jdevoo/river.ipynb

## river.ipynb
{
 "cells": [
  {
   "cell_type": "raw",
   "id": "babcaf00-b901-43fe-a550-845914aeba91",
   "metadata": {},
   "source": [
    "from river import datasets\n",
    "from river import linear_model\n",
    "from river import metrics\n",
    "from river import compose\n",
    "from river import preprocessing\n",
    "from river import stats\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ed5c4d58",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Phishing websites.\n",
      "\n",
      "This dataset contains features from web pages that are classified as phishing or not.\n",
      "\n",
      "    Name  Phishing                                                             \n",
      "    Task  Binary classification                                                \n",
      " Samples  1,250                                                                \n",
      "Features  9                                                                    \n",
      "  Sparse  False                                                                \n",
      "    Path  /usr/local/lib/python3.9/dist-packages/river/datasets/phishing.csv.gz\n"
     ]
    }
   ],
   "source": [
    "dataset = datasets.Phishing()\n",
    "print(dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "005a5763",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'empty_server_form_handler': 1.0, 'popup_window': 0.5, 'https': 1.0, 'request_from_other_domain': 1.0, 'anchor_from_other_domain': 1.0, 'is_popular': 0.5, 'long_url': 0.0, 'age_of_domain': 0, 'ip_in_url': 0}\n"
     ]
    }
   ],
   "source": [
    "for x, y in dataset:\n",
    "    pass\n",
    "\n",
    "print(x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a2733936",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ROCAUC: 89.36%"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = linear_model.LogisticRegression()\n",
    "\n",
    "metric = metrics.ROCAUC()\n",
    "\n",
    "for x, y in dataset:\n",
    "    y_pred = model.predict_proba_one(x)\n",
    "    model.learn_one(x, y)\n",
    "    metric.update(y, y_pred)\n",
    "\n",
    "metric"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "cbafa66b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div class=\"component pipeline\"><details class=\"component estimator\"><summary><pre class=\"estimator-name\">StandardScaler</pre></summary><code class=\"estimator-params\">\n",
       "{'counts': Counter(),\n",
       " 'means': defaultdict(&lt;class 'float'&gt;, {}),\n",
       " 'vars': defaultdict(&lt;class 'float'&gt;, {}),\n",
       " 'with_std': True}\n",
       "\n",
       "</code></details><details class=\"component estimator\"><summary><pre class=\"estimator-name\">LogisticRegression</pre></summary><code class=\"estimator-params\">\n",
       "{'_weights': {},\n",
       " '_y_name': None,\n",
       " 'clip_gradient': 1000000000000.0,\n",
       " 'initializer': Zeros (),\n",
       " 'intercept': 0.0,\n",
       " 'intercept_init': 0.0,\n",
       " 'intercept_lr': Constant({'learning_rate': 0.01}),\n",
       " 'l2': 0.0,\n",
       " 'loss': Log({'weight_pos': 1.0, 'weight_neg': 1.0}),\n",
       " 'optimizer': SGD({'lr': Constant({'learning_rate': 0.01}), 'n_iterations': 0})}\n",
       "\n",
       "</code></details></div><style scoped>\n",
       ".estimator {\n",
       "    padding: 1em;\n",
       "    border-style: solid;\n",
       "    background: white;\n",
       "}\n",
       "\n",
       ".pipeline {\n",
       "    display: flex;\n",
       "    flex-direction: column;\n",
       "    align-items: center;\n",
       "    background: linear-gradient(#000, #000) no-repeat center / 3px 100%;\n",
       "}\n",
       "\n",
       ".union {\n",
       "    display: flex;\n",
       "    flex-direction: row;\n",
       "    align-items: center;\n",
       "    justify-content: center;\n",
       "    padding: 1em;\n",
       "    border-style: solid;\n",
       "    background: white\n",
       "}\n",
       "\n",
       ".wrapper {\n",
       "    display: flex;\n",
       "    flex-direction: column;\n",
       "    align-items: center;\n",
       "    justify-content: center;\n",
       "    padding: 1em;\n",
       "    border-style: solid;\n",
       "    background: white;\n",
       "}\n",
       "\n",
       ".wrapper > .estimator {\n",
       "    margin-top: 1em;\n",
       "}\n",
       "\n",
       "/* Vertical spacing between steps */\n",
       "\n",
       ".component + .component {\n",
       "    margin-top: 2em;\n",
       "}\n",
       "\n",
       ".union > .estimator {\n",
       "    margin-top: 0;\n",
       "}\n",
       "\n",
       ".union > .pipeline {\n",
       "    margin-top: 0;\n",
       "}\n",
       "\n",
       "/* Spacing within a union of estimators */\n",
       "\n",
       ".union > .component + .component {\n",
       "    margin-left: 1em;\n",
       "}\n",
       "\n",
