","iopub.status.idle":"2022-05-10T02:10:44.724055Z","shell.execute_reply.started":"2022-05-10T02:10:44.696714Z","shell.execute_reply":"2022-05-10T02:10:44.723050Z"},"trusted":true},"execution_count":9,"outputs":[{"execution_count":9,"output_type":"execute_result","data":{"text/plain":"        t_dat                                        customer_id  article_id  \\\n0  2018-09-20  000058a12d5b43e67d225668fa1f8d618c13dc232df0ca...   663713001   \n1  2018-09-20  000058a12d5b43e67d225668fa1f8d618c13dc232df0ca...   541518023   \n2  2018-09-20  00007d2de826758b65a93dd24ce629ed66842531df6699...   505221004   \n3  2018-09-20  00007d2de826758b65a93dd24ce629ed66842531df6699...   685687003   \n4  2018-09-20  00007d2de826758b65a93dd24ce629ed66842531df6699...   685687004   \n\n      price  sales_channel_id  \n0  0.050831                 2  \n1  0.030492                 2  \n2  0.015237                 2  \n3  0.016932                 2  \n4  0.016932                 2  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>t_dat</th>\n      <th>customer_id</th>\n      <th>article_id</th>\n      <th>price</th>\n      <th>sales_channel_id</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>2018-09-20</td>\n      <td>000058a12d5b43e67d225668fa1f8d618c13dc232df0ca...</td>\n      <td>663713001</td>\n      <td>0.050831</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>2018-09-20</td>\n      <td>000058a12d5b43e67d225668fa1f8d618c13dc232df0ca...</td>\n      <td>541518023</td>\n      <td>0.030492</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>2018-09-20</td>\n      <td>00007d2de826758b65a93dd24ce629ed66842531df6699...</td>\n      <td>505221004</td>\n      <td>0.015237</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>2018-09-20</td>\n      <td>00007d2de826758b65a93dd24ce629ed66842531df6699...</td>\n      <td>685687003</td>\n      <td>0.016932</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>2018-09-20</td>\n      <td>00007d2de826758b65a93dd24ce629ed66842531df6699...</td>\n      <td>685687004</td>\n      <td>0.016932</td>\n      <td>2</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"markdown","source":"## 3. Recommendations","metadata":{}},{"cell_type":"markdown","source":"### Customer's Purchases within Last Week","metadata":{}},{"cell_type":"code","source":"# find each customer's most recent day of purchases\ntmp = train_data.groupby('customer_id').t_dat.max().reset_index()\ntmp.columns = ['customer_id','max_dat']\ntmp['max_dat'] = pd.to_datetime(tmp['max_dat'])\ntmp.head()","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:10:44.730070Z","iopub.execute_input":"2022-05-10T02:10:44.732911Z","iopub.status.idle":"2022-05-10T02:13:03.846090Z","shell.execute_reply.started":"2022-05-10T02:10:44.732844Z","shell.execute_reply":"2022-05-10T02:13:03.845337Z"},"trusted":true},"execution_count":10,"outputs":[{"execution_count":10,"output_type":"execute_result","data":{"text/plain":"                                         customer_id    max_dat\n0  00000dbacae5abe5e23885899a1fa44253a17956c6d1c3... 2020-09-05\n1  0000423b00ade91418cceaf3b26c6af3dd342b51fd051e... 2020-07-08\n2  000058a12d5b43e67d225668fa1f8d618c13dc232df0ca... 2020-09-15\n3  00005ca1c9ed5f5146b52ac8639a40ca9d57aeff4d1bd2... 2019-06-09\n4  00006413d8573cd20ed7128e53b7b13819fe5cfc2d801f... 2020-08-12","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>customer_id</th>\n      <th>max_dat</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>00000dbacae5abe5e23885899a1fa44253a17956c6d1c3...</td>\n      <td>2020-09-05</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0000423b00ade91418cceaf3b26c6af3dd342b51fd051e...</td>\n      <td>2020-07-08</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>000058a12d5b43e67d225668fa1f8d618c13dc232df0ca...</td>\n      <td>2020-09-15</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>00005ca1c9ed5f5146b52ac8639a40ca9d57aeff4d1bd2...</td>\n      <td>2019-06-09</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>00006413d8573cd20ed7128e53b7b13819fe5cfc2d801f...