morebikes-analysis-per-station-training.ipynb

{"cells":[{"source":"<a href=\"https://www.kaggle.com/desmondrn/morebikes-analysis-per-station-training?scriptVersionId=84845102\" target=\"_blank\"><img align=\"left\" alt=\"Kaggle\" title=\"Open in Kaggle\" src=\"https://kaggle.com/static/images/open-in-kaggle.svg\"></a>","metadata":{},"cell_type":"markdown","outputs":[],"execution_count":0},{"cell_type":"code","execution_count":1,"id":"97c19269","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:21.730913Z","iopub.status.busy":"2022-01-10T01:34:21.730299Z","iopub.status.idle":"2022-01-10T01:34:21.930123Z","shell.execute_reply":"2022-01-10T01:34:21.93072Z","shell.execute_reply.started":"2022-01-09T23:56:02.686258Z"},"papermill":{"duration":0.237179,"end_time":"2022-01-10T01:34:21.931038","exception":false,"start_time":"2022-01-10T01:34:21.693859","status":"completed"},"tags":[]},"outputs":[],"source":["%reset -f"]},{"cell_type":"code","execution_count":2,"id":"24670338","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:21.993788Z","iopub.status.busy":"2022-01-10T01:34:21.993145Z","iopub.status.idle":"2022-01-10T01:34:21.995479Z","shell.execute_reply":"2022-01-10T01:34:21.994892Z","shell.execute_reply.started":"2022-01-09T23:56:03.192086Z"},"papermill":{"duration":0.034218,"end_time":"2022-01-10T01:34:21.995606","exception":false,"start_time":"2022-01-10T01:34:21.961388","status":"completed"},"tags":[]},"outputs":[],"source":["# !pip uninstall -y scikit-learn\n","# !pip install scikit-learn==0.24.1"]},{"cell_type":"code","execution_count":3,"id":"5a32ff97","metadata":{"_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","execution":{"iopub.execute_input":"2022-01-10T01:34:22.056579Z","iopub.status.busy":"2022-01-10T01:34:22.055933Z","iopub.status.idle":"2022-01-10T01:34:29.127492Z","shell.execute_reply":"2022-01-10T01:34:29.128213Z","shell.execute_reply.started":"2022-01-09T23:56:03.197141Z"},"papermill":{"duration":7.105065,"end_time":"2022-01-10T01:34:29.128467","exception":false,"start_time":"2022-01-10T01:34:22.023402","status":"completed"},"tags":[]},"outputs":[{"name":"stderr","output_type":"stream","text":["\n","User settings:\n","\n","   KMP_AFFINITY=granularity=fine,verbose,compact,1,0\n","   KMP_BLOCKTIME=0\n","   KMP_DUPLICATE_LIB_OK=True\n","   KMP_INIT_AT_FORK=FALSE\n","   KMP_SETTINGS=1\n","   KMP_WARNINGS=0\n","\n","Effective settings:\n","\n","   KMP_ABORT_DELAY=0\n","   KMP_ADAPTIVE_LOCK_PROPS='1,1024'\n","   KMP_ALIGN_ALLOC=64\n","   KMP_ALL_THREADPRIVATE=128\n","   KMP_ATOMIC_MODE=2\n","   KMP_BLOCKTIME=0\n","   KMP_CPUINFO_FILE: value is not defined\n","   KMP_DETERMINISTIC_REDUCTION=false\n","   KMP_DEVICE_THREAD_LIMIT=2147483647\n","   KMP_DISP_NUM_BUFFERS=7\n","   KMP_DUPLICATE_LIB_OK=true\n","   KMP_ENABLE_TASK_THROTTLING=true\n","   KMP_FORCE_REDUCTION: value is not defined\n","   KMP_FOREIGN_THREADS_THREADPRIVATE=true\n","   KMP_FORKJOIN_BARRIER='2,2'\n","   KMP_FORKJOIN_BARRIER_PATTERN='hyper,hyper'\n","   KMP_GTID_MODE=3\n","   KMP_HANDLE_SIGNALS=false\n","   KMP_HOT_TEAMS_MAX_LEVEL=1\n","   KMP_HOT_TEAMS_MODE=0\n","   KMP_INIT_AT_FORK=true\n","   KMP_LIBRARY=throughput\n","   KMP_LOCK_KIND=queuing\n","   KMP_MALLOC_POOL_INCR=1M\n","   KMP_NUM_LOCKS_IN_BLOCK=1\n","   KMP_PLAIN_BARRIER='2,2'\n","   KMP_PLAIN_BARRIER_PATTERN='hyper,hyper'\n","   KMP_REDUCTION_BARRIER='1,1'\n","   KMP_REDUCTION_BARRIER_PATTERN='hyper,hyper'\n","   KMP_SCHEDULE='static,balanced;guided,iterative'\n","   KMP_SETTINGS=true\n","   KMP_SPIN_BACKOFF_PARAMS='4096,100'\n","   KMP_STACKOFFSET=64\n","   KMP_STACKPAD=0\n","   KMP_STACKSIZE=8M\n","   KMP_STORAGE_MAP=false\n","   KMP_TASKING=2\n","   KMP_TASKLOOP_MIN_TASKS=0\n","   KMP_TASK_STEALING_CONSTRAINT=1\n","   KMP_TEAMS_THREAD_LIMIT=4\n","   KMP_TOPOLOGY_METHOD=all\n","   KMP_USE_YIELD=1\n","   KMP_VERSION=false\n","   KMP_WARNINGS=false\n","   OMP_AFFINITY_FORMAT='OMP: pid %P tid %i thread %n bound to OS proc set {%A}'\n","   OMP_ALLOCATOR=omp_default_mem_alloc\n","   OMP_CANCELLATION=false\n","   OMP_DEFAULT_DEVICE=0\n","   OMP_DISPLAY_AFFINITY=false\n","   OMP_DISPLAY_ENV=false\n","   OMP_DYNAMIC=false\n","   OMP_MAX_ACTIVE_LEVELS=1\n","   OMP_MAX_TASK_PRIORITY=0\n","   OMP_NESTED: deprecated; max-active-levels-var=1\n","   OMP_NUM_THREADS: value is not defined\n","   OMP_PLACES: value is not defined\n","   OMP_PROC_BIND='intel'\n","   OMP_SCHEDULE='static'\n","   OMP_STACKSIZE=8M\n","   OMP_TARGET_OFFLOAD=DEFAULT\n","   OMP_THREAD_LIMIT=2147483647\n","   OMP_WAIT_POLICY=PASSIVE\n","   KMP_AFFINITY='verbose,warnings,respect,granularity=fine,compact,1,0'\n","\n","2022-01-10 01:34:29.066842: I tensorflow/core/common_runtime/process_util.cc:146] Creating new thread pool with default inter op setting: 2. Tune using inter_op_parallelism_threads for best performance.\n"]}],"source":["%%capture\n","\n","import numpy as np \n","import pandas as pd\n","import seaborn as sns\n","import matplotlib.pyplot as plt\n","\n","from sklearn.preprocessing import StandardScaler\n","from sklearn.preprocessing import MinMaxScaler\n","\n","import joblib\n","\n","import tensorflow as tf\n","mae = tf.keras.losses.MeanAbsoluteError()    ## The loss for this task\n","_ = mae([1], [2])                            ## Trick to initialize tensorflow here !\n","\n","import os"]},{"cell_type":"markdown","id":"df42f03e","metadata":{"papermill":{"duration":0.026099,"end_time":"2022-01-10T01:34:29.184637","exception":false,"start_time":"2022-01-10T01:34:29.158538","status":"completed"},"tags":[]},"source":["# Objective\n","The goal of this notebook is to analyse the `motorbikes2021` dataset and identify the best features and best per-station machine learning algorithms to use in the next phases of the competition. Our task is to predict the number of bikes at a station in 3h time."]},{"cell_type":"markdown","id":"1caae763","metadata":{"papermill":{"duration":0.025961,"end_time":"2022-01-10T01:34:29.236944","exception":false,"start_time":"2022-01-10T01:34:29.210983","status":"completed"},"tags":[]},"source":["# Load data for a few stations and analyse\n","\n","**_Which features are the most important for learning the number of bikes at a station?_** The answer will focus on a particular station (identified by its id `STATION_ID`). Now some features were engineered and provided to us in the dataset. These are described by `short`, `short_temp`, `full`, `full_temp`, `short_full`, and `short_full_temp`. Change the variable `FEATURE_SET` to either of these values to get different results."]},{"cell_type":"code","execution_count":4,"id":"9db03d68","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:29.295071Z","iopub.status.busy":"2022-01-10T01:34:29.294454Z","iopub.status.idle":"2022-01-10T01:34:29.307211Z","shell.execute_reply":"2022-01-10T01:34:29.307706Z","shell.execute_reply.started":"2022-01-09T23:56:03.215441Z"},"papermill":{"duration":0.044491,"end_time":"2022-01-10T01:34:29.307878","exception":false,"start_time":"2022-01-10T01:34:29.263387","status":"completed"},"tags":[]},"outputs":[{"data":{"text/plain":["['example_leaderboard_submission.csv', 'Models', 'test.csv', 'Train']"]},"execution_count":4,"metadata":{},"output_type":"execute_result"}],"source":["INPUT_DIR = '/kaggle/input/morebikes2021'\n","\n","STATION_ID = 252                  ## Change here to train models for other stations\n","FEATURE_SET = 'short_full_temp'   ## Change the features here (use 'all' to use all possible features)\n","\n","SHUFFLE_SAMPLES = True           ## Change here to shuffle the dataset before spliting\n","TRAIN_SIZE = 0.80                ## Change the train proportion here\n","\n","os.listdir(INPUT_DIR)"]},{"cell_type":"code","execution_count":5,"id":"957df6ca","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:29.364534Z","iopub.status.busy":"2022-01-10T01:34:29.363581Z","iopub.status.idle":"2022-01-10T01:34:29.418193Z","shell.execute_reply":"2022-01-10T01:34:29.4177Z","shell.execute_reply.started":"2022-01-09T23:56:03.232975Z"},"papermill":{"duration":0.084028,"end_time":"2022-01-10T01:34:29.418335","exception":false,"start_time":"2022-01-10T01:34:29.334307","status":"completed"},"tags":[]},"outputs":[{"data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>station</th>\n","      <th>latitude</th>\n","      <th>longitude</th>\n","      <th>numDocks</th>\n","      <th>timestamp</th>\n","      <th>year</th>\n","      <th>month</th>\n","      <th>day</th>\n","      <th>hour</th>\n","      <th>weekday</th>\n","      <th>...</th>\n","      <th>temperature.C</th>\n","      <th>relHumidity.HR</th>\n","      <th>airPressure.mb</th>\n","      <th>precipitation.l.m2</th>\n","      <th>bikes_3h_ago</th>\n","      <th>full_profile_3h_diff_bikes</th>\n","      <th>full_profile_bikes</th>\n","      <th>short_profile_3h_diff_bikes</th>\n","      <th>short_profile_bikes</th>\n","      <th>bikes</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>252</td>\n","      <td>39.484577</td>\n","      <td>-0.401273</td>\n","      <td>15</td>\n","      <td>1.412114e+09</td>\n","      <td>2014</td>\n","      <td>10</td>\n","      <td>1</td>\n","      <td>0</td>\n","      <td>Wednesday</td>\n","      <td>...</td>\n","      <td>21.3</td>\n","      <td>85.0</td>\n","      <td>855.3</td>\n","      <td>0.0</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>6.0</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>252</td>\n","      <td>39.484577</td>\n","      <td>-0.401273</td>\n","      <td>15</td>\n","      <td>1.412118e+09</td>\n","      <td>2014</td>\n","      <td>10</td>\n","      <td>1</td>\n","      <td>1</td>\n","      <td>Wednesday</td>\n","      <td>...</td>\n","      <td>21.1</td>\n","      <td>86.0</td>\n","      <td>1000.6</td>\n","      <td>0.0</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>6.0</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>252</td>\n","      <td>39.484577</td>\n","      <td>-0.401273</td>\n","      <td>15</td>\n","      <td>1.412122e+09</td>\n","      <td>2014</td>\n","      <td>10</td>\n","      <td>1</td>\n","      <td>2</td>\n","      <td>Wednesday</td>\n","      <td>...</td>\n","      <td>20.9</td>\n","      <td>86.0</td>\n","      <td>880.6</td>\n","      <td>0.0</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>5.0</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>252</td>\n","      <td>39.484577</td>\n","      <td>-0.401273</td>\n","      <td>15</td>\n","      <td>1.412125e+09</td>\n","      <td>2014</td>\n","      <td>10</td>\n","      <td>1</td>\n","      <td>3</td>\n","      <td>Wednesday</td>\n","      <td>...</td>\n","      <td>20.4</td>\n","      <td>88.0</td>\n","      <td>859.8</td>\n","      <td>0.0</td>\n","      <td>6.0</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>5.0</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>252</td>\n","      <td>39.484577</td>\n","      <td>-0.401273</td>\n","      <td>15</td>\n","      <td>1.412129e+09</td>\n","      <td>2014</td>\n","      <td>10</td>\n","      <td>1</td>\n","      <td>4</td>\n","      <td>Wednesday</td>\n","      <td>...</td>\n","      <td>20.3</td>\n","      <td>87.0</td>\n","      <td>898.1</td>\n","      <td>0.0</td>\n","      <td>6.0</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>NaN</td>\n","      <td>5.0</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>5 rows × 25 columns</p>\n","</div>"],"text/plain":["   station   latitude  longitude  numDocks     timestamp  year  month  day  \\\n","0      252  39.484577  -0.401273        15  1.412114e+09  2014     10    1   \n","1      252  39.484577  -0.401273        15  1.412118e+09  2014     10    1   \n","2      252  39.484577  -0.401273        15  1.412122e+09  2014     10    1   \n","3      252  39.484577  -0.401273        15  1.412125e+09  2014     10    1   \n","4      252  39.484577  -0.401273        15  1.412129e+09  2014     10    1   \n","\n","   hour    weekday  ...  temperature.C  relHumidity.HR  airPressure.mb  \\\n","0     0  Wednesday  ...           21.3            85.0           855.3   \n","1     1  Wednesday  ...           21.1            86.0          1000.6   \n","2     2  Wednesday  ...           20.9            86.0           880.6   \n","3     3  Wednesday  ...           20.4            88.0           859.8   \n","4     4  Wednesday  ...           20.3            87.0           898.1   \n","\n","   precipitation.l.m2  bikes_3h_ago  full_profile_3h_diff_bikes  \\\n","0                 0.0           NaN                         NaN   \n","1                 0.0           NaN                         NaN   \n","2                 0.0           NaN                         NaN   \n","3                 0.0           6.0                         NaN   \n","4                 0.0           6.0                         NaN   \n","\n","   full_profile_bikes  short_profile_3h_diff_bikes  short_profile_bikes  bikes  \n","0                 NaN                          NaN                  NaN    6.0  \n","1                 NaN                          NaN                  NaN    6.0  \n","2                 NaN                          NaN                  NaN    5.0  \n","3                 NaN                          NaN                  NaN    5.0  \n","4                 NaN                          NaN                  NaN    5.0  \n","\n","[5 rows x 25 columns]"]},"execution_count":5,"metadata":{},"output_type":"execute_result"}],"source":["def get_train_path(station_id):\n","    return INPUT_DIR+'/Train/Train/station_'+str(station_id)+'_deploy.csv'\n","\n","df = pd.read_csv(get_train_path(STATION_ID))\n","\n","df.head()"]},{"cell_type":"code","execution_count":6,"id":"7cdea968","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:29.484656Z","iopub.status.busy":"2022-01-10T01:34:29.483974Z","iopub.status.idle":"2022-01-10T01:34:31.072066Z","shell.execute_reply":"2022-01-10T01:34:31.072564Z","shell.execute_reply.started":"2022-01-09T23:56:03.280831Z"},"papermill":{"duration":1.62678,"end_time":"2022-01-10T01:34:31.072746","exception":false,"start_time":"2022-01-10T01:34:29.445966","status":"completed"},"tags":[]},"outputs":[{"data":{"image/png":"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\n","text/plain":["<Figure size 1080x720 with 6 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["def plot_corr_map(station_id, ax, labels=True):\n","    df = pd.read_csv(get_train_path(station_id))\n","    df = df.drop(['station', 'weekday'], axis=1)\n","\n","    s = df.corrwith(df['bikes'])\n","    if not labels: s = np.array(s)\n","    df = pd.DataFrame({'bikes':s})\n","\n","    ax.set_title('Station '+str(station_id))\n","    sns.heatmap(df, annot=True, ax=ax)\n","    \n","fig, ax = plt.subplots(1, 3, figsize=(15,10)) \n","\n","## Investigate correlation between features and target variable for specific stations\n","plot_corr_map(201, ax=ax[0])\n","plot_corr_map(202, ax=ax[1], labels=False)\n","plot_corr_map(STATION_ID, ax=ax[2], labels=False)"]},{"cell_type":"markdown","id":"fa4385d2","metadata":{"papermill":{"duration":0.029272,"end_time":"2022-01-10T01:34:31.132964","exception":false,"start_time":"2022-01-10T01:34:31.103692","status":"completed"},"tags":[]},"source":["Analysing the correlation between predictor variables and our target variable for three seperate stations indicate that some 'weather' and 'timepoint' features will be important for training. On the other hand, features relating to the station and its location will be dropped (`station`, `latitude` and`longitude`). The same goes for non-numeric features like `weekday` or constant features like `year` or `month`. In order to use all these interresting features for training, the `FEATURE_SET` variable above can be set `all`."]},{"cell_type":"code","execution_count":7,"id":"6f942bfa","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:31.202627Z","iopub.status.busy":"2022-01-10T01:34:31.20185Z","iopub.status.idle":"2022-01-10T01:34:31.204Z","shell.execute_reply":"2022-01-10T01:34:31.204446Z","shell.execute_reply.started":"2022-01-09T23:56:05.255636Z"},"papermill":{"duration":0.041685,"end_time":"2022-01-10T01:34:31.204643","exception":false,"start_time":"2022-01-10T01:34:31.162958","status":"completed"},"tags":[]},"outputs":[],"source":["def preprocess(df):\n","    df = df.drop(['weekday', 'latitude', 'longitude','year', 'month'], axis=1)\n","#     df = df.drop(['numDocks'], axis=1)\n","    df = df.drop(['station'], axis=1)\n","    df['timestamp'] = (df['timestamp'] - 1412114400) / 3600    ## 1412114400 is the closest to the Epoch in the dataset\n","    return df\n","\n","def get_peters_features():\n","    \"\"\"\n","    Features suggested by the competition\n","    \"\"\"\n","    short = ['bikes_3h_ago', 'short_profile_3h_diff_bikes', 'short_profile_bikes']\n","    short_temp = short + ['temperature.C']\n","    full = ['bikes_3h_ago', 'full_profile_3h_diff_bikes', 'full_profile_bikes']\n","    full_temp = full + ['temperature.C']\n","    short_full = ['bikes_3h_ago', 'short_profile_3h_diff_bikes', 'short_profile_bikes', 'full_profile_3h_diff_bikes', 'full_profile_bikes']\n","    short_full_temp = short_full + ['temperature.C']\n","    \n","    if FEATURE_SET=='short':\n","        return short\n","    elif FEATURE_SET=='short_temp':\n","        return short_temp\n","    elif FEATURE_SET=='full':\n","        return full\n","    elif FEATURE_SET=='full_temp':\n","        return full_temp\n","    elif FEATURE_SET=='short_full':\n","        return short_full\n","    elif FEATURE_SET=='short_full_temp':\n","        return short_full_temp\n","    else:\n","        return 'all'\n","\n","def select_peters_features(df):\n","    features_to_use = get_peters_features()\n","    if features_to_use == 'all':\n","        return df.dropna()\n","    else:\n","        if 'bikes' in list(df.columns):\n","            features_to_use += ['bikes']\n","        return df[features_to_use].dropna()"]},{"cell_type":"code","execution_count":8,"id":"b01efda3","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:31.268071Z","iopub.status.busy":"2022-01-10T01:34:31.267479Z","iopub.status.idle":"2022-01-10T01:34:31.285089Z","shell.execute_reply":"2022-01-10T01:34:31.285557Z","shell.execute_reply.started":"2022-01-09T23:56:05.267525Z"},"papermill":{"duration":0.051149,"end_time":"2022-01-10T01:34:31.28574","exception":false,"start_time":"2022-01-10T01:34:31.234591","status":"completed"},"tags":[]},"outputs":[],"source":["def make_learning_data(station_id):\n","    df = pd.read_csv(get_train_path(station_id))\n","    \n","    df = preprocess(df)\n","    if SHUFFLE_SAMPLES:\n","        df = df.sample(frac=1, random_state=12)\n","\n","    df = select_peters_features(df)\n","\n","    df_y = df['bikes']\n","    df_X = df.drop(['bikes'], axis=1)\n","\n","    return np.array(df_X), np.array(df_y)\n","\n","X, y = make_learning_data(STATION_ID)"]},{"cell_type":"code","execution_count":9,"id":"43dac304","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:31.349554Z","iopub.status.busy":"2022-01-10T01:34:31.348867Z","iopub.status.idle":"2022-01-10T01:34:31.358354Z","shell.execute_reply":"2022-01-10T01:34:31.357745Z","shell.execute_reply.started":"2022-01-09T23:56:05.296844Z"},"papermill":{"duration":0.041844,"end_time":"2022-01-10T01:34:31.358535","exception":false,"start_time":"2022-01-10T01:34:31.316691","status":"completed"},"tags":[]},"outputs":[{"name":"stdout","output_type":"stream","text":["Train shapes: (457, 6) (457,)\n","Val shapes: (115, 6) (115,)\n"]}],"source":["\"\"\" Split the data between training and validation \"\"\"\n","def split_dataset(X, y, train_size):\n","    dataset_size = len(X)\n","    assert dataset_size == len(y)\n","    assert dataset_size >= train_size\n","\n","    X_train, y_train = X[:train_size], y[:train_size]\n","    X_val, y_val = X[train_size:], y[train_size:]\n","\n","    return (X_train, y_train), (X_val, y_val)\n","\n","(X_train, y_train), (X_val, y_val) = split_dataset(X, y, int(TRAIN_SIZE*len(X)))\n","\n","print(\"Train shapes:\", X_train.shape, y_train.shape)\n","print(\"Val shapes:\", X_val.shape, y_val.shape)"]},{"cell_type":"markdown","id":"6619ff70","metadata":{"papermill":{"duration":0.0293,"end_time":"2022-01-10T01:34:31.417622","exception":false,"start_time":"2022-01-10T01:34:31.388322","status":"completed"},"tags":[]},"source":["After imputing missing values, we've split the loaded data for training and validation while avoiding shuffling. We will train our models on 80% of the data (approximately 25 days) and evaluate the models on the remaining 6 last days of the month. The idea is that the model that performs best on this slightly drifted data is most likely to perform best on the more drifted testing data (for the next 3 months). Here, the 80/20% split is for demonstration purpose; in production setting, we went for a 95/5% division to achieve better performance."]},{"cell_type":"markdown","id":"8c25cce6","metadata":{"papermill":{"duration":0.030018,"end_time":"2022-01-10T01:34:31.479045","exception":false,"start_time":"2022-01-10T01:34:31.449027","status":"completed"},"tags":[]},"source":["# Per-station trainings\n","\n","We identified several machine learning algorithms that can be interesting for our task. Each algorithm will be trained, tuned and stored for later comparison at the end of the notebook. Our unique metric for evaluation is the MAE, which tells us by how many (on average) our predictions differ from the actual number of bikes at the stations. \n","\n","Given the wide range of values our features can take, most algorithms will require a form of scaling before fitting, hence the widespread usage of the `Pipleline` class in what follows."]},{"cell_type":"code","execution_count":10,"id":"e2699d38","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:31.543807Z","iopub.status.busy":"2022-01-10T01:34:31.542797Z","iopub.status.idle":"2022-01-10T01:34:31.546844Z","shell.execute_reply":"2022-01-10T01:34:31.547385Z","shell.execute_reply.started":"2022-01-09T23:56:05.307319Z"},"papermill":{"duration":0.038813,"end_time":"2022-01-10T01:34:31.547574","exception":false,"start_time":"2022-01-10T01:34:31.508761","status":"completed"},"tags":[]},"outputs":[],"source":["\"\"\" All models that will be trained and their scores \"\"\"\n","\n","potential_models = []"]},{"cell_type":"markdown","id":"19c4c131","metadata":{"papermill":{"duration":0.029925,"end_time":"2022-01-10T01:34:31.608255","exception":false,"start_time":"2022-01-10T01:34:31.57833","status":"completed"},"tags":[]},"source":["## Linear Regression\n","\n","Given the nature of the variables, linear regression is the most intuitive choice here. We opted for Ridge regularization to avoid massive weights, while keeping as many features as possible (not setting weights to 0). We also considered using polynomial and interaction features (up to degree 3) which, depending on the initial set of features we chose, improved the results."]},{"cell_type":"code","execution_count":11,"id":"3375ec94","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:31.671748Z","iopub.status.busy":"2022-01-10T01:34:31.670764Z","iopub.status.idle":"2022-01-10T01:34:32.647589Z","shell.execute_reply":"2022-01-10T01:34:32.64709Z","shell.execute_reply.started":"2022-01-09T23:56:05.322378Z"},"papermill":{"duration":1.009655,"end_time":"2022-01-10T01:34:32.647736","exception":false,"start_time":"2022-01-10T01:34:31.638081","status":"completed"},"tags":[]},"outputs":[{"data":{"image/png":"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\n","text/plain":["<Figure size 1296x360 with 3 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["from sklearn.pipeline import Pipeline\n","from sklearn.preprocessing import PolynomialFeatures\n","from sklearn.linear_model import LinearRegression, Lasso, Ridge\n","from sklearn.linear_model import PoissonRegressor\n","from sklearn.model_selection import cross_val_score\n","\n","degrees = [1, 2, 3]\n","\n","plt.figure(figsize=(18, 5))\n","for i in range(len(degrees)):\n","    ax = plt.subplot(1, len(degrees), i + 1)\n","\n","    scaled_features = StandardScaler()\n","#     scaled_features = MinMaxScaler()\n","    polynomial_features = PolynomialFeatures(degree=degrees[i], include_bias=False)\n","    linear_regression = LinearRegression()\n","#     linear_regression = PoissonRegressor()\n","#     linear_regression = Lasso()\n","    linear_regression = Ridge()\n","    pipeline = Pipeline(\n","        [\n","            (\"scaled_features\", scaled_features),\n","            (\"polynomial_features\", polynomial_features),\n","            (\"linear_regression\", linear_regression),\n","        ]\n","    )\n","    pipeline.fit(X_train, y_train)\n","\n","    y_pred = np.rint(pipeline.predict(X_val))\n","    score = mae(y_val, y_pred).numpy()\n","    \n","    potential_models.append(('linear_reg_'+str(i+1), pipeline, score))\n","\n","    plt.plot(y_val, \"o\", label=\"True values\")\n","    plt.plot(y_pred, \".