</td>\n      <td>2020-08-12</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"# merge tables to link customer_id, article_id, max purchase date, and today\ntrain_data = train_data.merge(tmp,on=['customer_id'],how='left', suffixes=('', '_'))\ntrain_data['t_dat'] = pd.to_datetime(train_data['t_dat'])","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:13:03.847932Z","iopub.execute_input":"2022-05-10T02:13:03.848841Z","iopub.status.idle":"2022-05-10T02:13:27.609620Z","shell.execute_reply.started":"2022-05-10T02:13:03.848797Z","shell.execute_reply":"2022-05-10T02:13:27.608542Z"},"trusted":true},"execution_count":11,"outputs":[]},{"cell_type":"code","source":"# find all purchases made within the past week of customer's last purchase\ntrain_data['diff_dat'] = (train_data['max_dat'] - train_data['t_dat']).dt.days\ntrain_data['diff_dat_chrono'] = pd.to_datetime(train_data['t_dat'].max()) - train_data['t_dat']\ntrain_data = train_data.loc[train_data['diff_dat'] <= 6]\n\n# find all purchases with date within the past week\nrecent_trans = train_data.loc[(train_data['diff_dat_chrono'].dt.days <= 6) & (train_data['sales_channel_id'] == 2)]\nprint('Train shape:',train_data.shape)\nprint('Recent trans shape:', recent_trans.shape)","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:13:31.624280Z","iopub.execute_input":"2022-05-10T02:13:31.624579Z","iopub.status.idle":"2022-05-10T02:13:35.991199Z","shell.execute_reply.started":"2022-05-10T02:13:31.624547Z","shell.execute_reply":"2022-05-10T02:13:35.990268Z"},"trusted":true},"execution_count":12,"outputs":[{"name":"stdout","text":"Train shape: (5181535, 8)\nRecent trans shape: (165809, 8)\n","output_type":"stream"}]},{"cell_type":"code","source":"# add all purchases into history\npurchase_dict = {}\n\n#for i,x in enumerate(zip(train_data['customer_id'], train_data['article_id'])):\nfor i,x in enumerate(zip(train_data['customer_id'], train_data['article_id'])):\n    cust_id, art_id = x\n    if cust_id not in purchase_dict:\n        purchase_dict[cust_id] = {}\n    \n    if art_id not in purchase_dict[cust_id]:\n        purchase_dict[cust_id][art_id] = 0\n    \n    purchase_dict[cust_id][art_id] += 1\n    \nprint(len(purchase_dict))","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:13:53.611431Z","iopub.execute_input":"2022-05-10T02:13:53.611848Z","iopub.status.idle":"2022-05-10T02:14:01.251730Z","shell.execute_reply.started":"2022-05-10T02:13:53.611818Z","shell.execute_reply":"2022-05-10T02:14:01.250788Z"},"trusted":true},"execution_count":13,"outputs":[{"name":"stdout","text":"1362281\n","output_type":"stream"}]},{"cell_type":"markdown","source":"### Most Popular Items\n\nFill the remaining slots with the most popular items from this month","metadata":{}},{"cell_type":"code","source":"# merge tables to links transaction to the product type and its overall color\ncolor_data = train_data\narticles_df['prod_and_shade'] = articles_df['perceived_colour_value_name'] + ' ' + articles_df['product_type_name']\ntmp = articles_df.loc[:, ['article_id', 'prod_and_shade']]\ncolor_data = color_data.merge(tmp, on=['article_id'], how='left', suffixes=('', '_'))\nrecent_trans = recent_trans.merge(tmp, on=['article_id'], how='left', suffixes=('', '_'))","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:14:03.404161Z","iopub.execute_input":"2022-05-10T02:14:03.404809Z","iopub.status.idle":"2022-05-10T02:14:05.411424Z","shell.execute_reply.started":"2022-05-10T02:14:03.404771Z","shell.execute_reply":"2022-05-10T02:14:05.410655Z"},"trusted":true},"execution_count":14,"outputs":[]},{"cell_type":"code","source":"# create dictionary that shows frequency counts of product-color combos purchased by customer\n\"\"\"\nprod_and_shade = sorted(color_data.to_pandas().groupby(['customer_id']).prod_and_shade.value_counts().to_dict())\nprod_shade_i = {}\nprod_shade_list = []\nfor cust_id, prod in prod_and_shade:\n    if cust_id not in prod_shade_i:\n        prod_shade_list.append((cust_id, [prod]))\n        prod_shade_i[cust_id] = len(prod_shade_i)\n    else:\n        prod_shade_list[prod_shade_i[cust_id]][1].append(prod)\n\nprod_shade_dict = dict([item for item in prod_shade_list if len(item[1]) > 2])\n\"\"\"","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:14:09.602302Z","iopub.execute_input":"2022-05-10T02:14:09.602623Z","iopub.status.idle":"2022-05-10T02:14:09.610334Z","shell.execute_reply.started":"2022-05-10T02:14:09.602591Z","shell.execute_reply":"2022-05-10T02:14:09.609646Z"},"trusted":true},"execution_count":15,"outputs":[{"execution_count":15,"output_type":"execute_result","data":{"text/plain":"\"\\nprod_and_shade = sorted(color_data.to_pandas().groupby(['customer_id']).prod_and_shade.value_counts().to_dict())\\nprod_shade_i = {}\\nprod_shade_list = []\\nfor cust_id, prod in