\", label=\"Predictions\")\n","    ax.set_xlabel(\"instances\")\n","    plt.ylabel(\"# bikes\")\n","    plt.legend(loc=\"best\")\n","    plt.title(\"Degree: {}\\nMAE = {:.2f}\".format(degrees[i], score))"]},{"cell_type":"code","execution_count":12,"id":"073247e0","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:32.713744Z","iopub.status.busy":"2022-01-10T01:34:32.713001Z","iopub.status.idle":"2022-01-10T01:34:32.717538Z","shell.execute_reply":"2022-01-10T01:34:32.717953Z","shell.execute_reply.started":"2022-01-09T23:56:06.051298Z"},"papermill":{"duration":0.039262,"end_time":"2022-01-10T01:34:32.718117","exception":false,"start_time":"2022-01-10T01:34:32.678855","status":"completed"},"tags":[]},"outputs":[{"data":{"text/plain":["array([ 2.50644847,  0.46833421,  0.09366008,  0.46833421,  0.09366008,\n","       -0.09900921])"]},"execution_count":12,"metadata":{},"output_type":"execute_result"}],"source":["### See what the weights look like\n","potential_models[0][1].named_steps['linear_regression'].coef_"]},{"cell_type":"markdown","id":"d9108937","metadata":{"papermill":{"duration":0.030985,"end_time":"2022-01-10T01:34:32.780554","exception":false,"start_time":"2022-01-10T01:34:32.749569","status":"completed"},"tags":[]},"source":["## k-Nearest Neighbors\n","\n","Our task can also be viewed as a multiclass classification problem, with exactly `numDocks+1` classes. The first algorithm we considered for this approach is the low-bias and high-variance k-Nearest Neighbors algorithm with `k=2`. Compared to other models, we found that kNN produces better results when the input data is not normalized."]},{"cell_type":"code","execution_count":13,"id":"53e4dc6a","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:32.849552Z","iopub.status.busy":"2022-01-10T01:34:32.848568Z","iopub.status.idle":"2022-01-10T01:34:32.92641Z","shell.execute_reply":"2022-01-10T01:34:32.927018Z","shell.execute_reply.started":"2022-01-09T23:56:06.061256Z"},"papermill":{"duration":0.115157,"end_time":"2022-01-10T01:34:32.927195","exception":false,"start_time":"2022-01-10T01:34:32.812038","status":"completed"},"tags":[]},"outputs":[{"data":{"text/plain":["KNeighborsClassifier(n_neighbors=2)"]},"execution_count":13,"metadata":{},"output_type":"execute_result"}],"source":["from sklearn.neighbors import KNeighborsClassifier\n","KNN = KNeighborsClassifier(n_neighbors=2)\n","\n","KNN.fit(X_train, y_train)"]},{"cell_type":"code","execution_count":14,"id":"24d1ffb2","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:32.995537Z","iopub.status.busy":"2022-01-10T01:34:32.994861Z","iopub.status.idle":"2022-01-10T01:34:33.204376Z","shell.execute_reply":"2022-01-10T01:34:33.204985Z","shell.execute_reply.started":"2022-01-09T23:56:06.0815Z"},"papermill":{"duration":0.245167,"end_time":"2022-01-10T01:34:33.205183","exception":false,"start_time":"2022-01-10T01:34:32.960016","status":"completed"},"tags":[]},"outputs":[{"data":{"image/png":"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\n","text/plain":["<Figure size 1296x360 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["def plot_results(y_pred, title):\n","    fig, ax = plt.subplots(1, 1, figsize=(18, 5))\n","\n","    plt.plot(y_val, \"o\", label=\"True values\")\n","    ax.plot(y_pred, \".\", label=\"Predictions\")\n","    ax.set_xlabel(\"instances\")\n","    ax.set_ylabel(\"# bikes\")\n","    ax.legend(loc=\"best\")\n","    ax.set_title(\n","        title+\" \\nMAE = {:.2f}\".format(\n","            mae(y_val, y_pred).numpy()\n","        )\n","    )\n","\n","y_pred = KNN.predict(X_val)\n","\n","potential_models.append(('knn', KNN, mae(y_val, y_pred).numpy()))\n","\n","plot_results(y_pred, title='kNN')"]},{"cell_type":"markdown","id":"18aba946","metadata":{"papermill":{"duration":0.032477,"end_time":"2022-01-10T01:34:33.277671","exception":false,"start_time":"2022-01-10T01:34:33.245194","status":"completed"},"tags":[]},"source":["## Logistic regression\n","\n","*Logistic regression with multiple classes is not properly implemented in our version of scikit-learn (version 0.23), so we left it out, even though it could have brought interesting insights into the data.*"]},{"cell_type":"code","execution_count":15,"id":"da91feb9","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:33.346815Z","iopub.status.busy":"2022-01-10T01:34:33.346184Z","iopub.status.idle":"2022-01-10T01:34:33.350324Z","shell.execute_reply":"2022-01-10T01:34:33.350864Z","shell.execute_reply.started":"2022-01-09T23:56:06.388792Z"},"papermill":{"duration":0.040399,"end_time":"2022-01-10T01:34:33.351033","exception":false,"start_time":"2022-01-10T01:34:33.310634","status":"completed"},"tags":[]},"outputs":[],"source":["# from sklearn.linear_model import LogisticRegressionCV\n","\n","## LogReg = LogisticRegressionCV(multi_class='multinomial', solver='lbfgs', max_iter=400)\n","# LogReg = Pipeline([(\"minmax_scaler\", MinMaxScaler()), \n","#                    (\"logistic_reg\", LogisticRegressionCV(multi_class='ovr', solver='lbfgs', max_iter=200))])\n","\n","# LogReg.fit(X_train, y_train)"]},{"cell_type":"code","execution_count":16,"id":"5bab3a5a","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:33.424477Z","iopub.status.busy":"2022-01-10T01:34:33.42386Z","iopub.status.idle":"2022-01-10T01:34:33.425461Z","shell.execute_reply":"2022-01-10T01:34:33.425886Z","shell.execute_reply.started":"2022-01-09T23:56:06.393416Z"},"papermill":{"duration":0.041536,"end_time":"2022-01-10T01:34:33.426036","exception":false,"start_time":"2022-01-10T01:34:33.3845","status":"completed"},"tags":[]},"outputs":[],"source":["# y_pred = LogReg.predict(X_val)\n","# potential_models.append(('logistic_reg', LogReg, mae(y_val, y_pred).numpy()))\n","# plot_results(y_pred, title='Logistic regression')"]},{"cell_type":"markdown","id":"b252d3d4","metadata":{"papermill":{"duration":0.032368,"end_time":"2022-01-10T01:34:33.492084","exception":false,"start_time":"2022-01-10T01:34:33.459716","status":"completed"},"tags":[]},"source":["## Decision Tree\n","\n","Another high-variance model we tested is decision trees. We placed no limits on the number of leafs nor the minimum features required to split a node.  "]},{"cell_type":"code","execution_count":17,"id":"df7f8ba8","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:33.560698Z","iopub.status.busy":"2022-01-10T01:34:33.56005Z","iopub.status.idle":"2022-01-10T01:34:33.596019Z","shell.execute_reply":"2022-01-10T01:34:33.595531Z","shell.execute_reply.started":"2022-01-09T23:56:06.408044Z"},"papermill":{"duration":0.071402,"end_time":"2022-01-10T01:34:33.59617","exception":false,"start_time":"2022-01-10T01:34:33.524768","status":"completed"},"tags":[]},"outputs":[{"data":{"text/plain":["Pipeline(steps=[('minmax_scaler', MinMaxScaler()),\n","                ('decision_tree', DecisionTreeClassifier(random_state=12))])"]},"execution_count":17,"metadata":{},"output_type":"execute_result"}],"source":["from sklearn.tree import DecisionTreeClassifier\n","\n","# DcTree = DecisionTreeClassifier(random_state=12)\n","DcTree = Pipeline([(\"minmax_scaler\", MinMaxScaler()), (\"decision_tree\", DecisionTreeClassifier(random_state=12))])\n","\n","DcTree.fit(X_train, y_train)"]},{"cell_type":"code","execution_count":18,"id":"2b66bb77","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:33.667795Z","iopub.status.busy":"2022-01-10T01:34:33.667123Z","iopub.status.idle":"2022-01-10T01:34:33.870261Z","shell.execute_reply":"2022-01-10T01:34:33.869194Z","shell.execute_reply.started":"2022-01-09T23:56:06.430815Z"},"papermill":{"duration":0.241359,"end_time":"2022-01-10T01:34:33.870448","exception":false,"start_time":"2022-01-10T01:34:33.629089","status":"completed"},"tags":[]},"outputs":[{"data":{"image/png":"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\n","text/plain":["<Figure size 1296x360 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["y_pred = DcTree.predict(X_val)\n","potential_models.append(('decision_tree', DcTree, mae(y_val, y_pred).numpy()))\n","plot_results(y_pred, title='Decision Tree')"]},{"cell_type":"markdown","id":"81cacc70","metadata":{"papermill":{"duration":0.043773,"end_time":"2022-01-10T01:34:33.957541","exception":false,"start_time":"2022-01-10T01:34:33.913768","status":"completed"},"tags":[]},"source":["## Random Forest\n","\n","In an attempt to reduce the variance and improve results obtianed from decision trees, we tested random forests."]},{"cell_type":"code","execution_count":19,"id":"2441eb5b","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:34.047941Z","iopub.status.busy":"2022-01-10T01:34:34.047075Z","iopub.status.idle":"2022-01-10T01:34:34.323308Z","shell.execute_reply":"2022-01-10T01:34:34.322826Z","shell.execute_reply.started":"2022-01-09T23:56:06.745977Z"},"papermill":{"duration":0.320152,"end_time":"2022-01-10T01:34:34.323489","exception":false,"start_time":"2022-01-10T01:34:34.003337","status":"completed"},"tags":[]},"outputs":[{"data":{"text/plain":["Pipeline(steps=[('minmax_scaler', MinMaxScaler()),\n","                ('random_forest',\n","                 RandomForestClassifier(n_jobs=-1, random_state=12))])"]},"execution_count":19,"metadata":{},"output_type":"execute_result"}],"source":["from sklearn.ensemble import RandomForestClassifier\n","\n","# RdFor = RandomForestClassifier(n_estimators=300, max_leaf_nodes=None, n_jobs=-1, random_state=12)\n","RdFor = Pipeline([(\"minmax_scaler\", MinMaxScaler()), (\"random_forest\", RandomForestClassifier(n_estimators=100, n_jobs=-1, random_state=12))])\n","\n","RdFor.fit(X_train, y_train)"]},{"cell_type":"code","execution_count":20,"id":"2ef4f7ac","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:34.400877Z","iopub.status.busy":"2022-01-10T01:34:34.400182Z","iopub.status.idle":"2022-01-10T01:34:34.698149Z","shell.execute_reply":"2022-01-10T01:34:34.697562Z","shell.execute_reply.started":"2022-01-09T23:56:07.008525Z"},"papermill":{"duration":0.340734,"end_time":"2022-01-10T01:34:34.698289","exception":false,"start_time":"2022-01-10T01:34:34.357555","status":"completed"},"tags":[]},"outputs":[{"data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAABBwAAAFcCAYAAABr+63OAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8/fFQqAAAACXBIWXMAAAsTAAALEwEAmpwYAABEYElEQVR4nO3dfXxcdZ3//fcnmUQisISbiqZUU10tg9y0GkqUDXKBS1CRVoRdtLsUV9dlr91L1LVK6vVTua7l5rfhx6qsK+IdRftzYTFWVDTeIJKVjSUlQIEh3hFsE4QumHI3mkzy+f0xk2xakswkc86cc2Zez8ejj8mc28/5fr/ne2Y+Pec75u4CAAAAAAAIUl3UAQAAAAAAgOpDwgEAAAAAAASOhAMAAAAAAAgcCQcAAAAAABA4Eg4AAAAAACBwJBwAAAAAAEDgSDgAAICizOwTZvbVqOMAAADJQcIBAICEMrNhM8ua2TNm9lszu97MDoo6rnKY2almNlU4pul/36rg/lvNzM0sVal9AgBQrUg4AACQbG9194MkrZa0RlJXtOEEYtTdD5r1762L3YCZ1YcRGAAAKB0JBwAAqoC7/1ZSr/KJB0mSmV1iZr8ys6fN7EEze9useRea2X+Y2VVm9jsze9jM3jRr/koz+0lh3R9IOmL2/szsbDN7wMzGzOx2M0vPmjdsZpvM7D4ze9bMvmhmR5rZdwvb+6GZHbrYYzSzdGFfY4V9nz1r3vVm9lkzu9XMnpX0f5lZi5l93cz2FI7vfbOWX2tmA2b2lJk9ZmZXF2bdUXgdK9xd8brFxgkAAPJIOAAAUAXM7ChJb5L0y1mTfyWpQ9Ihki6V9FUze8ms+SdJGlI+mfBPkr5oZlaY978l7SjM+/8lbZy1r1dJ+pqk90taJulWSd8ys8ZZ2367pD+V9CpJb5X0XUmbC8vXSXqfFsHMGiR9S9L3Jb1I0v8jaauZrZq12DslXSbpYEl3Fpa/V9JySadLer+ZdRaW/ZSkT7n7H0l6haSbCtNPKbw2F+6u+M/FxAkAAP4bCQcAAJJtm5k9LWmXpMclfXx6hrv/u7uPuvuUu98o6ReS1s5a9xF3/7y7T0raIuklko40s5dKOlHS/3D3P7j7Hcp/eZ/255K+4+4/cPcJSVdJapL0+lnLXOPuj7n7iKQ+ST9z90F3/72kbyj/+Md8Wgp3MUz/+zNJ7ZIOknSlu4+7+22Svi3pHbPW+6a7/9TdpyQdJ2mZu/9/heV/Lenzks4vLDsh6Y/N7Ah3f8bd+xcsZQAAsGgkHAAASLb17n6wpFMlHa1Zjz6Y2QVmds/0F3dJx2rfRyN+O/2Huz9X+PMgSS2Sfufuz85a9pFZf7fMfl/4gr9L+TsJpj026+/sHO8XGtxy1N2bZ/27qbDPXYV9zY5p9j53zfr7ZdovcaH8HRZHFua/W/m7Lx4ys7vM7KwF4gEAAEvACMwAAFQBd/+JmV2v/N0G683sZcr/j/7pkv7T3SfN7B5JNv9WZjwq6VAzO3BW0uGlkrzw96jydxBIkgqPYayQNBLEscxjVNIKM6ublXR4qaSfz1rGZ/29S9LD7v7KuTbm7r+Q9A4zq5N0jqSbzezw/bYBAADKwB0OAABUj09K+lMzO0HSgcp/ed4jSWb2LuXvcCjK3R+RNCDpUjNrNLM/UX4chmk3SXqLmZ1eGFvhHyT9QflxE8LyM0nPSfqwmTWY2amFmP5tnuW3S3razD5iZk1mVm9mx5rZiZJkZn9hZssKyYuxwjpTypfXlKSXh3coAADUBhIOAABUCXffI+kGSR9z9wcl/S9J/6n84wzHSfrpIjb3TuUHlXxS+XEhbpi1nyFJfyHpGkn/pfwX/7e6+3gAhzGnwrbfqvzAmP8l6V8lXeDuD82z/KSks5T/1Y6HC+t8QfkBNCXpTEkPmNkzyg8geb67ZwuPllwm6aeFRzHawzomAACqnblz5yAAAAAAAAgWdzgAAAAAAIDAkXAAAAAAAACBI+EAAAAAAAACR8IBAAAAAAAEjoQDAAAAAAAIHAkHAABqhJkNm9m4mR2x3/RBM3Mza91v+icK00/ab/qFZjZpZs/s968l4HivM7MhM5syswuLLPtPZrbLzJ4ys0fMbPOseR1zxOpm9vYg4wUAAPsi4QAAQG15WNI7pt+Y2XGSXrj/QmZmki6Q9GThdX//6e4H7fdvNOBY75X0f0u6u4RlvyjpaHf/I0mvl7TBzM6RJHfvmx2npLMkPSPpewHHCwAAZiHhAABAbfmK9k0gbJR0wxzLdUh6iaT3STrfzBorENs+3P0z7v4jSb8vYdkhd3921qQpSX88z+IbJd283/IAACBgJBwAAKgt/ZL+yMzSZlYv6XxJX51juY2SviXppsL7ty51h2Z2n5mNzfPvX5e63Tn2c4mZPSNpt6QDJf3vOZY5UNK5krYEtV8AADC3VNQBAACAipu+y+EnkjKSRmbPNLMXSjpP0gXuPmFmNxeW//qsxdrNbGzW+yfc/RVz7czdjw8w9nm5+5Vm9j8lrZa0XtLeORY7R9J/KX/sAAAgRNzhAABA7fmKpHdKulBzP07xNkk5SbcW3m+V9CYzWzZrmX53b571b85kQ6V53qCkrKRL51hko6Qb3N0rGxkAALWHhAMAADXG3R9RfvDIN0