prod_and_shade:\\n    if cust_id not in prod_shade_i:\\n        prod_shade_list.append((cust_id, [prod]))\\n        prod_shade_i[cust_id] = len(prod_shade_i)\\n    else:\\n        prod_shade_list[prod_shade_i[cust_id]][1].append(prod)\\n\\nprod_shade_dict = dict([item for item in prod_shade_list if len(item[1]) > 2])\\n\""},"metadata":{}}]},{"cell_type":"code","source":"train_data = train_data.loc[train_data['t_dat'] >= pd.to_datetime('2020-09-01')]\ntrain_data = train_data.merge(tmp, on=['article_id'], how='left', suffixes=('', '_'))\nprint(train_data.shape)","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:14:11.947027Z","iopub.execute_input":"2022-05-10T02:14:11.947489Z","iopub.status.idle":"2022-05-10T02:14:12.362447Z","shell.execute_reply.started":"2022-05-10T02:14:11.947438Z","shell.execute_reply":"2022-05-10T02:14:12.361460Z"},"trusted":true},"execution_count":16,"outputs":[{"name":"stdout","text":"(692104, 9)\n","output_type":"stream"}]},{"cell_type":"code","source":"# fill spots with top 3 items of customer's most frequent product-color category\n\"\"\"\ncustom_freq1 = []\nif cust_id in prod_shade_dict:\n    custom_freq1 = recent_trans.loc[recent_trans['prod_and_shade'] ==  prod_shade_dict[cust_id][0]]\n    if len(custom_freq1) >= 3:\n        custom_freq1 = list(custom_freq1['article_id'].value_counts().nlargest(3).index)\n    else:\n        custom_freq1 = list(custom_freq1['article_id'].value_counts().index)\ncustom_freq1 = ['0' + str(item) for item in custom_freq1]\n\"\"\"\n\ntrans_slice = train_data[['customer_id','article_id', 't_dat']]\n\ndef calc_pairs(train):\n    # Calculate all articles purchased together\n    dt = train.groupby(['customer_id','t_dat'])['article_id'].agg(list).rename('pair').reset_index()\n    df = train[['customer_id', 't_dat', 'article_id']].merge(dt, on=['customer_id', 't_dat'], how='left')\n    del dt\n    \n    df = df[['article_id', 'pair']].explode(column='pair') # Explode the rows vs list of articles\n    df = df.loc[df['article_id'] != df['pair']].reset_index(drop=True) # Discard duplicates\n    df = df.groupby(['article_id', 'pair']).size().rename('count').reset_index() # Count how many times each pair combination happens\n    df = df.sort_values(['article_id' ,'count'], ascending=False).reset_index(drop=True) # Sort by frequency\n    \n    # Pick most frequent pair\n    df['rank'] = df.groupby('article_id')['pair'].cumcount()\n    df = df.loc[df['rank'] == 0].reset_index(drop=True)\n    del df['rank']\n    \n    return df\n\n# customers who bought an item also bought these two items\npairs = calc_pairs(trans_slice)\npairs_dict = dict(zip(pairs['article_id'], pairs['pair']))\npairs.head()","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:15:08.950282Z","iopub.execute_input":"2022-05-10T02:15:08.951206Z","iopub.status.idle":"2022-05-10T02:15:16.278750Z","shell.execute_reply.started":"2022-05-10T02:15:08.951162Z","shell.execute_reply":"2022-05-10T02:15:16.277629Z"},"trusted":true},"execution_count":17,"outputs":[{"execution_count":17,"output_type":"execute_result","data":{"text/plain":"   article_id       pair  count\n0   956217002  816832001      4\n1   953763001  730683062      6\n2   953450001  448509014      2\n3   952938001  785018003      2\n4   952267001  671607001      6","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>article_id</th>\n      <th>pair</th>\n      <th>count</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>956217002</td>\n      <td>816832001</td>\n      <td>4</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>953763001</td>\n      <td>730683062</td>\n      <td>6</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>953450001</td>\n      <td>448509014</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>952938001</td>\n      <td>785018003</td>\n      <td>2</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>952267001</td>\n      <td>671607001</td>\n      <td>6</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"code","source":"benchmark = submission[['customer_id']]\nprediction_list = []\ntop12_str = list((train_data['article_id'].value_counts()).index)[:12]\ntop12 = ['0' + str(item) for item in top12_str]\nbase_pred = ' '.join(top12)\n\nfor i, cust_id in enumerate(submission['customer_id'].values.reshape((-1,))):\n    if (i % 100000 == 0): print(i, end=' ')\n    if cust_id in purchase_dict:\n        # top 6 past purchases\n        purch = ['0' + str(item[0]) for