vqmWORjZIOkvQbM/utpH+X1KB8kmLRzOyBOX4lYvrftUs8jGJSkvZJgpjZCkmnau4kCwAACBiPVAAAUJveLelQd3/WzGY+D5jZckmnS3qTpPtmLf9+5R+r+NRid+Tur15KgIWBKuskmaQGMztA0ri7T+23XJ2kv1Z+vIkxSSdK+jtJV+y3yb+UdKe7/2op8QAAgMXhDgcAAGqQu//K3QfmmPWXku5x9++7+2+n/0n6tKTjzezYwnKvm+NuhRMDDvP7yj8a8XpJ1xX+PkWSzGyDmT0wa9m3SfqVpKeVHwTzmsK/2S4Qg0UCAFAxxiOMAAAAAAAgaNzhAAAAAAAAAkfCAQAAAAAABI6EAwAAAAAACBwJBwAAAAAAEDgSDgAAAAAAIHCp4otE74gjjvDW1taowwAAAAAAALPs2LHjv9x92VzzEpFwaG1t1cDAXD8VDgAAAAAAomJmj8w3j0cqAAAAAABA4Eg4AAAAAACAwJFwAAAAAAAAgUvEGA4AAAAAAMw2MTGh3bt36/e//33UodSEAw44QEcddZQaGhpKXoeEAwAAAAAgcXbv3q2DDz5Yra2tMrOow6lq7q4nnnhCu3fv1sqVK0tej0cqAAAAAACJ8/vf/16HH344yYYKMDMdfvjhi76bhIQDAAAAACCRSDZUzlLKOrSEg5l9ycweN7P755j3D2bmZnZEWPsHAGB/2wZHdPKVt2nlJd/RyVfepm2DI1GHBAAAEuqJJ57Q6tWrtXr1ar34xS/W8uXLZ96Pj49HEtOpp56qgYGBSPY9lzDHcLhe0r9IumH2RDNbIekMSb8Jcd8AAOxj2+CIunp2Kp3LaF19Rv170+rqyX8YWL9mecTRAQCApDn88MN1zz33SJI+8YlP6KCDDtKHPvShmfm5XE6pVG0PmxjaHQ7ufoekJ+eY9c+SPizJw9o3AAD76+4dUjqX0dbGy/XB1L9ra+PlSucy6u4dijo0AABQAZW40/HCCy/URRddpJNOOkkf/vCH9YlPfEJXXXXVzPxjjz1Ww8PDkqSvfvWrWrt2rVavXq2/+Zu/0eTk5D7b+t73vqfzzjtv5v3tt9+us846S5L0t3/7t2pra9OrX/1qffzjH58zloMOOmjm75tvvlkXXnihJGnPnj16+9vfrhNPPFEnnniifvrTn0qSfvKTn8zcobFmzRo9/fTTZZdHRcdwMLN1kkbc/d4Sln2vmQ2Y2cCePXsqEB0AoJqNjmXVXpdRg3JK2ZQalFN7XUajY9moQwMAACGbvtNxZCwrlzQyllVXz85Qkg67d+/WnXfeqauvvnreZTKZjG688Ub99Kc/1T333KP6+npt3bp1n2Xe+MY36mc/+5meffZZSdKNN96o888/X5J02WWXaWBgQPfdd59+8pOf6L777is5vosvvlgf+MAHdNddd+nrX/+63vOe90iSrrrqKn3mM5/RPffco76+PjU1NS320J+nYvd3mNkLJW1W/nGKotz9OknXSVJbWxt3QwAAytLS3KT+vWlNKCV5ThNKqX8qrZbm8i+mAAAg3rp7h5Sd2PcOguzEpLp7hwJ/tPK8885TfX39gsv86Ec/0o4dO3TiiSfmY8lm9aIXvWifZVKplM4880x961vf0rnnnqvvfOc7+qd/+idJ0k033aTrrrtOuVxOjz76qB588EEdf/zxJcX3wx/+UA8++ODM+6eeekrPPPOMTj75ZH3wgx/Uhg0bdM455+ioo45azGHPqZIPlLxC0kpJ9xZGtzxK0t1mttbdf1vBOAAANWhT5yp19Yxrw/hmtddl1D+VViaV1hWdq6IODQAAhGy+OxrDuNPxwAMPnPk7lUppampq5v30z0q6uzZu3KgrrrhiwW2df/75+pd/+Rcddthhamtr08EHH6yHH35YV111le666y4deuihuvDCC+f8ucrZvyoxe/7U1JT6+/t1wAEH7LP8JZdcore85S269dZbdfLJJ6u3t1dHH3304g5+PxV7pMLdd7r7i9y91d1bJe2W9BqSDQCASli/ZrmuOOc4PXbICfrs5Do9dsgJuuKc4xgwEgCAGjDfHY1h3+nY2tqqu+++W5J099136+GHH5YknX766br55pv1+OOPS5KefPJJPfLII89b/w1veIPuvvtuff7zn595nOKpp57SgQceqEMOOUSPPfaYvvvd78657yOPPFKZTEZTU1P6xje+MTP9jDPO0DXXXDPzfnrgy1/96lc67rjj9JGPfEQnnniiHnroobKPP7Q7HMzsa5JOlXSEme2W9HF3/2JY+wMAoJj1a5aTYAAAoAbl73Tcuc9jFU0N9doU8p2Ob3/723XDDTfo1a9+tU466SS96lWvkiQdc8wx+sd//EedccYZmpqaUkNDgz7zmc/oZS972T7r19fX66yzztL111+vLVu2SJJOOOEErVmzRkcffbRWrFihk08+ec59X3nllTrrrLO0bNkytbW16ZlnnpEkffrTn9bf/d3f6fjjj1cul9Mpp5yia6+9Vp/85Cf14x//WHV1dXr1q1+tN73pTWUfv7nHf3iEtrY2j9NviQIAAAAAopXJZJROp0teftvgiLp7hzQ6llVLc5M2da7iPyIWaa4yN7Md7t421/K1/aOgAAAAAICawJ2OlVfRn8UEAAAAAAC1gYQDAAAAAAAIHAkHAAAAAAAQOBIOAAAAAAAgcCQcAAAAAABA4Eg4AAAAAACwBPX19Vq9erWOPfZYnXfeeXruueeWvK0LL7xQN998syTpPe95jx588MF5l7399tt15513zry/9tprdcMNNyx532Eh4QAAAAAAwBI0NTXpnnvu0f3336/GxkZde+21+8zP5XJL2u4XvvAFHXPMMfPO3z/hcNFFF+mCCy5Y0r7CRMIBAAAAAIAydXR06Je//KVuv/12dXR06Oyzz9YxxxyjyclJbdq0SSeeeKKOP/54fe5zn5Mkubv+/u//XqtWrdIb3/hGPf744zPbOvXUUzUwMCBJ+t73vqfXvOY1OuGEE3T66adreHhY1157rf75n/9Zq1evVl9fnz7xiU/oqquukiTdc889am9v1/HHH6+3ve1t+t3vfjezzY985CNau3atXvWqV6mvr0+S9MADD2jt2rVavXq1jj/+eP3iF78IrExSgW0JQEVsGxxRd++QRseyamlu0qbOVVq/ZnnUYQEAgCrG5w9UjV3bpeE+qbVDWrE2sM3mcjl997vf1ZlnnilJuvvuu3X//fdr5cqVuu6663TIIYforrvu0h/+8AedfPLJOuOMMzQ4OKihoSE9+OCDeuyxx3TMMcfor/7qr/bZ7p49e/TXf/3XuuOOO7Ry5Uo9+eSTOuyww3TRRRfpoIMO0oc+9CFJ0o9+9KOZdS644AJdc801esMb3qCPfexjuvTSS/XJT35yJs7t27fr1ltv1aWXXqof/vCHuvbaa3XxxRdrw4YNGh8f1+TkZGDlQsIBSJBtgyPq6tmpdC6jdfUZ9e9Nq6tnXJK46AMAgFBMf/7ITuS/hIyMZdXVs1MSnz+QMLu2S1vOlibHpfpGaeMtZScdstmsVq9eLSl/h8O73/1u3XnnnVq7dq1WrlwpSfr+97+v++67b2Z8hr179+oXv/iF7rjjDr3jHe9QfX29WlpadNpppz1v+/39/TrllFNmtnXYYYctGM/evXs1NjamN7zhDZKkjRs36rzzzpuZf84550iSXvva12p4eFiS9LrXvU6XXXaZdu/erXPOOUevfOUrl14g+yHhACRId++Q0rmMtjZergblNKGUNoxvVndvIxd8AAAQiu7eoZlkw7TsxKS6e4f4/IFkGe7LJxt8Mv863Fd2wmF6DIf9HXjggTN/u7uuueYadXZ27rPMrbfeWta+l+IFL3iBpPxgl9PjS7zzne/USSedpO985zt685vfrM997nNzJj+WgjEcgAQZHcuqvS6jBuWUsik1KKf2uoxGx7JRhwYAAKrUfJ8z+PyBxGntyN/ZYPX519aOiuy2s7NTn/3sZzUxMSFJ+vnPf65nn31Wp5xyim688UZNTk7q0Ucf1Y9//OPnrdve3q477rhDDz/8sCTpySeflCQdfPDBevrpp5+3/CGHHKJDDz10ZnyGr3zlKzN3O8zn17/+tV7+8pfrfe97n9atW6f77ruvrOOdjTscgARpaW5S/960JpSSPH+HQ/9UWi3NTVGHBgAAqlRLc5NG5kgu8PkDibNibf4xihDGcFjIe97zHg0PD+s1r3mN3F3Lli3Ttm3b9La3vU233XabjjnmGL30pS/V6173uuetu2zZMl133XU655xzNDU1pRe96EX6wQ9+oLe+9a0699xz9c1vflPXXHPNPuts2bJFF110kZ577jm9/OUv15e//OUF47vpppv0la98RQ0NDXrxi1+szZs3B3bs5u6BbSwsbW1tPj1CJ1DLZo/h0F6XUf9UWplUWleccxy3NAIAgFDsP4aDJDU11PP5A5HLZDJKp9NRh1FT5ipzM9vh7m1zLc8dDkCCTF/Uu3sb9dmxV6mluUlXMEo0AAAI0X9//uBXKgAsDgkHIGHWr1nOBR4AAFQUnz8ALAWDRgIAAAAAgMCRcAAAAAAAJFISxiSsFkspaxIOAAAAAIDEOeCAA/TEE0+QdKgAd9cTTzyhAw44YFHrMYYDAAAAACBxjjrqKO3evVt79uyJOpSacMABB+ioo45a1DokHAAAAAAAidPQ0KCVK1dGHQYWwCMVAAAAAAAgcCQcAAAAAABA4Eg4AAAAAACAwJFwAAAAAAAAgQst4WBmXzKzx83s/lnTus3sITO7z8y+YWbNYe0fAAAAAABEJ8w7HK6XdOZ+034g6Vh3P17SzyV1hbh/LNG2wRGdfOVtWnnJd3Tylbdp2+BI1CEBQEXQ/wEAAAQntJ/FdPc7zKx1v2nfn/W2X9K5Ye0fS7NtcERdPTuVzmW0rj6j/r1pdfWMS5LWr1kecXQAEJ7p/i87MSlJGhnLqqtnpyT6P6Bidm2Xhvuk1g5pxdqoo8H+kl4/SY8fSKDQEg4l+CtJN0a4f8yhu3dI6VxGWxsvV4NymlBKG8Y3q7u3kQ/cAKpad+/QTLJhWnZiUt29Q/R/QCXs2i5tOVuaHJfqG6WNt/ClME6SXj9Jjx9IqEgGjTSzj0rKSdq6wDLvNbMBMxvYs2dP5YKrcaNjWbXXZdSgnFI2pQbl1F6X0ehYNurQACBU8/Vz9H9AhQz35b8M+mT+dbgv6ogwW9LrJ+nxAwlV8YSDmV0o6SxJG9zd51vO3a9z9zZ3b1u2bFnF4qt1Lc1N6p9Ka0Ip5bxOE0qpfyqtluamqEMDgFDN18/R/wEV0tqR/59nq8+/tnZEHRFmS3r9JD1+IKEq+kiFmZ0p6cOS3uDuz1Vy3yjNps5V6uoZ14bxzWqvy6h/Kq1MKq0rOldFHRoAhCrf/+3c57GKpoZ6baL/Aypjxdr8be48Yx9PSa+fpMcPJJQtcJNBeRs2+5qkUyUdIekxSR9X/lcpXiDpicJi/e5+UbFttbW1+cDAQChx4vm2DY6ou3dIo2NZtTQ3aVPnKp5fBlAT6P8AAAAWx8x2uHvbnPPCSjgEiYQDAAAAAADxs1DCIZJBIwEAAAAAQHUj4QAAAAAAAAJHwgEAAAAAAASOhAMAAAAAAAgcCQcAAAAAABA4Eg4AAAAAACBwJBwAAAAAAEDgSDgAAAAAAIDAkXAAAAAAAACBS0UdQC3aNjii7t4hjY5l1dLcpE2dq7R+zfKowwIARIzrAwCEg/4ViAYJhwrbNjiirp6dyk5MSpJGxrLq6tkpSXR6AFDDuD4AFbBruzTcJ7V2SCvWRh0NKoT+FYgOj1RUWHfv0ExnNy07Manu3qGIIgIAxAHXByBku7ZLW86Wbrss/7pre9QRoULoX4HokHCosNGx7KKmAwBqA9cHIGTDfdLkuOST+dfhvqgjQoXQvwLRIeFQYS3NTYuaDgCoDVwfgJC1dkj1jZLV519bO6KOCBVC/wpEh4RDhW3qXKWmhvp9pjU11GtT56qIIgIAxAHXByBkK9ZKG2+RTvto/pUxHGoG/SsQHQaNrLDpgWkYJRcAMBvXB6ACVqwl0VCD6F+B6Ji7Rx1DUW1tbT4wMBB1GAAAAAAAYBYz2+HubXPN45EKAAAAAAAQOBIOAAAAAAAgcCQcAAAAAABA4Eg4AAAAAACAwJFwAAAAAAAAgSPhAAAAAAAAAkfCAQAAAAAABI6EAwAAAAAACBwJBwAAAAAAELjQEg5m9iUze9zM7p817TAz+4GZ/aLwemhY+wcAAAAAANEJ8w6H6yWdud+0SyT9yN1fKelHhfcAAEiStg2O6OQrb9PKS76jk6+8TdsGR6IOCQAAAEsUWsLB3e+Q9OR+k9dJ2lL4e4uk9WHtHwCQLNsGR9TVs1NH7r1Xf1v/TR2591519ewk6RC0Xdulvv+VfwUAAAhRqsL7O9LdHy38/VtJR1Z4/wCAmOruHVI6l9HWxsvVoJwmlNKG8c3q7m3U+jXLow6vOuzaLm05W5ocl+obpY23SCvWRh0VAACoUpENGunuLsnnm29m7zWzATMb2LNnTwUjAwBEYXQsq/a6jBqUU8qm1KCc2usyGh3LRh1a9RjuyycbfDL/OtwXdUQAAKCKVTrh8JiZvUSSCq+Pz7egu1/n7m3u3rZs2bKKBQgAiEZLc5P6p9KaUEo5r9OEUuqfSquluSnq0KpHa0f+zgarz7+2dkQdEQAAqGKVfqTiFkkbJV1ZeP1mhfcPAIipTZ2r1NUzrg3jm9Vel1H/VFqZVFpXdK6KOrTqsWJt/jGK4b58soHHKQAAQIhCSziY2dcknSrpCDPbLenjyicabjKzd0t6RNKfhbV/AECyTI/T0N3bqM+OvUotzU26onMV4zcEbcVaEg0AAKAiLD+UQry1tbX5wMBA1GEAAAAAAIBZzGyHu7fNNS+yQSMBAAAAAED1IuEAAAAAAAACR8IBAAAAAAAEjoQDAAAAAAAIHAkHAAAAAAAQOBIOAAAAAAAgcCQcAAAAAABA4Eg4AAAAAACAwJFwAAAAAAAAgSPhAAAAAAAAApeKOgAkz7bBEXX3Dml0LKuW5iZt6lyl9WuWRx0WABU/P5N+/iY9/qSr9vKv9uOLu6jLP+r9lyvp8SP+aGNYChIOcbVruzTcJ7V2SCvWRh3NjG2DI+rq2ansxKQkaWQsq66enZJEhwNEbPr8TOcyWlefUf/etLp6xiXlz8+kn7/Fjg/hqvbyT/r5kXRRl3/S23fS48ciRPQdIepzFMlFwiGOdm2XtpwtTY5L9Y3Sxltik3To7h2a6WimZScm1d07RGcDRKy7d0jpXEZbGy9Xg3KaUEobxjeru7dR69csT/z5W+z4EK5qL/+knx9JF3X5J719Jz1+lCjC7whRn6NILsZwiKPhvnxH4pP51+G+qCOaMTqWXdR0AJUzOpZVe11GDcopZVNqUE7tdZmZ8zPp52+x40O4qr38k35+JF3U5Z/09p30+FGiCL8jRH2OIrlIOMRRa0c+a2n1+dfWjqgjmtHS3LSo6QAqp6W5Sf1TaU0opZzXaUIp9U+lZ87PpJ+/xY4P4ar28k/6+ZF0UZd/0tt30uNHiSL8jhD1OYrkIuEQRyvW5m+ROu2jsXqcQpI2da5SU0P9PtOaGuq1qXNVRBEBmLapc5UyqbQ2jG/W1bnztGF8szKp9Mz5mfTzt9jxIVzVXv5JPz+SLuryT3r7Tnr8KFGE3xGiPkeRXObuUcdQVFtbmw8MDEQdBgoYoRaIL36lAmGq9vKv9uOLu6jLP+r9lyvp8SP+aGOYj5ntcPe2OeeRcAAAAAAAAEuxUMKBRyoAAAAAAEDgSDgAAAAAAIDAkXAAAAAAAACBI+EAAAAAAAACR8IBAAAAAAAEjoQDAAAAAAAIHAkHAAAAAAAQOBIOAAAAAAAgcJEkHMzsA2b2gJndb2ZfM7MDoogDAAAAAACEI1XpHZrZcknvk3SMu2fN7CZJ50u6vtKxxNW2wRF19w5pdCyrluYmbepcpfVrlkcdFoAAcH4jTOW2L9pnuCjfaFH+KCbubYT4kEQVTzjM2m+TmU1IeqGk0YjiiM6u7dJwn9TaIa1YOzN52+CIunp2KjsxKUkaGcuqq2enJO17ws6zPoD4Kvn8Dhv9x8ISWj7ltq/p9dO5jNbVZ9S/N62unvGS1y9Z3Ms3pPhic/6HLab1WzPljyWL+2fw2LThcr/DoOZUPOHg7iNmdpWk30jKSvq+u3+/0nFEatd2acvZ0uS4VN8obbxl5oTt7h2aOVGnZScm1d079N8n6wLrA4ivks7vsNF/LCzB5VNu++ruHVI6l9HWxsvVoJwmlNKG8c3q7m0Mrn3GvXxDjC8W53/YYly/NVH+KEvcP4PHog2X+x0GNaniYziY2aGS1klaKalF0oFm9hdzLPdeMxsws4E9e/ZUOsxwDfflT1SfzL8O983MGh3LzrnKPtMXWB9AfJV0foeN/mNhCS6fctvX6FhW7XUZNSinlE2pQTm112WCbZ9xL98Q44vF+R+2GNdvTZQ/yhL3z+CxaMPlfodBTSqacDCzi83sjyzvi2Z2t5mdUcY+3yjpYXff4+4TknokvX7/hdz9Ondvc/e2ZcuWlbG7GGrtyGcFrT7/2toxM6uluWnOVfaZvsD6AOKrpPM7bPQfC0tw+ZTbvlqam9Q/ldaEUsp5nSaUUv9UOtj2GffyDTG+WJz/YYtx/dZE+aMscf8MHos2XO53GNSkUu5w+Ct3f0rSGZIOlfSXkq4sY5+/kdRuZi80M5N0uqRMGdtLnhVr87cgnfbR592KtalzlZoa6vdZvKmhXps6V5W0PoD4Kun8Dhv9x8ISXD7ltq9NnauUSaW1YXyzrs6dpw3jm5VJpYNtn3Ev3xDji8X5H7YY129NlD/KEvfP4LFow+V+h0FNMndfeAGz+9z9eDP7lKTb3f0bZjbo7muWvFOzSyX9uaScpEFJ73H3P8y3fFtbmw8MDCx1d4nDCK9A9eL8Rpj4lYp4o3yjRfmjmLi3EeJDXJnZDndvm3NeCQmHL0tarvyYCydIqlc+8fDaoAOdT60lHAAAAAAASIKFEg6l/ErFuyWtlvRrd3/OzA6X9K4A4wMAAAAAAFWmlDEcXNIxkt5XeH+gpANCiwgAAAAAACReKQmHf5X0OknvKLx/WtJnQosIAAAAAAAkXimPVJzk7q8xs0FJcvffmVljyHEBAAAAAIAEK+UOhwkzq1f+0QqZ2TJJU6FGBQAAAAAAEq2UhMOnJX1D0ovM7DJJ/yHp8lCjAgAAAAAAiVbKIxU3S9oh6XRJJmm9pMdCjAkAAAAAACRcKQmHHknr3f0hSTKzl0j6gaTXhhkYAAAAAABIrlIeqdgm6SYzqzezVkm9krrCDAoAAAAAACRb0Tsc3P3zhV+l2CapVdLfuPudIccFAAAAAAASbN6Eg5l9cPZbSS+VdI+kdjNrd/erQ44N89g2OKLu3iGNjmXV0tykTZ2rtH7N8qjDAiTRPhEt2t/CKB+gdsX9/I97fFIyYgTiZqE7HA7e733PPNNRQdsGR9TVs1PZiUlJ0shYVl09OyWJDi9Iu7ZLw31Sa4e0Ym3U0STGdPtM5zJaV59R/960unrGJS2yfZZb/nGvv2o/vogE1v6qFNePmOD8RQTi3j8moX9KQoxAHM2bcHD3SysZCErT3Ts009FNy05Mqrt3iM4uKLu2S1vOlibHpfpGaeMtfCgsUXfvkNK5jLY2Xq4G5TShlDaMb1Z3b2Pp7bPc8o97/VX78UUokPZXxbh+xADnLyIS9/4xCf1TEmIE4mjeQSPN7JOF12+Z2S37/6tYhNjH6Fh2UdOxBMN9+Q+DPpl/He6LOqLEGB3Lqr0uowbllLIpNSin9rrM4tpnueUf9/qr9uOLUCDtr4px/YgBzl9EJO79YxL6pyTECMTRQo9UfKXwelUlAkFpWpqbNDJHx9bS3BRBNFWqtSP/P0/T/wPV2hF1RInR0tyk/r1pTSglef5/UPqn0otrn+WWf9zrr9qPL0KBtL8qxvUjBjh/EZG4949J6J+SECMQR+buxRfK/0rF0ZJc0pC7j4cd2GxtbW0+MDBQyV3G1v7Pj0lSU0O9rjjnOG7nChLP2C7J7GdE2+sy6p9KK5NKL759VvsYB9V+fBEJrP1VKa4fMcH5iwjEvX9MQv+UhBiBqJjZDndvm3NesYSDmb1F0rWSfqX8r1WsVP6nMb8bdKDzIeGwL0bIRZzRPhEl2t/CKB+gdsX9/I97fFIyYgSiUG7C4SFJZ7n7LwvvXyHpO+5+dOCRzoOEAwAAAAAA8bNQwmHeQSNneXo62VDwa0lPBxIZAAAAAACoSvMOGmlm5xT+HDCzWyXdpPwYDudJuqsCsQEAAAAAgIRa6Fcq3jrr78ckvaHw9x5JB4QWEQAAAAAASLx5Ew7u/q5KBgIAAAAAAKpHKWM4AAAAAAAALAoJBwAAAAAAEDgSDgAAAAAAIHBFEw5m9v/O+vsF4YYDAAAAAACqwbwJBzP7iJm9TtK5syb/Z/ghAQAAAACApFvoDoeHJJ0n6eVm1mdmn5d0uJmtKnenZtZsZjeb2UNmlikkNgAAAAAAQJWY92cxJY1J2izp1MK/tKQzJF1iZqvc/fVl7PdTkr7n7ueaWaOkF5axLQAJsm1wRN29Qxody6qluUmbOldp/ZrlJc+vdZRPtCh/xFm57ZP2DZQn6edg1PtHdVoo4dAp6WOSXiHpakn3SXrW3d9Vzg7N7BBJp0i6UJLcfVzSeDnbBJAM2wZH1NWzU+lcRuvqM+rfm1ZXT/70X79m+cz87MSkJGlkLKuunp0z82sd5ROtYu23ZLu2S8N9UmuHtGJtSNGi1pTbPulfgPIEdQ6WfY1ZIq5xCMu8CQd33yxJZnavpK9Ieo2kZWb2H5J+5+5vXeI+V0raI+nLZnaCpB2SLnb3Z5e4PQAJ0d07pHQuo62Nl6tBOU0opQ3jm9Xd26j1a5aru3do5sPutOzEpLp7h/jAK1E+ESvWfkuya7u05Wxpclyqb5Q23sIHMgSi3PZJ/wKUJ4hzsOxrTITxS+IahzmV8rOYve4+4O7XSdrt7n8iqZy7HFLKJy8+6+5rJD0r6ZL9FzKz95rZgJkN7Nmzp4zdAYiL0bGs2usyalBOKZtSg3Jqr8todCw7M3++9UD5RK1Y+y3JcF/+g5hP5l+H+8ILGDWl3PZJ/wKUJ4hzsOxrTBm4xiEsRRMO7v7hWW8vLEz7rzL2uVv5xMXPCu9vVj4Bsf9+r3P3NndvW7ZsWRm7AxAXLc1N6p9Ka0Ip5bxOE0qpfyqtluammfnzrQfKJ2rF2m9JWjvy/+tj9fnX1o7wAkZNKbd90r8A5QniHCz7GlMGrnEIy0JjODyPu99b7g7d/bdmtqsw8OSQpNMlPVjudgHE36bOVerqGdeG8c1qr8uofyqtTCqtKzpXzZq/c5/bepsa6rWps+wfx6kKlE+0irXfkqxYm7/FlOdbEbBy2yf9C1CeYM7BMq8xZeAah7CYu1d+p2arJX1BUqOkX0t6l7v/br7l29rafGBgoELRAQgTv1JRHsonWpQ/4izpI+QDSZf0czDq/SO5zGyHu7fNOS+KhMNikXAAAAAAACB+Fko4lDJoJAAAAAAAwKKQcAAAAAAAAIEj4QAAAAAAAAJHwgEAAAAAAASOhAMAAAAAAAgcCQcAAAAAABA4Eg4AAAAAACBwJBwAAAAAAEDgSDgAAAAAAIDApaIOALVn2+CIunuHNDqWVUtzkzZ1rtL6Ncsrtv1y54cdX9wlPX4AqFVc38JV68cPYGG12keQcMDS7NouDfdJrR3SirUlr7ZtcERdPTuVnZiUJI2MZdXVs1OSAjnhprefzmW0rj6j/r1pdfWMz2y/1PlF44vp8Yct6fGjQpZ4flSNqI8/6v0jlopd/4Lafq1eH2r9+GfQ/0SvWB3EvY7iHt8Shd0HxxkJByzeru3SlrOlyXGpvlHaeEvJHUJ379DMxXhadmJS3b1DgZxs3b1DSucy2tp4uRqU04RS2jC+Wd29jVq/ZnlJ84vGF+PjD1vS40cFlHF+VIWojz/q/SO2il3/gth+LV8fav34JdH/xEGxOoh7HcU9vjKE3QfHGWM4YPGG+/IdgU/mX4f7Sl51dCy7qOmLNTqWVXtdRg3KKWVTalBO7XWZme2XMr9ofDE+/rAlPX5UQBnnR1WI+vij3j9iq9j1L4jtL2Z6tan145dE/xMHxeog7nUU9/jKEHYfHGckHLB4rR35rKPV519bO0petaW5aVHTF6uluUn9U2lNKKWc12lCKfVPpWe2X8r8ovHF+PjDlvT4UQFlnB9VIerjj3r/iK1i178gtr+Y6dWm1o9fEv1PHBSrg7jXUdzjK0PYfXCcmbtHHUNRbW1tPjAwEHUYmC2gMQwkqamhXlecc1zgYzi012XUP5VWJpWe2X6p84vGF9PjD1vS40eFVOnzlyWL+vij3j9iqdj1L6jt1+r1odaPfwb9T/QYwyGWwu6Do2ZmO9y9bc55JBxQaVGPks2vVJQn6fEDQK3i+hauWj9+AAur5j6ChAMAAAAAAAjcQgkHxnAAAAAAAACBI+EAAAAAAAACR8IBAAAAAAAEjoQDAAAAAAAIHAkHAAAAAAAQOBIOAAAAAAAgcCQcAAAAAABA4Eg4AAAAAACAwJFwAAAAAAAAgYss4WBm9WY2aGbfjioGAAAAAAAQjlSE+75YUkbSH0UYA7Bo2wZH1N07pNGxrFqam7Spc5XWr1kedVg1o9zyp/7KU6z8ar18KR+UI+7tg/4XC6F+o5f0a1Dc48PSRJJwMLOjJL1F0mWSPhhFDFVv13ZpuE9q7ZBWrI06mqqxbXBEXT07lZ2YlCSNjGXV1bNTkhbXIca9fmIaX7nlXzP1F5Lp8kvnMlpXn1H/3rS6esYl5csvsPJNqGLHX+vlg4UVO7+iFlT/G9fjK1mx/j/s60NMrz9V1b/FtIyLSfo1KO7xYemiusPhk5I+LOngiPZf3XZtl7acLU2OS/WN0sZbEtVhxll379BMRzgtOzGp7t6h0jvDuNdPjOMrt/xrov5C1N07pHQuo62Nl6tBOU0opQ3jm9Xd26j1a5YHU74JVuz4a718sLBi51fUguh/43x8JSnW/4d9fYjx9adq+rcYl3ExSb8GxT0+LF3Fx3Aws7MkPe7uO4os914zGzCzgT179lQouiox3JfvKH0y/zrcF3VEVWN0LLuo6XOKe/3EOL5yy78m6i9Eo2NZtddl1KCcUjalBuXUXpeZKb9AyjfBih1/rZcPFlbs/IpaEP1vnI+vJMX6/7CvDzG+/lRN/xbjMi4m6deguMeHpYti0MiTJZ1tZsOS/k3SaWb21f0Xcvfr3L3N3duWLVtW6RiTrbUjn5W1+vxra0fUEVWNluamRU2fU9zrJ8bxlVv+NVF/IWppblL/VFoTSinndZpQSv1T6ZnyC6R8E6zY8dd6+WBhxc6vqAXR/8b5+EpSrP8P+/oQ4+tP1fRvMS7jYpJ+DYp7fFi6iicc3L3L3Y9y91ZJ50u6zd3/otJxVLUVa/O3gJ320UTdCpYEmzpXqamhfp9pTQ312tS5qvSNxL1+YhxfueVfE/UXok2dq5RJpbVhfLOuzp2nDeOblUmlZ8ovkPJNsGLHX+vlg4UVO7+iFkT/G+fjK0mx/j/s60OMrz9V07/FuIyLSfo1KO7xYenM3aPbudmpkj7k7mcttFxbW5sPDAxUJCagGEbQjRajpEcr6SNgh43yQTni3j7of7EQ6jd6Sb8GxT0+zM/Mdrh725zzokw4lIqEAwAAAAAA8bNQwiGKMRwAAAAAAECVI+EAAAAAAAACR8IBAAAAAAAEjoQDAAAAAAAIHAkHAAAAAAAQOBIOAAAAAAAgcCQcAAAAAABA4Eg4AAAAAACAwJFwAAAAAAAAgUtFHQCwWNsGR9TdO6TRsaxampu0qXOV1q9ZHnVYVSPp5Rt2/MW2X+7+k17+QDXj/KxtUdd/1PsHgKUg4YBo7NouDfdJrR3SirUlr7ZtcERdPTuVnZiUJI2MZdXVs1OS4nXRXeLxBbb+ErefmPKdR9jxF9t+ufufXj+dy2hdfUb9e9Pq6hkPLH4kRNjnf9JFVD4ln99Jr7+kxx+SqK+PUe8fSAz6sNgh4YDK27Vd2nK2NDku1TdKG28puUPo7h2audhOy05Mqrt3KD4X3DKOL5D1y9h+Isp3AWHHX2z75e6/u3dI6VxGWxsvV4NymlBKG8Y3q7u3MRHljwCEff4nXYTlU9L5nfT6S3r8IYr6+hj1/oFEoA+LJcZwQOUN9+U7Ap/Mvw73lbzq6Fh2UdMjUcbxBbJ+GdtPRPkuIOz4i22/3P2PjmXVXpdRg3JK2ZQalFN7XSYx5Y8AhH3+J12E5VPS+Z30+kt6/CGK+voY9f6BRKAPiyUSDqi81o581tHq86+tHSWv2tLctKjpkSjj+AJZv4ztJ6J8FxB2/MW2X+7+W5qb1D+V1oRSynmdJpRS/1Q6MeWPAIR9/iddhOVT0vmd9PpLevwhivr6GPX+gUSgD4slEg6ovBVr87c4nfbRRd/qtKlzlZoa6veZ1tRQr02dq4KOcunKOL5A1i9j+4ko3wWEHX+x7Ze7/02dq5RJpbVhfLOuzp2nDeOblUmlE1P+CEDY53/SRVg+JZ3fSa+/pMcfoqivj1HvH0gE+rBYMnePOoai2trafGBgIOowEBOM0hyupJcvv1IBICycn7Ut6vqPev8AMB8z2+HubXPOI+EAAAAAAACWYqGEA49UAAAAAACAwJFwAAAAAAAAgSPhAAAAAAAAAkfCAQAAAAAABI6EAwAAAAAACBwJBwAAAAAAEDgSDgAAAAAAIHAkHAAAAAAAQOBIOAAAAAAAgMBVPOFgZivM7Mdm9qCZPWBmF1c6BgAAAAAAEK5UBPvMSfoHd7/bzA6WtMPMfuDuD0YQSyi2DY6ou3dIo2NZtTQ3aVPnKq1fszzqsCqm1o+/XOWWH+UPoFbR/1U36jdclC+wMM6Rpal4wsHdH5X0aOHvp80sI2m5pKpIOGwbHFFXz05lJyYlSSNjWXX17JSkmmiQFTv+Xdul4T6ptUNasTa47Uas3PKr9fYXmHLbV9LbZ9LjR7QiOn+m+790LqN19Rn1702rq2dc0n79H+07kUq+vlG/SxKbzw/F6o/6RURtIDbnSAJFcYfDDDNrlbRG0s+ijCNI3b1DMw1xWnZiUt29QzXRGCty/Lu2S1vOlibHpfpGaeMtVXPRKbf8ar39BaLc9pX09pn0+BGtCM+f7t4hpXMZbW28XA3KaUIpbRjfrO7exv/u/2jfiVXS9Y36XbJYfH4oVn/ULyJsA7E4RxIqskEjzewgSV+X9H53f2qO+e81swEzG9izZ0/lA1yi0bHsoqZXm4oc/3BfvqPxyfzrcF9w245YueVX6+0vEOW2r6S3z6THj2hFeP6MjmXVXpdRg3JK2ZQalFN7XWbf/o/2nVglXd+o3yWLxeeHYvVH/SLCNhCLcyShIkk4mFmD8smGre7eM9cy7n6du7e5e9uyZcsqG2AZWpqbFjW92lTk+Fs78llNq8+/tnYEt+2IlVt+td7+AlFu+0p6+0x6/IhWhOdPS3OT+qfSmlBKOa/ThFLqn0rv2//RvhOrpOsb9btksfj8UKz+qF9E2AZicY4klLl7ZXdoZpK2SHrS3d9fyjptbW0+MDAQalxB2f/5HklqaqjXFeccVxO321Ts+Kv0Gb5yy6/W219gGMMh2fEjWjEYw6G9LqP+qbQyqfTz+z/adyKVfH2jfpckNp8fGMMBxcRkDAeJz9izmdkOd2+bc14ECYc/kdQnaaekqcLkze5+63zrJCnhIDGCaa0ff7n4lQoAWBr6v+pG/YaL8gUWxjkyv1glHJYiaQkHAAAAAABqwUIJh8gGjQQAAAAAANWLhAMAAAAAAAgcCQcAAAAAABA4Eg4AAAAAACBwJBwAAAAAAEDgSDgAAAAAAIDAkXAAAAAAAACBI+EAAAAAAAACR8IBAAAAAAAEjoQDAAAAAAAIXCrqAIBqs21wRN29Qxody6qluUmbOldp/ZrlUYeFElF/iLO4t89i8ZUbf9yPP2y1fvxJF/b5AVS7ar/GRL3/sJBwCMuu7dJwn9TaIa1YG3U0zxf3+BJq2+CIunp2KjsxKUkaGcuqq2enJFVFhzEj6vYT0v5rpv6iFnX7Sai4t89i8ZUbf2yOP6L2O3386VxG6+oz6t+bVlfPuKR41H/FJLT/KFZ/1G8NSWgbjlpg15h5yr/c9Uuev8TjSzJz96hjKKqtrc0HBgaiDqN0u7ZLW86WJsel+kZp4y3x6lDiHl+CnXzlbRoZyz5v+vLmJv30ktMiiCgEUbefEPdfE/UXtajbT4LFvX0Wi6/c+GNx/BG235OvvE1H7r1XWxsvV4NymlBKG8Y367FDTohF/VdEgvuPYvVH/daIBLfhqAVyjVmg/Mtdv6T5ZRxf3JnZDndvm2seYziEYbgv39B8Mv863Bd1RPuKe3wJNjpHR7HQ9ESKuv2EuP+aqL+oRd1+Eizu7bNYfOXGH4vjj7D9jo5l1V6XUYNyStmUGpRTe10mNvVfEQnuP4rVH/VbIxLchqMWyDVmgfIvd/2S5i8gFte4kJBwCENrRz6rZfX519aOqCPaV9zjS7CW5qZFTU+kqNtPiPuvifqLWtTtJ8Hi3j6LxVdu/LE4/gjbb0tzk/qn0ppQSjmv04RS6p9Kx6b+KyLB/Uex+qN+a0SC23DUArnGLFD+5a5f0vwFxOIaFxIeqQhL3J/Pint8CbX/81eS1NRQryvOOS7xz1/tI+r2U6ExHKQqrb+oRd1+Eiru7bNYfOXGH5vjj8EYDu11GfVPpZVJpWNT/xWT0P6jWP1RvzUkoW04aoFdY0ocw2Gx65c8f4nHF3cLPVJBwgEIWLWOMFsrqD/EWdzbZ7WPIB61Wj/+pONXKoDyVPs1Jur9l4OEAwAAAAAACByDRgIAAAAAgIoi4QAAAAAAAAJHwgEAAAAAAASOhAMAAAAAAAgcCQcAAAAAABA4Eg4AAAAAACBwJBwAAAAAAEDgSDgAAAAAAIDAkXAAAAAAAACBS0WxUzM7U9KnJNVL+oK7XxlFHNVq2+CIunuHNDqWVUtzkzZ1rtL6NctLnh+1uMeXdJQvFhJ2+6D9YSFJb3+072hR/vFWbv0k/fOtlIwYwxT18XMNiEbFEw5mVi/pM5L+VNJuSXeZ2S3u/mClY6lG2wZH1NWzU9mJSUnSyFhWXT07JUnr1ywvOj8wu7ZLw31Sa4e0Yu2i40/nMlpXn1H/3rS6esaDj6/azVP+Fav/uFti+6x2YbcP2l9AqrT9Jr391Uz7jmn7q5nyD1tI9Vvu57tSP9/G+fNjrbfRqOuo5PIv8ztM2fUb0z62HFE8UrFW0i/d/dfuPi7p3yStiyCOqtTdOzTT0KdlJybV3TtU0vxA7NoubTlbuu2y/Ouu7SWv2t07pHQuo62Nl+uDqX/X1sbLlc5lgo2v2i1Q/hWp/7gro31Wu7DbB+0vAFXcfpPe/mqifce4/dVE+YctxPot9/NdKZ9v4/75sdbbaNR1VFL5l/kdpuz6jXEfW44oEg7LJe2a9X53Ydo+zOy9ZjZgZgN79uypWHBJNzqWXXB6sfmBGO6TJscln8y/DveVvOroWFbtdRk1KKeUTalBObXXZYKNr9otUP4Vqf+4K6N9Vruw2wftLwBV3H6T3v5qon3HuP3VRPmHLcT6LffzXSmfb+P++bHW22jUdVRS+Zf5HWYx0+cU4z62HLEdNNLdr3P3NndvW7ZsWdThJEZLc9OC04vND0Rrh1TfKFl9/rW1o+RVW5qb1D+V1oRSynmdJpRS/1Q62Piq3QLlX5H6j7sy2me1C7t90P4CUMXtN+ntrybad4zbX02Uf9hCrN9yP9+V8vk27p8fa72NRl1HJZV/md9hFjN9TjHuY8sRRcJhRNKKWe+PKkxDADZ1rlJTQ/0+05oa6rWpc1VJ8wOxYq208RbptI/mXxfx/NGmzlXKpNLaML5ZV+fO04bxzcqk0sHGV+0WKP+K1H/cldE+q13Y7YP2F4Aqbr9Jb3810b5j3P5qovzDFmL9lvv5rpTPt3H//FjrbTTqOiqp/Mv8DlN2/ca4jy2HuXtld2iWkvRzSacrn2i4S9I73f2B+dZpa2vzgYGBCkWYfEkfxTfu8SUd5YuFMIIzopT09kf7jhblH2/8SkUyYgxT1MfPNSA8ZrbD3dvmnFfphIMkmdmbJX1S+Z/F/JK7X7bQ8iQcAAAAAACIn4USDhX/WUxJcvdbJd0axb4BAAAAAED4YjtoJAAAAAAASC4SDgAAAAAAIHAkHAAAAAAAQOBIOAAAAAAAgMCRcAAAAAAAAIEj4QAAAAAAAAJHwgEAAAAAAATO3D3qGIoysz2SHok6jiU4QtJ/RR0EahbtD1Gi/SFKtD9EifaHqNEGUWkvc/dlc81IRMIhqcxswN3boo4DtYn2hyjR/hAl2h+iRPtD1GiDiBMeqQAAAAAAAIEj4QAAAAAAAAJHwiFc10UdAGoa7Q9Rov0hSrQ/RIn2h6jRBhEbjOEAAAAAAAACxx0OAAAAAAAgcCQcQmBmZ5rZkJn90swuiToeVDczW2FmPzazB83sATO7uDD9MDP7gZn9ovB6aNSxonqZWb2ZDZrZtwvvV5rZzwr94I1m1hh1jKheZtZsZjeb2UNmljGz19EHolLM7AOF6+/9ZvY1MzuAPhBhMbMvmdnjZnb/rGlz9neW9+lCO7zPzF4TXeSoVSQcAmZm9ZI+I+lNko6R9A4zOybaqFDlcpL+wd2PkdQu6e8Kbe4SST9y91dK+lHhPRCWiyVlZr3/n5L+2d3/WNLvJL07kqhQKz4l6XvufrSkE5Rvi/SBCJ2ZLZf0Pklt7n6spHpJ54s+EOG5XtKZ+02br797k6RXFv69V9JnKxQjMIOEQ/DWSvqlu//a3ccl/ZukdRHHhCrm7o+6+92Fv59W/oP2cuXb3ZbCYlskrY8kQFQ9MztK0lskfaHw3iSdJunmwiK0P4TGzA6RdIqkL0qSu4+7+5joA1E5KUlNZpaS9EJJj4o+ECFx9zskPbnf5Pn6u3WSbvC8fknNZvaSigQKFJBwCN5ySbtmvd9dmAaEzsxaJa2R9DNJR7r7o4VZv5V0ZFRxoep9UtKHJU0V3h8uaczdc4X39IMI00pJeyR9ufBYzxfM7EDRB6IC3H1E0lWSfqN8omGvpB2iD0Rlzdff8b0EkSPhAFQJMztI0tclvd/dn5o9z/M/R8NP0iBwZnaWpMfdfUfUsaBmpSS9RtJn3X2NpGe13+MT9IEIS+FZ+XXKJ75aJB2o59/uDlQM/R3ihoRD8EYkrZj1/qjCNCA0ZtagfLJhq7v3FCY/Nn3bXOH18ajiQ1U7WdLZZjas/CNkpyn/PH1z4fZiiX4Q4dotabe7/6zw/mblExD0gaiEN0p62N33uPuEpB7l+0X6QFTSfP0d30sQORIOwbtL0isLoxM3Kj9w0C0Rx4QqVnhe/ouSMu5+9axZt0jaWPh7o6RvVjo2VD9373L3o9y9Vfn+7jZ33yDpx5LOLSxG+0No3P23knaZ2arCpNMlPSj6QFTGbyS1m9kLC9fj6fZHH4hKmq+/u0XSBYVfq2iXtHfWoxdARVj+rhsEyczerPwzzfWSvuTul0UbEaqZmf2JpD5JO/Xfz9BvVn4ch5skvVTSI5L+zN33H2QICIyZnSrpQ+5+lpm9XPk7Hg6TNCjpL9z9DxGGhypmZquVH7S0UdKvJb1L+f9UoQ9E6MzsUkl/rvyvRg1Keo/yz8nTByJwZvY1SadKOkLSY5I+Lmmb5ujvCkmwf1H+MZ/nJL3L3QciCBs1jIQDAAAAAAAIHI9UAAAAAACAwJFwAAAAAAAAgSPhAAAAAAAAAkfCAQAAAAAABI6EAwAAAAAACBwJBwAA8DxmducS11tvZscEHQ8AAEgeEg4AAOB53P31S1x1vSQSDgAAgIQDAAB4PjN7pvB6qpndbmY3m9lDZrbVzKww70oze9DM7jOzq8zs9ZLOltRtZveY2SvM7K/N7C4zu9fMvm5mLyyse72ZfdrM7jSzX5vZubP2/REz21lY58rCtFeY2ffMbIeZ9ZnZ0YXp55nZ/YVl76h0OQEAgPmlog4AAADE3hpJr5Y0Kumnkk42s4ykt0k62t3dzJrdfczMbpH0bXe/WZLMbMzdP1/4+x8lvVvSNYXtvkTSn0g6WtItkm42szdJWifpJHd/zswOKyx7naSL3P0XZnaSpH+VdJqkj0nqdPcRM2sOuRwAAMAikHAAAADFbHf33ZJkZvdIapXUL+n3kr5oZt+W9O151j22kGholnSQpN5Z87a5+5SkB83syMK0N0r6srs/J0nu/qSZHSTp9ZL+vXBzhSS9oPD6U0nXm9lNknrKPE4AABAgEg4AAKCYP8z6e1JSyt1zZrZW0umSzpX098rfcbC/6yWtd/d7zexCSafOs13T/Ookjbn76v1nuPtFhTse3iJph5m91t2fKHpEAAAgdIzhAAAAFq1w18Eh7n6rpA9IOqEw62lJB89a9GBJj5pZg6QNJWz6B5LeNWush8Pc/SlJD5vZeYVpZmYnFP5+hbv/zN0/JmmPpBUBHB4AAAgACQcAALAUB0v6tpndJ+k/JH2wMP3fJG0ys0Eze4Wk/yHpZ8o/+vBQsY26+/eUH89hoPD4xocKszZIereZ3SvpAeXHeZDyA1TuNLP7Jd0p6d4gDg4AAJTP3D3qGAAAAAAAQJXhDgcAAAAAABA4Eg4AAAAAACBwJBwAAAAAAEDgSDgAAAAAAIDAkXAAAAAAAACBI+EAAAAAAAACR8IBAAAAAAAEjoQDAAAAAAAI3P8BiI3Y5hjAyk0AAAAASUVORK5CYII=\n","text/plain":["<Figure size 1296x360 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["y_pred = RdFor.predict(X_val)\n","potential_models.append(('random_forest', RdFor, mae(y_val, y_pred).numpy()))\n","plot_results(y_pred, title='Random Forest')"]},{"cell_type":"markdown","id":"2d16e422","metadata":{"papermill":{"duration":0.03449,"end_time":"2022-01-10T01:34:34.768342","exception":false,"start_time":"2022-01-10T01:34:34.733852","status":"completed"},"tags":[]},"source":["## SVM - One vs. Rest\n","\n","Another algorithm we investigated for classification is Support Vector Machine. Since this is an inherently binary classification algorithm, we tested two techniques for multiclass classification: \n","- the efficient [One vs. Rest](https://scikit-learn.org/stable/modules/generated/sklearn.multiclass.OneVsRestClassifier.html),\n","- the more costly [One vs. One](https://scikit-learn.org/stable/modules/generated/sklearn.multiclass.OneVsOneClassifier.html)."]},{"cell_type":"code","execution_count":21,"id":"c69a7746","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:34.842474Z","iopub.status.busy":"2022-01-10T01:34:34.841835Z","iopub.status.idle":"2022-01-10T01:34:34.89385Z","shell.execute_reply":"2022-01-10T01:34:34.893291Z","shell.execute_reply.started":"2022-01-09T23:56:07.329206Z"},"papermill":{"duration":0.090632,"end_time":"2022-01-10T01:34:34.893999","exception":false,"start_time":"2022-01-10T01:34:34.803367","status":"completed"},"tags":[]},"outputs":[{"data":{"text/plain":["Pipeline(steps=[('minmax_scaler', MinMaxScaler()),\n","                ('svm_ovr',\n","                 OneVsRestClassifier(estimator=LinearSVC(random_state=12)))])"]},"execution_count":21,"metadata":{},"output_type":"execute_result"}],"source":["from sklearn.multiclass import OneVsRestClassifier\n","from sklearn.svm import LinearSVC\n","\n","# SVM1 = OneVsRestClassifier(LinearSVC(random_state=12))\n","SVM1 = Pipeline([(\"minmax_scaler\", MinMaxScaler()), (\"svm_ovr\", OneVsRestClassifier(LinearSVC(random_state=12)))])\n","\n","SVM1.fit(X_train, y_train)"]},{"cell_type":"code","execution_count":22,"id":"4542e8ab","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:34.970286Z","iopub.status.busy":"2022-01-10T01:34:34.969704Z","iopub.status.idle":"2022-01-10T01:34:35.16557Z","shell.execute_reply":"2022-01-10T01:34:35.166052Z","shell.execute_reply.started":"2022-01-09T23:56:07.389111Z"},"papermill":{"duration":0.23611,"end_time":"2022-01-10T01:34:35.166225","exception":false,"start_time":"2022-01-10T01:34:34.930115","status":"completed"},"tags":[]},"outputs":[{"data":{"image/png":"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\n","text/plain":["<Figure size 1296x360 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["y_pred = SVM1.predict(X_val)\n","potential_models.append(('svm_ovr', SVM1, mae(y_val, y_pred).numpy()))\n","plot_results(y_pred, title='SVM - OvR')"]},{"cell_type":"markdown","id":"c74af5b8","metadata":{"papermill":{"duration":0.036361,"end_time":"2022-01-10T01:34:35.23838","exception":false,"start_time":"2022-01-10T01:34:35.202019","status":"completed"},"tags":[]},"source":["## SVM - One vs. One"]},{"cell_type":"code","execution_count":23,"id":"61a131de","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:35.320835Z","iopub.status.busy":"2022-01-10T01:34:35.319704Z","iopub.status.idle":"2022-01-10T01:34:35.469185Z","shell.execute_reply":"2022-01-10T01:34:35.468298Z","shell.execute_reply.started":"2022-01-09T23:56:07.701112Z"},"papermill":{"duration":0.195233,"end_time":"2022-01-10T01:34:35.469391","exception":false,"start_time":"2022-01-10T01:34:35.274158","status":"completed"},"tags":[]},"outputs":[{"data":{"text/plain":["Pipeline(steps=[('minmax_scaler', MinMaxScaler()),\n","                ('svm_ovo',\n","                 OneVsOneClassifier(estimator=SVC(random_state=12)))])"]},"execution_count":23,"metadata":{},"output_type":"execute_result"}],"source":["from sklearn.multiclass import OneVsOneClassifier\n","from sklearn.svm import SVC\n","\n","# SVM2 = OneVsOneClassifier(SVC(random_state=12))\n","SVM2 = Pipeline([(\"minmax_scaler\", MinMaxScaler()), (\"svm_ovo\", OneVsOneClassifier(SVC(random_state=12)))])\n","\n","SVM2.fit(X_train, y_train)"]},{"cell_type":"code","execution_count":24,"id":"12aa60c8","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:35.560917Z","iopub.status.busy":"2022-01-10T01:34:35.560255Z","iopub.status.idle":"2022-01-10T01:34:35.889895Z","shell.execute_reply":"2022-01-10T01:34:35.888703Z","shell.execute_reply.started":"2022-01-09T23:56:07.904185Z"},"papermill":{"duration":0.374821,"end_time":"2022-01-10T01:34:35.890042","exception":false,"start_time":"2022-01-10T01:34:35.515221","status":"completed"},"tags":[]},"outputs":[{"data":{"image/png":"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\n","text/plain":["<Figure size 1296x360 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["y_pred = SVM2.predict(X_val)\n","potential_models.append(('svm_ovo', SVM2, mae(y_val, y_pred).numpy()))\n","plot_results(y_pred, title='SVM - OvO')"]},{"cell_type":"markdown","id":"992a365b","metadata":{"papermill":{"duration":0.038185,"end_time":"2022-01-10T01:34:35.966106","exception":false,"start_time":"2022-01-10T01:34:35.927921","status":"completed"},"tags":[]},"source":["## Poisson regression\n","\n","An interresting remark is that the number of bikes at a station (sometimes) appears to follow a Poisson distribution, with the station more often empty than full. This justifies envisioning the Poisson regression as a model for individual stations."]