item in sorted((purchase_dict[cust_id]).items(), key=lambda x: x[1], reverse=True)]\n        purch_slice = purch[:6] if len(purch) >= 6 else purch\n        \n        # items also bought for customer's top 3 purchases\n        also_bought_slice = []\n        for i in range(len(purch_slice)):\n            if int(purch_slice[i][1:]) in pairs_dict:\n                also_bought = pairs_dict[int(purch_slice[i][1:])]\n                also_bought_slice.append('0' + str(also_bought))\n\n        # fill remaining spots with popular items and remove repeats\n        top12_slice = top12[:(12-len(purch_slice)-len(also_bought_slice))]\n        top12_rev_slice = top12[(12-len(purch_slice)-len(also_bought_slice)):]\n        for item in top12_slice:\n            if item in purch_slice or item in also_bought_slice: \n                top12_slice.remove(item)\n                top12_slice.append(top12_rev_slice.pop(0))\n\n        s = ' '.join(purch_slice + also_bought_slice + top12_slice)\n    else:\n        s = base_pred\n    prediction_list.append(s)","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:15:16.280569Z","iopub.execute_input":"2022-05-10T02:15:16.280846Z","iopub.status.idle":"2022-05-10T02:15:37.001273Z","shell.execute_reply.started":"2022-05-10T02:15:16.280815Z","shell.execute_reply":"2022-05-10T02:15:37.000292Z"},"trusted":true},"execution_count":18,"outputs":[{"name":"stdout","text":"0 100000 200000 300000 400000 500000 600000 700000 800000 900000 1000000 1100000 1200000 1300000 ","output_type":"stream"}]},{"cell_type":"code","source":"benchmark['prediction'] = prediction_list\nbenchmark.head()","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:15:37.003361Z","iopub.execute_input":"2022-05-10T02:15:37.003591Z","iopub.status.idle":"2022-05-10T02:15:37.161472Z","shell.execute_reply.started":"2022-05-10T02:15:37.003563Z","shell.execute_reply":"2022-05-10T02:15:37.160402Z"},"trusted":true},"execution_count":19,"outputs":[{"execution_count":19,"output_type":"execute_result","data":{"text/plain":"                                         customer_id  \\\n0  00000dbacae5abe5e23885899a1fa44253a17956c6d1c3...   \n1  0000423b00ade91418cceaf3b26c6af3dd342b51fd051e...   \n2  000058a12d5b43e67d225668fa1f8d618c13dc232df0ca...   \n3  00005ca1c9ed5f5146b52ac8639a40ca9d57aeff4d1bd2...   \n4  00006413d8573cd20ed7128e53b7b13819fe5cfc2d801f...   \n\n                                          prediction  \n0  0568601043 0779781015 0751471001 0909370001 09...  \n1  0826211002 0889616001 0751471001 0909370001 09...  \n2  0794321007 0805000001 0751471001 0909370001 09...  \n3  0742079001 0732413001 0751471001 0909370001 09...  \n4  0896152002 0730683050 0927530004 0791587015 09...  ","text/html":"<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>customer_id</th>\n      <th>prediction</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>00000dbacae5abe5e23885899a1fa44253a17956c6d1c3...</td>\n      <td>0568601043 0779781015 0751471001 0909370001 09...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0000423b00ade91418cceaf3b26c6af3dd342b51fd051e...</td>\n      <td>0826211002 0889616001 0751471001 0909370001 09...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>000058a12d5b43e67d225668fa1f8d618c13dc232df0ca...</td>\n      <td>0794321007 0805000001 0751471001 0909370001 09...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>00005ca1c9ed5f5146b52ac8639a40ca9d57aeff4d1bd2...</td>\n      <td>0742079001 0732413001 0751471001 0909370001 09...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>00006413d8573cd20ed7128e53b7b13819fe5cfc2d801f...</td>\n      <td>0896152002 0730683050 0927530004 0791587015 09...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"},"metadata":{}}]},{"cell_type":"markdown","source":"## 4. Submission","metadata":{}},{"cell_type":"code","source":"benchmark.to_csv('submission.csv', index=False)","metadata":{"execution":{"iopub.status.busy":"2022-05-10T02:15:37.162944Z","iopub.execute_input":"2022-05-10T02:15:37.163172Z","iopub.status.idle":"2022-05-10T02:15:49.894532Z","shell.execute_reply.started":"2022-05-10T02:15:37.163147Z","shell.execute_reply":"2022-05-10T02:15:49.893660Z"},"trusted":true},"execution_count":20,"outputs":[]},{"cell_type":"markdown","source":"## Credits  \nhttps://www.kaggle.com/code/vanguarde/h-m-eda-first-look  \nhttps://www.kaggle.com/code/titericz/article-id-pairs-in-3s-using-cudf","metadata":{}}]}