},{"cell_type":"code","execution_count":25,"id":"dd0736d8","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:36.078234Z","iopub.status.busy":"2022-01-10T01:34:36.068155Z","iopub.status.idle":"2022-01-10T01:34:36.268705Z","shell.execute_reply":"2022-01-10T01:34:36.269158Z","shell.execute_reply.started":"2022-01-09T23:56:08.261673Z"},"papermill":{"duration":0.265331,"end_time":"2022-01-10T01:34:36.26935","exception":false,"start_time":"2022-01-10T01:34:36.004019","status":"completed"},"tags":[]},"outputs":[{"data":{"image/png":"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\n","text/plain":["<Figure size 432x288 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["plt.hist(y_train)\n","plt.title('Number of bikes at station '+str(STATION_ID));"]},{"cell_type":"code","execution_count":26,"id":"30ce49d6","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:36.360383Z","iopub.status.busy":"2022-01-10T01:34:36.356483Z","iopub.status.idle":"2022-01-10T01:34:36.364715Z","shell.execute_reply":"2022-01-10T01:34:36.364154Z","shell.execute_reply.started":"2022-01-09T23:56:08.530471Z"},"papermill":{"duration":0.056386,"end_time":"2022-01-10T01:34:36.364853","exception":false,"start_time":"2022-01-10T01:34:36.308467","status":"completed"},"tags":[]},"outputs":[{"data":{"text/plain":["Pipeline(steps=[('minmax_scaler', MinMaxScaler()),\n","                ('pois_reg', PoissonRegressor())])"]},"execution_count":26,"metadata":{},"output_type":"execute_result"}],"source":["from sklearn.linear_model import PoissonRegressor\n","\n","# Poisreg = PoissonRegressor()\n","Poisreg = Pipeline([(\"minmax_scaler\", MinMaxScaler()), (\"pois_reg\", PoissonRegressor())])\n","\n","Poisreg.fit(X_train, y_train)"]},{"cell_type":"code","execution_count":27,"id":"c79570dd","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:36.446631Z","iopub.status.busy":"2022-01-10T01:34:36.445995Z","iopub.status.idle":"2022-01-10T01:34:36.691185Z","shell.execute_reply":"2022-01-10T01:34:36.690725Z","shell.execute_reply.started":"2022-01-09T23:56:08.54857Z"},"papermill":{"duration":0.288292,"end_time":"2022-01-10T01:34:36.691323","exception":false,"start_time":"2022-01-10T01:34:36.403031","status":"completed"},"tags":[]},"outputs":[{"data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAABBwAAAFcCAYAAABr+63OAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8/fFQqAAAACXBIWXMAAAsTAAALEwEAmpwYAABDDklEQVR4nO3de3gdZbn//89NEmwEJBwqmlJsZUugYGlxUdDK4Qd8CQpCQdiiuKEqIntvt2wPUYp7K14/Eb47eET3xnoARH4Ku9aIcggIAlXkkDaVAiWgUKQph0oNxyVN0vv3x1qJaUjWSjIza07v13X1StfMWjP3PPPMMzP3euZZ5u4CAAAAAAAI0zZxBwAAAAAAALKHhAMAAAAAAAgdCQcAAAAAABA6Eg4AAAAAACB0JBwAAAAAAEDoSDgAAAAAAIDQkXAAACADzOwBMzs87jjSyMxOM7Ob4o4DAICsMXePOwYAAFBmZusk7SZpUNJLkm6Q9HF3fzHOuAAAACaLHg4AACTPe9x9e0kHSCpI+o+Y45kyM6tP8vIAAEB0SDgAAJBQ7t6rUg+H/STJzI4vPzrRZ2a3mdk+Q+81s3VmdlT5/wvMrMvMnjezp83sa+Xp08zsx2b2bHkZ95rZbuV5zWZ2rZltMrM/mtlHRyz7fDO7xsx+ZGYvlGMojBe3mbmZ/auZPSLpkfK048xsdXm9d5rZ3BHvP8DMusvL/l8zu9rMvlyed7iZrTezz5nZU5IuM7NtzOxcM/tTeVuuMbOdJ7CNi83s0fJ6HjOz00ZM/+2IeN5R/txz5b/vGDHvNjP7f83sd+Xl3GRmu051HwMAkGUkHAAASCgzmynp3ZK6zWwvST+R9O+Spku6XtIvzWzbMT76TUnfdPfXSdpT0jXl6WdI2lHSTEm7SDpbUrE876eS1ktqlnSypK+Y2REjlnl8+T1Nkq6V9O0q4S+SdJCkOWY2X9IPJX2svN7vSrrWzF5Tjv/nki6XtHN5G08ctaw3lOe9SdJZkv6tvPzDyvH+VdJ3Km2jmW0n6VuS3uXuO0h6h6TVo4MuJy6uK793F0lfk3Sdme0y4m0fkPQhSa+XtK2kz1QpCwAAcomEAwAAydNhZn2SfivpdklfkfQ+Sde5+83u3i/pYkmNKt04j9Yv6R/MbFd3f9Hd7xoxfRdJ/+Dug+6+0t2fLyc2Fkr6nLv/zd1XS/q+pNNHLPO37n69uw9KulLS/lW24UJ33+TuRZWSBN9197vL671C0iuSDi7/q5f0LXfvd/flku4Ztawtkr7o7q+Ul3e2pM+7+3p3f0XS+ZJOLj9uMeY2jljOfmbW6O5PuvsDY8R9rKRH3P1Kdx9w959IekjSe0a85zJ3f7gcyzWS5lUpCwAAcomEAwAAybPI3Zvc/U3u/i/lG9tmSY8PvcHdt0h6QtKMMT7/EUl7SXqo/EjAceXpV0rqlPRTM9tgZv9lZg3lZW9y9xdGLOPxUct+asT/X5Y0rcp4Ck+M+P+bJH26/IhDXzmZMrO83mZJvb71KNYjPytJG939b6OW9/MRy1qr0iCbu423je7+kkpJm7MlPWlm15nZ3mPEvVU5l1Uri+3HLgIAAPKNhAMAAOmwQaUbbUmSmZlKN+29o9/o7o+4+/tV6vL/fyUtM7Ptyj0IvuTuc1TqGXGcSr0YNkja2cx2GLGYPcZa9iSMTiBcUE6iDP17bbn3wJOSZpS3Z8jMCssaWt67Ri1vmrv3VthGuXunu/8fSW9UqdfC98aIe6tyLgtaFgAA5BIJBwAA0uEaScea2ZHlXgmfVumxhDtHv9HMPmhm08u9IPrKk7eY2f9jZm81szpJz6v0+MEWd3+ivJwLy4MuzlWpl8SPQ4r9e5LONrODrGQ7Mzu2nOD4vUq9Ez5uZvVmdoKkBVWWd6mkC8zsTeXtnV7+nMbbRjPbzcxOKI/l8IqkF1V6xGK06yXtZWYfKMfzPklzJP0qcCkAAJAzJBwAAEgBd++R9EFJl0j6i0pjCrzH3TeP8fZjJD1gZi+qNIDkqeXHMt4gaZlKN+JrVRof4sryZ94vaZZK3/D/XKUxE34dUuxdkj6q0kCTf5X0R0mLy/M2SzpJpQRHX3kbf6VSUmA831Rp4MqbzOwFSXepNEClNP42biPpU+Xt26TSgJP/PEasz6rUK+LTkp6V9FlJx7n7X6aw6QAA5Jpt/cgkAABAvMzsbkmXuvtlcccCAACmjh4OAAAgVmZ2mJm9ofwIwxmS5kq6Me64AABAMJVGlwYAAKiFFpXGqNhO0qOSTnb3J+MNCQAABMUjFQAAAAAAIHQ8UgEAAAAAAEJHwgEAAAAAAISOhAMAADlhZuvMbLOZ7TpqereZuZnNGjX9/PL0g0ZNX2xmg2b24qh/zSHGupeZ/cLMNprZJjPrNLOWKp85ysxWmdlLZrbezP5xjPecXt6mM8OKFQAAjI2EAwAA+fKYpPcPvTCzt0p67eg3mZlJOl3SpvLf0X7v7tuP+rchxDibJF2r0oCSu0m6R9Ivxnuzmc2R9P9J+rykHSXtL2nlqPfsJOk8SQ+EGCcAABgHCQcAAPLlSm2dQDhD0o/GeN8hkt4o6ROSTjWzbWsQ2zB3v8fdf+Dum9y9X9LXJbWY2S7jfOQ/JH3X3W9w9wF3f9bd/zTqPRdK+pakv0QYOgAAKCPhAABAvtwl6XVmto+Z1Uk6VdKPx3jfGZJ+qdLPVUrSe6a6QjO7z8z6xvn33xNczKGSnnL3Z8eZf3B5XWvM7Ekz+7GZ7TwihgWSCpIunep2AACAyamPOwAAAFBzQ70cbpe0VlLvyJlm9lpJp0g63d37zWxZ+f0/G/G2g82sb8TrZ919z7FW5u5zgwRrZrtL+o6kT1V42+6S/knS0ZI2SLpC0iWSTisnVv5b0sfdfUvpaREAABA1Eg4AAOTPlZLukDRbYz9OcaKkAUnXl19fJenXZjbd3TeWp93l7u+MOlAzmy7pJkn/7e4/qfDWoqTL3P3h8ue+IunX5Xn/Iuk+d78r0mABAMBWeKQCAICccffHVRo88t2Slo/xljMkbS/pz2b2lKT/ldQg6QNTWZ+ZPTDGL1oM/Rv3EYfyII83SbrW3S+ospr7JPmI1yP/f6SkE83sqfL2vEPSV83s21PZHgAAMDH0cAAAIJ8+Imknd3/JzIavB8xshko36O9S6SZ+yL+r9FjFNye7Inffd7KfMbPXSeqU9Dt3P3cCH7lM0n+a2Y8lPSXpXEm/Ks9bLGnaiPcul7RM0g8mGxcAAJg4ejgAAJBD7v4nd+8aY9Y/SVrt7je5+1ND/1T6dYe5ZrZf+X1vH6O3woEhhniipAMlfWjUOvaQJDM7zcyGf97S3X+o0uMhd0t6XNIrKv3Chty9b9S2bJb0vLs/F2K8AABgFHP36u8CAAAAAACYBHo4AAAAAACA0JFwAAAAAAAAoSPhAAAAAAAAQkfCAQAAAAAAhI6EAwAAAAAACF199bfEb9ddd/VZs2bFHQYAAAAAABhh5cqVf3H36WPNS0XCYdasWerqGuunwgEAAAAAQFzM7PHx5vFIBQAAAAAACB0JBwAAAAAAEDoSDgAAAAAAIHSpGMMBAAAAAICR+vv7tX79ev3tb3+LO5RcmDZtmnbffXc1NDRM+DMkHAAAAAAAqbN+/XrtsMMOmjVrlsws7nAyzd317LPPav369Zo9e/aEP8cjFQAAAACA1Pnb3/6mXXbZhWRDDZiZdtlll0n3JiHhAAAAAABIJZINtTOVso4s4WBmPzSzZ8zs/jHmfdrM3Mx2jWr9AACM1tHdq4UX3arZ516nhRfdqo7u3rhDAgAAKfXss89q3rx5mjdvnt7whjdoxowZw683b94cS0yHH364urq6Yln3WKIcw+FySd+W9KORE81spqSjJf05wnUDALCVju5eLVm+RsX+QUlSb19RS5avkSQtmj8jztAAAEAK7bLLLlq9erUk6fzzz9f222+vz3zmM8PzBwYGVF+f72ETI+vh4O53SNo0xqyvS/qsJI9q3QAAjNbe2TOcbBhS7B9Ue2dPTBEBAIBaqkVPx8WLF+vss8/WQQcdpM9+9rM6//zzdfHFFw/P32+//bRu3TpJ0o9//GMtWLBA8+bN08c+9jENDm59nXLjjTfqlFNOGX5922236bjjjpMk/fM//7MKhYL23XdfffGLXxwzlu233374/8uWLdPixYslSRs3btR73/teHXjggTrwwAP1u9/9TpJ0++23D/fQmD9/vl544YXA5VHTMRzM7ARJve7+hwm89ywz6zKzro0bN9YgOgBAlm3oK05qOgAAyI6hno69fUW5/t7TMYqkw/r163XnnXfqa1/72rjvWbt2ra6++mr97ne/0+rVq1VXV6errrpqq/ccddRRuvvuu/XSSy9Jkq6++mqdeuqpkqQLLrhAXV1duu+++3T77bfrvvvum3B855xzjj75yU/q3nvv1c9+9jOdeeaZkqSLL75Y3/nOd7R69WqtWLFCjY2Nk930V6lZ/w4ze62k81R6nKIqd18qaakkFQoFekMAAAJpbmpU7xjJheam4CdTAACQbJV6Oob9aOUpp5yiurq6iu+55ZZbtHLlSh144IGlWIpFvf71r9/qPfX19TrmmGP0y1/+UieffLKuu+46/dd//Zck6ZprrtHSpUs1MDCgJ598Ug8++KDmzp07ofh+/etf68EHHxx+/fzzz+vFF1/UwoUL9alPfUqnnXaaTjrpJO2+++6T2ewx1fKBkj0lzZb0h/LolrtLWmVmC9z9qRrGAQDIobbWlq3GcJCkxoY6tbW2xBgVAACohVr2dNxuu+2G/19fX68tW7YMvx76WUl31xlnnKELL7yw4rJOPfVUffvb39bOO++sQqGgHXbYQY899pguvvhi3Xvvvdppp520ePHiMX+ucuSvSoycv2XLFt11112aNm3aVu8/99xzdeyxx+r666/XwoUL1dnZqb333ntyGz9KzR6pcPc17v56d5/l7rMkrZd0AMkGAEAtLJo/Qxee9FbNaGqUSZrR1KgLT3orA0YCAJAD4/VojLqn46xZs7Rq1SpJ0qpVq/TYY49Jko488kgtW7ZMzzzzjCRp06ZNevzxx1/1+cMOO0yrVq3S9773veHHKZ5//nltt9122nHHHfX000/rhhtuGHPdu+22m9auXastW7bo5z//+fD0o48+Wpdccsnw66GBL//0pz/prW99qz73uc/pwAMP1EMPPRR4+yPr4WBmP5F0uKRdzWy9pC+6+w+iWh8AANUsmj+DBAMAADkUV0/H9773vfrRj36kfffdVwcddJD22msvSdKcOXP05S9/WUcffbS2bNmihoYGfec739Gb3vSmrT5fV1en4447TpdffrmuuOIKSdL++++v+fPna++999bMmTO1cOHCMdd90UUX6bjjjtP06dNVKBT04osvSpK+9a1v6V//9V81d+5cDQwM6NBDD9Wll16qb3zjG/rNb36jbbbZRvvuu6/e9a53Bd5+c0/+8AiFQsGT9FuiAAAAAIB4rV27Vvvss8+E39/R3av2zh5t6CuqualRba0tfBExSWOVuZmtdPfCWO/P94+CAgAAAABygZ6OtVfTn8UEAAAAAAD5QMIBAAAAAACEjoQDAAAAAAAIHQkHAAAAAAAQOhIOAAAAAAAgdCQcAAAAAACYgrq6Os2bN0/77befTjnlFL388stTXtbixYu1bNkySdKZZ56pBx98cNz33nbbbbrzzjuHX1966aX60Y9+NOV1R4WEAwAAAAAAU9DY2KjVq1fr/vvv17bbbqtLL710q/kDAwNTWu73v/99zZkzZ9z5oxMOZ599tk4//fQprStKJBwAAAAAAAjokEMO0R//+EfddtttOuSQQ3T88cdrzpw5GhwcVFtbmw488EDNnTtX3/3udyVJ7q6Pf/zjamlp0VFHHaVnnnlmeFmHH364urq6JEk33nijDjjgAO2///468sgjtW7dOl166aX6+te/rnnz5mnFihU6//zzdfHFF0uSVq9erYMPPlhz587ViSeeqL/+9a/Dy/zc5z6nBQsWaK+99tKKFSskSQ888IAWLFigefPmae7cuXrkkUdCK5P60JYEoCY6unvV3tmjDX1FNTc1qq21RYvmz4g7LAAAkGFcfyAznrhHWrdCmnWINHNBaIsdGBjQDTfcoGOOOUaStGrVKt1///2aPXu2li5dqh133FH33nuvXnnlFS1cuFBHH320uru71dPTowcffFBPP/205syZow9/+MNbLXfjxo366Ec/qjvuuEOzZ8/Wpk2btPPOO+vss8/W9ttvr8985jOSpFtuuWX4M6effrouueQSHXbYYfrCF76gL33pS/rGN74xHOc999yj66+/Xl/60pf061//WpdeeqnOOeccnXbaadq8ebMGBwdDKxcSDkCKdHT3asnyNSr2lxqB3r6ilixfI0mc9AEAQCS4/kBmPHGPdMXx0uBmqW5b6YxrAycdisWi5s2bJ6nUw+EjH/mI7rzzTi1YsECzZ8+WJN1000267777hsdneO655/TII4/ojjvu0Pvf/37V1dWpublZRxxxxKuWf9ddd+nQQw8dXtbOO+9cMZ7nnntOfX19OuywwyRJZ5xxhk455ZTh+SeddJIk6W1ve5vWrVsnSXr729+uCy64QOvXr9dJJ52kt7zlLVMvkFF4pAJIkfbOnuGT/ZBi/6DaO3tiiggAAGQd1x/IjHUrSskGHyz9Xbci8CKHxnBYvXq1LrnkEm277baSpO222274Pe6uSy65ZPh9jz32mI4++ujA656K17zmNZJKg10OjS/xgQ98QNdee60aGxv17ne/W7feemto6yPhAKTIhr7ipKYDAAAExfUHMmPWIaWeDVZX+jvrkJqstrW1Vf/zP/+j/v5+SdLDDz+sl156SYceeqiuvvpqDQ4O6sknn9RvfvObV3324IMP1h133KHHHntMkrRp0yZJ0g477KAXXnjhVe/fcccdtdNOOw2Pz3DllVcO93YYz6OPPqo3v/nN+sQnPqETTjhB9913X6DtHYlHKoAUaW5qVO8YJ/fmpsYYogEAAHnA9QcyY+aC0mMUEYzhUMmZZ56pdevW6YADDpC7a/r06ero6NCJJ56oW2+9VXPmzNEee+yht7/97a/67PTp07V06VKddNJJ2rJli17/+tfr5ptv1nve8x6dfPLJ+sUvfqFLLrlkq89cccUVOvvss/Xyyy/rzW9+sy677LKK8V1zzTW68sor1dDQoDe84Q0677zzQtt2c/fQFhaVQqHgQyN0Ank2+hlKSWpsqNOFJ72VZygBAEAkuP5AUq1du1b77LNP3GHkylhlbmYr3b0w1vvp4QCkyNBJnVGiAQBArXD9AWCqSDgAKbNo/gxO8AAAoKa4/gAwFQwaCQAAAAAAQkfCAQAAAACQSmkYkzArplLWJBwAAAAAAKkzbdo0PfvssyQdasDd9eyzz2ratGmT+hxjOAAAAAAAUmf33XfX+vXrtXHjxrhDyYVp06Zp9913n9RnSDgAAAAAAFKnoaFBs2fPjjsMVMAjFQAAAAAAIHQkHAAAAAAAQOhIOAAAAAAAgNCRcAAAAAAAAKGLLOFgZj80s2fM7P4R09rN7CEzu8/Mfm5mTVGtHwAAAAAAxCfKHg6XSzpm1LSbJe3n7nMlPSxpSYTrxxR1dPdq4UW3ava512nhRbeqo7s37pAAoCZo/wAAAMIT2c9iuvsdZjZr1LSbRry8S9LJUa0fU9PR3asly9eo2D8oSertK2rJ8jWSpEXzZ8QZGgBEivYPAAAgXHGO4fBhSTfEuH6Mob2zZ/hie0ixf1DtnT0xRQQAtUH7BwAAEK5YEg5m9nlJA5KuqvCes8ysy8y6Nm7cWLvgcm5DX3FS0wEgK2j/AAAAwlXzhIOZLZZ0nKTT3N3He5+7L3X3grsXpk+fXrP48q65qXFS0wEgK2j/AAAAwlXThIOZHSPps5KOd/eXa7luTExba4saG+q2mtbYUKe21paYIgKA2qD9AwAACFdkg0aa2U8kHS5pVzNbL+mLKv0qxWsk3WxmknSXu58dVQyYvKGB0do7e7Shr6jmpka1tbYwYBqAzKP9AwAACJdVeKohMQqFgnd1dcUdBgAAAAAAGMHMVrp7Yax5cf5KBQAAAAAAyCgSDgAAAAAAIHQkHAAAAAAAQOhIOAAAAAAAgNCRcAAAAAAAAKEj4QAAAAAAAEJHwgEAAAAAAISOhAMAAAAAAAgdCQcAAAAAABC6+rgDyKOO7l61d/ZoQ19RzU2Namtt0aL5M+IOCwAQM84PABAN2lcgHiQcaqyju1dLlq9RsX9QktTbV9SS5WskiUYPAHKM8wMARIP2FYgPj1TUWHtnz3BjN6TYP6j2zp6YIgIAJAHnBwCIBu0rEB8SDjW2oa84qekAgHzg/AAA0aB9BeJDwqHGmpsaJzUdAJAPnB8AIBq0r0B8SDjUWFtrixob6raa1thQp7bWlpgiAgAkAecHAIgG7SsQHwaNrLGhgWkYJRcAMBLnBwCIBu0rEB9z97hjqKpQKHhXV1fcYQAAAAAAgBHMbKW7F8aaxyMVAAAAAAAgdCQcAAAAAABA6Eg4AAAAAACA0JFwAAAAAAAAoSPhAAAAAAAAQkfCAQAAAAAAhI6EAwAAAAAACB0JBwAAAAAAEDoSDgAAAAAAIHSRJRzM7Idm9oyZ3T9i2s5mdrOZPVL+u1NU6wcAAAAAAPGJsofD5ZKOGTXtXEm3uPtbJN1Sfg0AgCSpo7tXCy+6VbPPvU4LL7pVHd29cYcEAACAKYos4eDud0jaNGryCZKuKP//CkmLolo/ACBdOrp7tWT5GvX2FeWSevuKWrJ8DUkHAACAlKr1GA67ufuT5f8/JWm3Gq8fAJBQ7Z09KvYPbjWt2D+o9s6emCICAABAELENGunuLsnHm29mZ5lZl5l1bdy4sYaRAQDisKGvOKnpAAAASLZaJxyeNrM3SlL57zPjvdHdl7p7wd0L06dPr1mAAIB4NDc1Tmo6AAAAkq3WCYdrJZ1R/v8Zkn5R4/UDABKqrbVFjQ11W01rbKhTW2tLTBEBAAAgiPqoFmxmP5F0uKRdzWy9pC9KukjSNWb2EUmPS/rHqNYPAEiXRfNnSCqN5bChr6jmpka1tbYMTwcAAEC6WGkohWQrFAre1dUVdxgAAAAAAGAEM1vp7oWx5sU2aCQAAAAAAMguEg4AAAAAACB0JBwAAAAAAEDoSDgAAAAAAIDQkXAAAAAAAAChI+EAAAAAAABCR8IBAAAAAACEjoQDAAAAAAAIHQkHAAAAAAAQOhIOAAAAAAAgdPVxB4D06ejuVXtnjzb0FdXc1Ki21hYtmj8j7rAAqPrxmfbjN+3xp13Wyz/r25d0cZd/3OsPKu3xI/moY5gKEg6YlI7uXi1ZvkbF/kFJUm9fUUuWr5EkGhwgZtWOz7Qfv2mPP+2yXv5Z376ki7v8415/UGmPH8lHHcNU8UgFJqW9s2e4oRlS7B9Ue2dPTBEBGFLt+Ez78Zv2+NMu6+Wf9e1LurjLP+71B5X2+JF81DFMFQkHTMqGvuKkpgOonWrHZ9qP37THn3ZZL/+sb1/SxV3+ca8/qLTHj+SjjmGqSDhgUpqbGic1HUDtVDs+0378pj3+tMt6+Wd9+5Iu7vKPe/1BpT1+JB91DFNFwgGT0tbaosaGuq2mNTbUqa21JaaIAAypdnym/fhNe/xpl/Xyz/r2JV3c5R/3+oNKe/xIPuoYpopBIzEpQ4PCMEItkDzVjs+0H79pjz/tsl7+Wd++pIu7/ONef1Bpjx/JRx3DVJm7xx1DVYVCwbu6uuIOAwAAAAAAjGBmK929MNY8HqkAAAAAAAChI+EAAAAAAABCR8IBAAAAAACEjoQDAAAAAAAIHQkHAAAAAAAQOhIOAAAAAAAgdCQcAAAAAABA6Eg4AAAAAACA0MWScDCzT5rZA2Z2v5n9xMymxREHAAAAAACIRn2tV2hmMyR9QtIcdy+a2TWSTpV0ea1jSaqO7l61d/ZoQ19RzU2Namtt0aL5M+IOC0AIOL4RpaD1i/oZLco3XpQ/qkl6HSE+pFHNEw4j1ttoZv2SXitpQ0xxJE5Hd6+WLF+jYv+gJKm3r6gly9dIEgcskHIc34hS0PpF/YwW5Rsvyh/VJL2OEB/SquaPVLh7r6SLJf1Z0pOSnnP3m2odR1K1d/YMH6hDiv2Dau/siSkiAGHh+EaUgtYv6me0KN94Uf6oJul1hPiQVjVPOJjZTpJOkDRbUrOk7czsg2O87ywz6zKzro0bN9Y6zNhs6CtOajqA9OD4RpSC1i/qZ7Qo33hR/qgm6XWE+JBWVRMOZnaOmb3OSn5gZqvM7OgA6zxK0mPuvtHd+yUtl/SO0W9y96XuXnD3wvTp0wOsLl2amxonNR1AenB8I0pB6xf1M1qUb7wof1ST9DpCfEirifRw+LC7Py/paEk7SfonSRcFWOefJR1sZq81M5N0pKS1AZaXKW2tLWpsqNtqWmNDndpaW2KKCEBYOL4RpaD1i/oZLco3XpQ/qkl6HSE+pNVEBo208t93S7rS3R8oJwqmxN3vNrNlklZJGpDULWnpVJeXNUODqjDCK5A9HN+IUtD6Rf2MFuUbL8of1SS9jhAf0srcvfIbzC6TNEOlMRf2l1Qn6TZ3f1v04ZUUCgXv6uqq1eoAAAAAAMAEmNlKdy+MNW8iPRw+ImmepEfd/WUz20XSh0KMDwAAAAAAZMxExnBwSXMkfaL8ejtJ0yKLCAAAAAAApN5EEg7/Lentkt5ffv2CpO9EFhEAAAAAAEi9iTxScZC7H2Bm3ZLk7n81s20jjgsAAAAAAKTYRHo49JtZnUqPVsjMpkvaEmlUAAAAAAAg1SaScPiWpJ9Ler2ZXSDpt5K+EmlUAAAAAAAg1SbySMUySSslHSnJJC2S9HSEMQEAAAAAgJSbSMJhuaRF7v6QJJnZGyXdLOltUQYGAAAAAADSayKPVHRIusbM6sxslqROSUuiDAoAAAAAAKRb1R4O7v698q9SdEiaJelj7n5nxHEBAAAAAIAUGzfhYGafGvlS0h6SVks62MwOdvevRRwbxtHR3av2zh5t6CuqualRba0tWjR/RtxhAZKon4gX9a8yygfIr6Qf/0mPT0pHjEDSVOrhsMOo18vHmY4a6uju1ZLla1TsH5Qk9fYVtWT5GkmiwUPsqJ+IE/WvMsoHyK+kH/9Jj09KR4xAEo07hoO7f6nSv1oGib9r7+wZbuiGFPsH1d7ZE1NEwN9RPxEn6l9llA+QX6Ed/0/cI634aulviNLQPqUhRiCJKj1S8Q13/3cz+6UkHz3f3Y+PNDKMaUNfcVLTgVqifiJO1L/KKB8gv0I5/p+4R7rieGlws1S3rXTGtdLMBcmJL2JpiBFIokqPVFxZ/ntxLQLBxDQ3Nap3jIatuakxhmiArVE/ESfqX2WUD5BfoRz/61aUkg0+WPq7bkVoCYc0tE9piBFIokqPVKws/71d0u8l/VXSJkm/L09DDNpaW9TYULfVtMaGOrW1tsQUEfB31E/EifpXGeUD5Fcox/+sQ0o9G6yu9HfWIcmKL2JpiBFIoqo/i2lmx0q6VNKfVPq1itlm9jF3vyHq4PBqQ4PSMEIukoj6iThR/yqjfID8CuX4n7mg9BjFuhWlZENIvRtCiy9iaYgRSCJzf9XwDFu/wewhSce5+x/Lr/eUdJ27712D+CRJhULBu7q6arU6AAAAAAAwAWa20t0LY80b95GKEV4YSjaUPSrphVAiAwAAAAAAmVTpVypOKv+3y8yul3SNSr9WcYqke2sQGwAAAAAASKlKYzi8Z8T/n5Z0WPn/GyVNiywiANU9cU8kz1ACAAAAQFjGTTi4+4dqGQiACYrwd7ABAAAAICwTGcMBefTEPdKKr5b+IlnG+h1sAAAAAEiYqj+LiRziG/RkG/od7KH9E+LvYAMAAABAWEg44NXG+gadhENyRPg72AAAAAAQlqoJBzP7D3f/cvn/r3H3V6IPC7HiG/Tkm7mARAMAAACARBt3DAcz+5yZvV3SySMm/z76kHIiyWMkDH2DfsTneZwCQPYkuf0FAADIkEo9HB6SdIqkN5vZivLrXcysxd17gqzUzJokfV/SfpJc0ofdPT/JjDSMkcA36EByVftZVH42dXxpaH8BAAAyolLCoU/SeZIOL//bR9LRks4tJx3eEWC935R0o7ufbGbbSnptgGWlD2MkIMc6unvV3tmjDX1FNTc1qq21RYvmz5jw/Ly7/ZbrdNCKxar3AQ1Yve4+5HIdduSxf38DN9SVBWx/c1E/SVilVtD6mYv6DUQo7cdg3OtHNlVKOLRK+oKkPSV9TdJ9kl5y9w8FWaGZ7SjpUEmLJcndN0vaHGSZqcMYCcipju5eLVm+RsX+QUlSb19RS5avkSQtmj+j6vy86+ju1aO3X6uFNqB62yL5gFbefq3+uvO8v5cPCc3KArS/uaifJKxSK2j9zEX9BiJUs2MwoqQwbQCiMu4YDu5+nrsfKWmdpCsl1Umabma/NbNfBljnbEkbJV1mZt1m9n0z2y7A8tKHMRKQU+2dPcMnsiHF/kG1d/ZMaH7etXf26Lf9e6tf9RrwbdSvev22f++ty2fohtrqSGiOJUD7m4v6OVbCCqkQtH7mon4DEarJMTiUFL71gtLfEMciog1AVCbys5id7t4lqcvM/tnd32lmuwZc5wGS/s3d7zazb0o6V9J/jnyTmZ0l6SxJ2mOPPQKsLqEYIwE5tKGvWHF6tfl5t6GvqF7tpdM2n6eDt1mru7bso1W+l2xk+fCzqdVNsf3NRf2kB15qBa2fuajfQISPjNXkGIywFyNtAKJSNeHg7p8d8XJxedpfAqxzvaT17n53+fUylRIOo9e7VNJSSSoUCh5gfdnEM7ZIoeamRvWOceJqbmqc0Py8GyqfVb6XVg3utdX0rZDQjEQu6icJq9QKWj9zUb+RbxE/MlaTYzDCpDBtAKIy7iMVY3H3PwRdobs/JekJM2spTzpS0oNBl5srEXanAqLU1tqixoa6raY1NtSprbVlQvPzLjflk9CfrcxN+c9cIB3yaZINKRO0fiamfif0+EcGRPzIWE2OwQgfy05MG4DMmcgjFVH4N0lXlX+h4lFJgQaizB0GhUNKDQ06NN4IyNXm510uyifBgxbmovyRWkHrZyLqd4KPf2RAxI+M1ewYDNqLcZxe0oloA5BJ5p78pxUKhYJ3dXXFHUZycEJG0vHID6ZqxVdLvbd8sDTw5RGfL33bPhL1K16UP6IykeMfCCLv7Rf3EIiIma1098JY8+Lq4YAgeMY2uLyfcKLEyQxBVPsGivoVL8o/OM4/42PQ0ujlvf7lfYyjMHpJ570OYdJIOKRV3hvMILhgjhaP/CCIaglV6le8KP9gJnL+ifpiPsk3C7X4QiXJ2x81rn/CkeY6FDSpRx3CFJBwQP5wwRwtvqFCUJUSqrWoX2m+mIwax3cw1c4/UV/Mp+FmIcovVMJI+KS5fajF9U+ay2ci0nAMVRI0qcc1NKaAhENWZb3BD4IL5mjxyE/ypbl9iLp+JeEb6CTLw/Ed5f6tdv6J+mI+7zcLQRM+ab/ZjPr6J+3lMxFZSNoESepNpA7l+RyJMZFwyKI8NPhBhHHBTGNaGY/8JFcY7UPc9T/K+hX3N9BpkOXjO+r9W+38E/UNYd57CAVN+KQ9YRN1wjDt5TMReU/aVKtDSY8/CZLcRkaEhENSBamMWci+Ri3IBTONKdIsaPuQhvofpH2K+xvoWkh7+x2lWuzfSuefqG8Ik9BDKE5BEz5Z6CEZZcIwC+VTTRqSNnH2kOAepLKkt5ERIeGQREErY96zr1HLwg1H0qX5ZJJ0QduHpNf/oO1T3N9Ah6HS8ZP39ruaJOzfqHuQxNlDKAmCJHzy8EhREHkpnyQnbeLuxcg9SGVpaCMjQMIhiYJWxjRkX6MW5zO4CCbtJ5OkC9o+JL3+h9E+xfkNdFDVjp80tN9xSvr+Tbqktw8TUe1mMu5HipKekA9aPknfvqgFbYPi7sXIPUhlWWgjp4CEQxKFURmTnH2NWtzP4CKYtJ9M0iBI+5D0+l+L9inuG45Kqh0/SWi/k35DkeT9m3QTaR+Svv+TLOsJ+axsX9A6HqQNSkIvxjzfg1ST9GuoiJBwSKKkV8aA8XV096q9s0cb+opqbmpUW2uLFs2fEVp4D9x5nVoGXlG9tmhg4BX13Hmd9n3f32Ostv4JxRegMY16+6MWefxpP5nkQZJvyJLefkat2vETd/lk5YYiySrc7ETdfpeW/6I29O2t5qYX1dbau/XyM77/Iz8/Zj0hn4Xti7uOZ6EXY6WETdznsICqtpEZRcIhqZJ8QS9NOb6O7l4tWb5Gxf5BSVJvX1FLlq+RpFAOuI7uXl19/y764Tb1kgbUr3p9+f5d9L69Sgd0tfXXIr4olx+1msQfxsmEb9DyLentZ5T1cyLHT5zlk4UbiiSrcLMTWvs9Tv2d0PIzvP9rcn5Mws1glLKwfUmo42nuxTiRhE3Sz/HjSPs9QBDbxB0A8qW9s2f4QBtS7B9Ue2dPaMv/ff+eOm3zefrawCk6bfN5+n3/nsPLr7b+ofkH2MP6l7pf6AB7eGrxPXGPtOKrpb+j4oty+6MWWvzjlM+wmQukQz499WTDFcdLt15Q+jveOoA41KJ+Bjl+ojZ0Q2F16b2hSLKxbnbKQmm/K9TfCS0/w/u/Juf3oZvBIz4//jfn1c6vSTaR7Uu6LNTxOM8hFdqwtEv7PUAQ9HBATW3oK05q+lSXv8r30qrBvV41vdr6N/QVdYA9rKu2/Yoayj0kTtt8nrr79hrzc2OqkJ0dWs8B9rAO3mat7tqyj1b5XqFtf9RC2X9RdzdMwrcL1dADI7/SUD+jFPe3Z1lX4RviUNrvCvV3QsvP8P6v2fm90re7cXfnD0MSvr0Oco7OcB2viSz0chlH1PdASUbCATXV3NSo3jEOrOamxposfyLzD35hrRo0oHrbIvmADt5mrZ7eYf+JB1Hhgqy5qVG7PfeHVyU0nt5xEsuPUSj7L+obrqSfrLJwQYipS3r9rIUk3FBkVYWbnVDa7wr1d8LLz+j+T8T5Pe8JzTCEcY7OaB2viQwnbKK+B0oyHqlATbW1tqixoW6raY0NdWprbanJ8icyv3ub/dSveg34NupXvbq32W9y8VXoTtfW2qJ3Njw0nNBo0IDe2fBQaNsftVD2X9TdDZPeJXMi3QXT3CUWlSW9fiL9xukOHUr7XaH+Rn1+T7pEnN+z0J0/bhnu0p8aSX4sMIA8t5H0cEA0xumONjQoStVRnKfYna3a8ic2/70654Z6/cNLq/XH7ebpfe86YXKDuVTIzi6aP0O3bzpeAyuWSz6gAavX2w49XoeFtP1Rm/D+q6QW2eskf7tQ7RtuekBkX5LrJzIrlPZbGrf+hrb8lErE+T3D3w7XDL3QEJE8t5Hm7nHHUFWhUPCurq64w8BEBb1hysMNV6ULjiRsf0ITHplRqXxXfLU0IJsPlr6lOuLzpUw/MITjE1GifgWT9PN73NJQv9IQI5AwZrbS3QtjzaOHA8IX9BnCPDyDWOkbzjC2P8jJkgui6FXa/3y7gko4PhGlLNSvuG8Wg57f444/yvWnpX7RCy3d4j6G8CokHPIqyoMx6A1T3m+4gm5/0BN6GhI+cZ9M6BKLuKTh+ER6pb1+Jf2GNumP1PErUrUR9zVMlsV9DGFMJBzyKOqDMegN00Q+n+XGOmj5BT2hJz3hE/fJpBbr59sVjCfpxyfSLe31K+k3tNXO73HHn/dfkaqFuK9hsi7uY0jK9j3KFJFwyKNaHIxBb5iy/jvT1QQpv6An9KR/wx73IydJOJklHSfb6CT9+ES6pb1+peGGNsmP1EW9/rTXrzBwDRGtidThoNcojNMyaSQc8ijuE1pQNNaVhXFCT/I37HE/cpL24ydqnGyDq3YxlOTjE9GLOqGX5vqV9hvauOPP+69I1QLXENGqVoejHtiee5QxkXDIo7hPaEHRWFeX5RN63I+cpP34iRon22BI2FQXdw+aONdP/agu7ee/uOOPe/1ZxzVE9IIO3FpJtc9zjzImEg55leYTCo11/OK+4I/zkZOg60+DIPuXk20wJGwqi/uGeyLrj7J9pH4A6Zf1a4gki3pg+zDuUeK+xo4ACQfEI+jBRGMdn7gv+IPK+6Ck1QTdvyQEgyFhU1ncN9zV1h91+0j9QN7l+fyM4IJeo0zk80HuUdJ+jT0OEg5RoUEcX0YPplTJ+6CJeR+UtJIw9i8JwakjYVNZGDfcUfbgibp9pH5wfZVneT8/IxxBr1GivMbJwjX2GGJLOJhZnaQuSb3uflxccUSCBrGyjB5MqcGgiZXlvX7WYv9yw1BZ2hM2Ue7foD2Uou7BU4vjJ+31Iwiur/It7+dnZF9Gr7Hj7OFwjqS1kl4XYwzRoEGsLOUHU0d3r9o7e7Shr6jmpka1tbZo0fwZcYc1cSkfNDFo+Vf9fMrrZ2BV9m+18qtavhm/YQhcPmlXi/0bpIdS1D14AraPSa8fkbe/1XB9lWiR199a/ORhyqX9HJT0+CKX0V5ssSQczGx3ScdKukDSp+KIIVJ5v2GpJsUHU0d3r5YsX6Ni/6AkqbevqCXL10hSehrEFA+aGLT8J/R5BvwZd/9WK78JlW+GbxhCKZ+0C2P/RvnIV4J7ICS9ftSk/a2G66vEqkn9rXZ+znhCu5q0n4MSE1/c13AZ7MW2TUzr/Yakz0raEtP6ozXUIB7x+dw1dhM2c4F0yKdTVzbtnT3DDeGQYv+g2jt7YopoClJcP4OW/4Q/H6R+Dl3w3HpB6e8T90x+GQlVrfwmVL5DNwxWl7kbhlDKJ+2C7t+gx0+19Se4/Ut6/ahZ+1tJgvdf3tWs/lY6P4+VcMyRtJ+DEhFfhq/h4lTzHg5mdpykZ9x9pZkdXuF9Z0k6S5L22GOP2gQXpgxmpyBt6CtOanpipbR+Bi3/muy/DH+DX638JlS+Ke7hVM3Qdh5gD+vgbdbqri37aJXvNbnySbug+7cWj3wltP1Lev1ITPub0P2Xd4movznvARPKOTpGiYgvw9dwcYrjkYqFko43s3dLmibpdWb2Y3f/4Mg3uftSSUslqVAoeO3DBF6tualRvWM0fM1NjTFEkz9By78m+y/DFzzVym/C5ZvRG4bmpkbt9twfdNW2X1GDBtSvep22+Tw9veP+w/Nz0X4E2b8pfuQrqKTXj1S0v4hNIvZvhhPaExHaOTpq4zyykIj4MnwNF6eaP1Lh7kvcfXd3nyXpVEm3jk42AEnV1tqixoa6raY1NtSprbUlpojyJWj512T/ZbjLb7Xyy/vx0dbaonc2PKQGDajetqhBA3pnw0OUz2Rk+PipJun1IxXtL2KTmP2b0kd2w5CKc3SFRxYSEV+Oz0FRivNXKoDUGRq0Jtcj6MYoaPnXbP+l9BvWaqqVX96Pj0XzZ+j2TcdrYMVyyQc0YPV626HH6zDKZ3IyevxUk/T6kZr2F7Fg/8YvFefoCo8sJCI+KbfnoCiZe/KfVigUCt7V1RV3GAAAVBf3CNcAACRRzn9JJMvMbKW7F8aaRw8HAADCxLcjAAC8Ws7H2cgrEg4AAAAAgOiRlM+dmg8aCQAAAAAAso+EAwAAAAAACB0JBwAAAAAAEDoSDgAAAAAAIHQkHAAAAAAAQOhIOAAAAAAAgNCRcAAAAAAAAKGrjzsAYLI6unvV3tmjDX1FNTc1qq21RYvmz4g7rMxIe/lGHX+15Qddf9rLH8gyjs98i3v/x71+AJgKEg5IlY7uXi1ZvkbF/kFJUm9fUUuWr5EkTrohSHv5Rh1/teUHXX/ayx/IMo7PfIt7/8e9fgCYKh6pQKq0d/YMn2yHFPsH1d7ZE1NE2ZL28o06/mrLD7r+tJc/kGUcn/kW9/6Pe/0AMFUkHJAqG/qKk5qOyUl7+UYdf7XlB11/2ssfyDKOz3yLe//HvX4AmCoSDkiV5qbGSU3H5KS9fKOOv9ryg64/7eUPZBnHZ77Fvf/jXj8ATBUJB6RKW2uLGhvqtprW2FCnttaWmCLKlrSXb9TxV1t+0PWnvfyBLOP4zLe493/c6weAqWLQSKTK0MBIjNIcjbSXb9TxV1t+0PWnvfyBLOP4zLe493/c6weAqTJ3jzuGqgqFgnd1dcUdBgAAAAAAGMHMVrp7Yax5PFIBAAAAAABCR8IBAAAAAACEjoQDAAAAAAAIHQkHAAAAAAAQOhIOAAAAAAAgdCQcAAAAAABA6Eg4AAAAAACA0JFwAAAAAAAAoSPhAAAAAAAAQlfzhIOZzTSz35jZg2b2gJmdU+sYAAAAAABAtOpjWOeApE+7+yoz20HSSjO72d0fjCGWSHR096q9s0cb+opqbmpUW2uLFs2fEXdYNZP37Q8qaPlR/gDyivYv29i/0aJ8gco4Rqam5gkHd39S0pPl/79gZmslzZCUiYRDR3evlixfo2L/oCSpt6+oJcvXSFIuKmTetz+ooOVH+QPIK9q/bGP/RovyBSrjGJm6WMdwMLNZkuZLujvOOMLU3tkzXBGHFPsH1d7ZE1NEtZX37Q8qaPlR/gDyivYv29i/0aJ8gco4RqYutoSDmW0v6WeS/t3dnx9j/llm1mVmXRs3bqx9gFO0oa84qelZk/ftDypo+VH+APKK9i/b2L/RonyByjhGpi6WhIOZNaiUbLjK3ZeP9R53X+ruBXcvTJ8+vbYBBtDc1Dip6VmT9+0PKmj5Uf4A8or2L9vYv9GifIHKOEamLo5fqTBJP5C01t2/Vuv1R62ttUWNDXVbTWtsqFNba0tMEdVW3rc/qKDlR/kDyCvav2xj/0aL8gUq4xiZujh+pWKhpH+StMbMVpennefu18cQS+iGBg3J6wimed/+oIKWH+UPIK9o/7KN/RstyheojGNk6szd446hqkKh4F1dXXGHAQAAAAAARjCzle5eGGterL9SAQAAAAAAsomEAwAAAAAACB0JBwAAAAAAEDoSDgAAAAAAIHQkHAAAAAAAQOhIOAAAAAAAgNCRcAAAAAAAAKEj4QAAAAAAAEJHwgEAAAAAAISOhAMAAAAAAAhdfdwBAFnT0d2r9s4ebegrqrmpUW2tLVo0f0bcYWGC2H9IsqTXz2rxBY0/6dsftbxvf9pFfXwAWZf1c0zc64+KuXvcMVRVKBS8q6sr7jCAqjq6e7Vk+RoV+weHpzU21OnCk96aiQYj69h/SLKk189q8QWNP+nbH7W8b3/aRX18AFmX9XNM3OsPysxWunthrHk8UgGEqL2zZ6uGQpKK/YNq7+yJKSJMBvsPSZb0+lktvqDxJ337o5b37U+7qI8PIOuyfo6Je/1RIuEAhGhDX3FS05Es7D8kWdLrZ7X4gsaf9O2PWt63P+2iPj6ArMv6OSbu9UeJhAMQouamxklNR7Kw/5BkSa+f1eILGn/Stz9qed/+tIv6+ACyLuvnmLjXHyUSDkCI2lpb1NhQt9W0xoY6tbW2xBQRJoP9hyRLev2sFl/Q+JO+/VHL+/anXdTHB5B1WT/HxL3+KPErFUCIhgZ1yeIIs3nA/kOSJb1+VosvaPxJ3/6o5X370y7q4wPIuqyfY+Jef5T4lQoAAAAAADAl/EoFAAAAAACoKRIOAAAAAAAgdCQcAAAAAABA6Eg4AAAAAACA0JFwAAAAAAAAoSPhAAAAAAAAQkfCAQAAAAAAhI6EAwAAAAAACB0JBwAAAAAAELr6OFZqZsdI+qakOknfd/eL4ogjqzq6e9Xe2aMNfUU1NzWqrbVFi+bPmPD8uCU9vrSjfFFJ1PWD+odK0l7/qN/xovyTLej+Sfv1rZSOGKMU9/ZzDoiHuXttV2hWJ+lhSf9H0npJ90p6v7s/ON5nCoWCd3V11SjCdOvo7tWS5WtU7B8cntbYUKcLT3qrFs2fUXV+3JIeX9pRvqgk6vpB/UMlaa9/1O94Uf7JFnT/pP36VqKOxr39nAOiZWYr3b0w1rw4HqlYIOmP7v6ou2+W9FNJJ8QQRya1d/ZsVdElqdg/qPbOngnNj1vS40s7yheVRF0/qH+oJO31j/odL8o/2YLun7Rf30rpiDFKcW8/54D4xJFwmCHpiRGv15enbcXMzjKzLjPr2rhxY82CS7sNfcWK06vNj1vS40s7yheVRF0/qH+oJO31j/odL8o/2YLun7Rf30rpiDFKcW8/54D4JHbQSHdf6u4Fdy9Mnz497nBSo7mpseL0avPjlvT40o7yRSVR1w/qHypJe/2jfseL8k+2oPsn7de3UjpijFLc2885ID5xJBx6Jc0c8Xr38jSEoK21RY0NdVtNa2yoU1try4Tmxy3p8aUd5YtKoq4f1D9Ukvb6R/2OF+WfbEH3T9qvb6V0xBiluLefc0B84viVinslvcXMZquUaDhV0gdiiCOThgYlGW+E1Grz45b0+NKO8kUlUdcP6h8qSXv9o37Hi/JPtqD7J+3Xt1I6YoxS3NvPOSA+Nf+VCkkys3dL+oZKP4v5Q3e/oNL7+ZUKAAAAAACSp9KvVMTRw0Hufr2k6+NYNwAAAAAAiF5iB40EAAAAAADpRcIBAAAAAACEjoQDAAAAAAAIHQkHAAAAAAAQOhIOAAAAAAAgdCQcAAAAAABA6Eg4AAAAAACA0Jm7xx1DVWa2UdLjcccxBbtK+kvcQSC3qH+IE/UPcaL+IU7UP8SNOohae5O7Tx9rRioSDmllZl3uXog7DuQT9Q9xov4hTtQ/xIn6h7hRB5EkPFIBAAAAAABCR8IBAAAAAACEjoRDtJbGHQByjfqHOFH/ECfqH+JE/UPcqINIDMZwAAAAAAAAoaOHAwAAAAAACB0JhwiY2TFm1mNmfzSzc+OOB9lmZjPN7Ddm9qCZPWBm55Sn72xmN5vZI+W/O8UdK7LLzOrMrNvMflV+PdvM7i63g1eb2bZxx4jsMrMmM1tmZg+Z2VozezttIGrFzD5ZPv/eb2Y/MbNptIGIipn90MyeMbP7R0wbs72zkm+V6+F9ZnZAfJEjr0g4hMzM6iR9R9K7JM2R9H4zmxNvVMi4AUmfdvc5kg6W9K/lOneupFvc/S2Sbim/BqJyjqS1I17/X0lfd/d/kPRXSR+JJSrkxTcl3ejue0vaX6W6SBuIyJnZDEmfkFRw9/0k1Uk6VbSBiM7lko4ZNW289u5dkt5S/neWpP+pUYzAMBIO4Vsg6Y/u/qi7b5b0U0knxBwTMszdn3T3VeX/v6DShfYMlerdFeW3XSFpUSwBIvPMbHdJx0r6fvm1STpC0rLyW6h/iIyZ7SjpUEk/kCR33+zufaINRO3US2o0s3pJr5X0pGgDERF3v0PSplGTx2vvTpD0Iy+5S1KTmb2xJoECZSQcwjdD0hMjXq8vTwMiZ2azJM2XdLek3dz9yfKspyTtFldcyLxvSPqspC3l17tI6nP3gfJr2kFEabakjZIuKz/W830z2060gagBd++VdLGkP6uUaHhO0krRBqK2xmvvuC9B7Eg4ABlhZttL+pmkf3f350fO89LP0fCTNAidmR0n6Rl3Xxl3LMitekkHSPofd58v6SWNenyCNhBRKT8rf4JKia9mSdvp1d3dgZqhvUPSkHAIX6+kmSNe716eBkTGzBpUSjZc5e7Ly5OfHuo2V/77TFzxIdMWSjrezNap9AjZESo9T99U7l4s0Q4iWuslrXf3u8uvl6mUgKANRC0cJekxd9/o7v2SlqvULtIGopbGa++4L0HsSDiE715JbymPTrytSgMHXRtzTMiw8vPyP5C01t2/NmLWtZLOKP//DEm/qHVsyD53X+Luu7v7LJXau1vd/TRJv5F0cvlt1D9Ext2fkvSEmbWUJx0p6UHRBqI2/izpYDN7bfl8PFT/aANRS+O1d9dKOr38axUHS3puxKMXQE1YqdcNwmRm71bpmeY6ST909wvijQhZZmbvlLRC0hr9/Rn681Qax+EaSXtIelzSP7r76EGGgNCY2eGSPuPux5nZm1Xq8bCzpG5JH3T3V2IMDxlmZvNUGrR0W0mPSvqQSl+q0AYicmb2JUnvU+lXo7olnanSc/K0gQidmf1E0uGSdpX0tKQvSurQGO1dOQn2bZUe83lZ0ofcvSuGsJFjJBwAAAAAAEDoeKQCAAAAAACEjoQDAAAAAAAIHQkHAAAAAAAQOhIOAAAAAAAgdCQcAAAAAABA6Eg4AACAVzGzO6f4uUVmNifseAAAQPqQcAAAAK/i7u+Y4kcXSSLhAAAASDgAAIBXM7MXy38PN7PbzGyZmT1kZleZmZXnXWRmD5rZfWZ2sZm9Q9LxktrNbLWZ7WlmHzWze83sD2b2MzN7bfmzl5vZt8zsTjN71MxOHrHuz5nZmvJnLipP29PMbjSzlWa2wsz2Lk8/xczuL7/3jlqXEwAAGF993AEAAIDEmy9pX0kbJP1O0kIzWyvpREl7u7ubWZO795nZtZJ+5e7LJMnM+tz9e+X/f1nSRyRdUl7uGyW9U9Lekq6VtMzM3iXpBEkHufvLZrZz+b1LJZ3t7o+Y2UGS/lvSEZK+IKnV3XvNrCnicgAAAJNAwgEAAFRzj7uvlyQzWy1plqS7JP1N0g/M7FeSfjXOZ/crJxqaJG0vqXPEvA533yLpQTPbrTztKEmXufvLkuTum8xse0nvkPS/5c4VkvSa8t/fSbrczK6RtDzgdgIAgBCRcAAAANW8MuL/g5Lq3X3AzBZIOlLSyZI+rlKPg9Eul7TI3f9gZoslHT7Ock3j20ZSn7vPGz3D3c8u93g4VtJKM3ubuz9bdYsAAEDkGMMBAABMWrnXwY7ufr2kT0ravzzrBUk7jHjrDpKeNLMGSadNYNE3S/rQiLEednb35yU9ZmanlKeZme1f/v+e7n63u39B0kZJM0PYPAAAEAISDgAAYCp2kPQrM7tP0m8lfao8/aeS2sys28z2lPSfku5W6dGHh6ot1N1vVGk8h67y4xufKc86TdJHzOwPkh5QaZwHqTRA5Rozu1/SnZL+EMbGAQCA4Mzd444BAAAAAABkDD0cAAAAAABA6Eg4AAAAAACA0JFwAAAAAAAAoSPhAAAAAAAAQkfCAQAAAAAAhI6EAwAAAAAACB0JBwAAAAAAEDoSDgAAAAAAIHT/P3wuuwrJSBaMAAAAAElFTkSuQmCC\n","text/plain":["<Figure size 1296x360 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["y_pred = Poisreg.predict(X_val)\n","potential_models.append(('poisson_reg', Poisreg, mae(y_val, y_pred).numpy()))\n","plot_results(y_pred, title='Poisson regression')"]},{"cell_type":"markdown","id":"c501c3fe","metadata":{"papermill":{"duration":0.039256,"end_time":"2022-01-10T01:34:36.769605","exception":false,"start_time":"2022-01-10T01:34:36.730349","status":"completed"},"tags":[]},"source":["## Neural Network\n","\n","The final model we tested is a 2-layered fully connected neural network. Even with untuned hyper-parameters and blatant overfitting, we found that it outperformed other models for most stations. However, in the spirit of fairness, we will **not** consider neural networks for the subsequent phases of this work."]},{"cell_type":"code","execution_count":28,"id":"491d4de7","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:36.854434Z","iopub.status.busy":"2022-01-10T01:34:36.853699Z","iopub.status.idle":"2022-01-10T01:34:37.080341Z","shell.execute_reply":"2022-01-10T01:34:37.079457Z","shell.execute_reply.started":"2022-01-09T23:56:08.849038Z"},"papermill":{"duration":0.271814,"end_time":"2022-01-10T01:34:37.080543","exception":false,"start_time":"2022-01-10T01:34:36.808729","status":"completed"},"tags":[]},"outputs":[{"name":"stdout","output_type":"stream","text":["Model: \"sequential\"\n","_________________________________________________________________\n","Layer (type)                 Output Shape              Param #   \n","=================================================================\n","flatten (Flatten)            (457, 6)                  0         \n","_________________________________________________________________\n","dense (Dense)                (457, 4096)               28672     \n","_________________________________________________________________\n","dense_1 (Dense)              (457, 1024)               4195328   \n","_________________________________________________________________\n","dense_2 (Dense)              (457, 28)                 28700     \n","=================================================================\n","Total params: 4,252,700\n","Trainable params: 4,252,700\n","Non-trainable params: 0\n","_________________________________________________________________\n"]}],"source":["\"\"\" Define the model  \"\"\"\n","KerasNet = tf.keras.Sequential([\n","\n","        tf.keras.layers.Flatten(),\n","\n","        tf.keras.layers.Dense(4096, activation='relu'),\n","#         tf.keras.layers.Dropout(0.25),\n","#         tf.keras.layers.Flatten(),\n","\n","#         tf.keras.layers.Dense(1024, activation='relu', kernel_regularizer=tf.keras.regularizers.l2(0.001)),\n","        tf.keras.layers.Dense(1024, activation='relu'),\n","#         tf.keras.layers.Dropout(0.25),\n","\n","        tf.keras.layers.Dense(28, activation='softmax')        \n","    \n","#         tf.keras.layers.Dense(28, activation='relu'),\n","#         tf.keras.layers.Dense(1)                    \n","                    \n","      ])\n","\n","## Model architecture test with a batch size of 2\n","KerasNet(X_train)\n","\n","KerasNet.summary()"]},{"cell_type":"code","execution_count":29,"id":"e09530c6","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:37.170247Z","iopub.status.busy":"2022-01-10T01:34:37.169263Z","iopub.status.idle":"2022-01-10T01:34:37.177246Z","shell.execute_reply":"2022-01-10T01:34:37.176705Z","shell.execute_reply.started":"2022-01-09T23:56:08.955508Z"},"papermill":{"duration":0.056141,"end_time":"2022-01-10T01:34:37.177388","exception":false,"start_time":"2022-01-10T01:34:37.121247","status":"completed"},"tags":[]},"outputs":[],"source":["\"\"\" Compile the model  \"\"\"\n","KerasNet.compile(\n","#     loss=mae,    \n","    loss=tf.losses.sparse_categorical_crossentropy,\n","    optimizer=tf.optimizers.Adam(1e-4),\n","    metrics=[tf.metrics.sparse_categorical_accuracy])"]},{"cell_type":"code","execution_count":30,"id":"a406ccd3","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:34:37.265274Z","iopub.status.busy":"2022-01-10T01:34:37.264636Z","iopub.status.idle":"2022-01-10T01:35:18.808347Z","shell.execute_reply":"2022-01-10T01:35:18.807597Z","shell.execute_reply.started":"2022-01-09T23:56:08.969678Z"},"papermill":{"duration":41.590731,"end_time":"2022-01-10T01:35:18.808562","exception":false,"start_time":"2022-01-10T01:34:37.217831","status":"completed"},"tags":[]},"outputs":[{"name":"stderr","output_type":"stream","text":["2022-01-10 01:34:37.335632: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:185] None of the MLIR Optimization Passes are enabled (registered 2)\n"]}],"source":["\"\"\" Train the model (probably overfits)  \"\"\"\n","\n","history = KerasNet.fit(\n","      X_train,\n","      y_train,\n","      batch_size=60,\n","      epochs=100,\n","      verbose=0,\n","#       steps_per_epoch=100,\n","      validation_data=(X_val, y_val),\n",")"]},{"cell_type":"code","execution_count":31,"id":"319719dd","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:35:18.899103Z","iopub.status.busy":"2022-01-10T01:35:18.898269Z","iopub.status.idle":"2022-01-10T01:35:19.276132Z","shell.execute_reply":"2022-01-10T01:35:19.275657Z","shell.execute_reply.started":"2022-01-09T23:56:36.567341Z"},"papermill":{"duration":0.426465,"end_time":"2022-01-10T01:35:19.276272","exception":false,"start_time":"2022-01-10T01:35:18.849807","status":"completed"},"tags":[]},"outputs":[{"data":{"image/png":"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\n","text/plain":["<Figure size 1296x360 with 2 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["\"\"\" Visualize crossentropy loss and accuracy \"\"\"\n","\n","fig, (ax2, ax1) = plt.subplots(1, 2, figsize=(18, 5))\n","\n","# summarize history for accuracy\n","ax1.plot(history.history['sparse_categorical_accuracy'])\n","ax1.plot(history.history['val_sparse_categorical_accuracy'])\n","ax1.set_title('KerasNet accuracy')\n","ax1.set_ylabel('accuracy')\n","ax1.set_xlabel('epochs')\n","ax1.legend(['train', 'val'], loc='upper left')\n","\n","# summarize history for loss\n","ax2.plot(history.history['loss'])\n","ax2.plot(history.history['val_loss'])\n","ax2.set_title('KerasNet loss')\n","ax2.set_ylabel('loss')\n","ax2.set_xlabel('epochs')\n","ax2.legend(['train', 'val'], loc='upper right');"]},{"cell_type":"code","execution_count":32,"id":"a5f5e7c6","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:35:19.366801Z","iopub.status.busy":"2022-01-10T01:35:19.365965Z","iopub.status.idle":"2022-01-10T01:35:19.75235Z","shell.execute_reply":"2022-01-10T01:35:19.752808Z","shell.execute_reply.started":"2022-01-09T23:56:37.048033Z"},"papermill":{"duration":0.434813,"end_time":"2022-01-10T01:35:19.752986","exception":false,"start_time":"2022-01-10T01:35:19.318173","status":"completed"},"tags":[]},"outputs":[{"data":{"image/png":"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\n","text/plain":["<Figure size 1296x360 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["\"\"\" Make a few predictions \"\"\"\n","# Evaluate the model on the test data \n","# results = KerasNet.evaluate(X_val[:,:,np.newaxis], y_val, batch_size=12)\n","# print(\"Results (loss and accuracy) on Station \"+str(201)+\" test dataset:\", results)\n","\n","# Generate predictions on new data\n","y_hat_pred_proba = KerasNet.predict(X_val)\n","y_pred = tf.cast(tf.argmax(y_hat_pred_proba, axis=1),tf.int32)\n","\n","# Predictions in regression scenario\n","# y_pred = KerasNet.predict(X_val)\n","\n","## Save the model\n","potential_models.append(('neural_net', KerasNet, mae(y_val, y_pred).numpy()))\n","\n","plot_results(y_pred, title='Neural Network')"]},{"cell_type":"markdown","id":"6ddb7c49","metadata":{"papermill":{"duration":0.042668,"end_time":"2022-01-10T01:35:19.838477","exception":false,"start_time":"2022-01-10T01:35:19.795809","status":"completed"},"tags":[]},"source":["# Sort the models and pick the best\n","\n","Here, we compare the models and plot the result in a bar chart. As we stated before, only the best model will be selected to make predictions for the corresponding station in the next notebooks."]},{"cell_type":"code","execution_count":33,"id":"77bb5ce9","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:35:19.926182Z","iopub.status.busy":"2022-01-10T01:35:19.925567Z","iopub.status.idle":"2022-01-10T01:35:20.173365Z","shell.execute_reply":"2022-01-10T01:35:20.173852Z","shell.execute_reply.started":"2022-01-09T23:56:37.493203Z"},"papermill":{"duration":0.293284,"end_time":"2022-01-10T01:35:20.17402","exception":false,"start_time":"2022-01-10T01:35:19.880736","status":"completed"},"tags":[]},"outputs":[{"data":{"image/png":"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\n","text/plain":["<Figure size 1080x360 with 1 Axes>"]},"metadata":{"needs_background":"light"},"output_type":"display_data"}],"source":["# potential_models\n","\n","labels = []\n","y_plot = []\n","for mod in potential_models:\n","    labels.append(mod[0])\n","    y_plot.append(mod[2])\n","    \n","x_plot = np.arange(len(y_plot))\n","fig, ax = plt.subplots(1, 1, figsize=(15,5)) \n","plt.bar(x_plot, height=y_plot, color=['black', 'red', 'green', 'blue', 'cyan'])\n","plt.xticks(ticks=x_plot, labels=labels)\n","plt.ylim((0, 5))\n","plt.ylabel(\"MAE score\")\n","plt.title(\"Comparison of different learning algorithms on station \"+str(STATION_ID))\n","plt.legend();"]},{"cell_type":"code","execution_count":34,"id":"4d0a637e","metadata":{"execution":{"iopub.execute_input":"2022-01-10T01:35:20.263615Z","iopub.status.busy":"2022-01-10T01:35:20.262933Z","iopub.status.idle":"2022-01-10T01:35:20.269118Z","shell.execute_reply":"2022-01-10T01:35:20.268635Z","shell.execute_reply.started":"2022-01-09T23:59:45.547829Z"},"papermill":{"duration":0.051982,"end_time":"2022-01-10T01:35:20.26926","exception":false,"start_time":"2022-01-10T01:35:20.217278","status":"completed"},"tags":[]},"outputs":[{"data":{"text/plain":["('neural_net', <keras.engine.sequential.Sequential at 0x7fe4adad5510>, 1)"]},"execution_count":34,"metadata":{},"output_type":"execute_result"}],"source":["potential_models = sorted(potential_models, key=lambda el:el[2])\n","\n","## Save the best model\n","# if potential_models[0][0] != 'neural_net':\n","#     save_name = str(STATION_ID)+'_'+potential_models[0][0]+'.pkl'\n","#     joblib.dump(potential_models[0][1], save_name)\n","# else:\n","#     save_name = str(STATION_ID)+'_'+potential_models[0][0]+'.h5'\n","#     potential_models[0][1].save(save_name)\n","\n","## The best model is ...\n","potential_models[0]"]}],"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.7.12"},"papermill":{"default_parameters":{},"duration":68.821952,"end_time":"2022-01-10T01:35:21.525166","environment_variables":{},"exception":null,"input_path":"__notebook__.ipynb","output_path":"__notebook__.ipynb","parameters":{},"start_time":"2022-01-10T01:34:12.703214","version":"2.3.3"}},"nbformat":4,"nbformat_minor":5}