       "/* Typography */\n",
       "\n",
       ".estimator-params {\n",
       "    display: block;\n",
       "    white-space: pre-wrap;\n",
       "    font-size: 120%;\n",
       "    margin-bottom: -1em;\n",
       "}\n",
       "\n",
       ".estimator > code,\n",
       ".wrapper > details > code {\n",
       "    background-color: white !important;\n",
       "}\n",
       "\n",
       ".estimator-name {\n",
       "    display: inline;\n",
       "    margin: 0;\n",
       "    font-size: 130%;\n",
       "}\n",
       "\n",
       "/* Toggle */\n",
       "\n",
       "summary {\n",
       "    display: flex;\n",
       "    align-items:center;\n",
       "    cursor: pointer;\n",
       "}\n",
       "\n",
       "summary > div {\n",
       "    width: 100%;\n",
       "}\n",
       "</style></div>"
      ],
      "text/plain": [
       "Pipeline (\n",
       "  StandardScaler (\n",
       "    with_std=True\n",
       "  ),\n",
       "  LogisticRegression (\n",
       "    optimizer=SGD (\n",
       "      lr=Constant (\n",
       "        learning_rate=0.01\n",
       "      )\n",
       "    )\n",
       "    loss=Log (\n",
       "      weight_pos=1.\n",
       "      weight_neg=1.\n",
       "    )\n",
       "    l2=0.\n",
       "    intercept_init=0.\n",
       "    intercept_lr=Constant (\n",
       "      learning_rate=0.01\n",
       "    )\n",
       "    clip_gradient=1e+12\n",
       "    initializer=Zeros ()\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model = compose.Pipeline(\n",
    "    preprocessing.StandardScaler(),\n",
    "    linear_model.LogisticRegression()\n",
    ")\n",
    "\n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c153c712",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "ROCAUC: 95.04%"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "metric = metrics.ROCAUC()\n",
    "\n",
    "for x, y in dataset:\n",
    "    y_pred = model.predict_proba_one(x)\n",
    "    model.learn_one(x, y)\n",
    "    metric.update(y, y_pred)\n",
    "\n",
    "metric"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0e230de9",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = datasets.AirlinePassengers()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9c62e8ed",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'month': datetime.datetime(1949, 1, 1, 0, 0)} 112\n"
     ]
    }
   ],
   "source": [
    "for x, y in datasets.AirlinePassengers():\n",
    "    print(x, y)\n",
    "    break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "e2b518d0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div><div class=\"component pipeline\"><details class=\"component estimator\"><summary><pre class=\"estimator-name\">get_ordinal_date</pre></summary><code class=\"estimator-params\">\n",
       "def get_ordinal_date(x):\n",
       "    return {'ordinal_date': x['month'].toordinal()}\n",
       "\n",
       "</code></details><details class=\"component estimator\"><summary><pre class=\"estimator-name\">StandardScaler</pre></summary><code class=\"estimator-params\">\n",
       "{'counts': Counter(),\n",
       " 'means': defaultdict(&lt;class 'float'&gt;, {}),\n",
       " 'vars': defaultdict(&lt;class 'float'&gt;, {}),\n",
       " 'with_std': True}\n",
       "\n",
       "</code></details><details class=\"component estimator\"><summary><pre class=\"estimator-name\">LinearRegression</pre></summary><code class=\"estimator-params\">\n",
       "{'_weights': {},\n",
       " '_y_name': None,\n",
       " 'clip_gradient': 1000000000000.0,\n",
       " 'initializer': Zeros (),\n",
       " 'intercept': 0.0,\n",
       " 'intercept_init': 0.0,\n",
       " 'intercept_lr': Constant({'learning_rate': 0.01}),\n",
       " 'l2': 0.0,\n",
       " 'loss': Squared({}),\n",
       " 'optimizer': SGD({'lr': Constant({'learning_rate': 0.01}), 'n_iterations': 0})}\n",
       "\n",
       "</code></details></div><style scoped>\n",
       ".estimator {\n",
       "    padding: 1em;\n",
       "    border-style: solid;\n",
       "    background: white;\n",
       "}\n",
       "\n",
       ".pipeline {\n",
       "    display: flex;\n",
       "    flex-direction: column;\n",
       "    align-items: center;\n",
       "    background: linear-gradient(#000, #000) no-repeat center / 3px 100%;\n",
       "}\n",
       "\n",
       ".union {\n",
       "    display: flex;\n",
       "    flex-direction: row;\n",
       "    align-items: center;\n",
       "    justify-content: center;\n",
       "    padding: 1em;\n",
       "    border-style: solid;\n",
       "    background: white\n",
       "}\n",
       "\n",
       ".wrapper {\n",
       "    display: flex;\n",
       "    flex-direction: column;\n",
       "    align-items: center;\n",
       "    justify-content: center;\n",
       "    padding: 1em;\n",
       "    border-style: solid;\n",
       "    background: white;\n",
       "}\n",
       "\n",
       ".wrapper > .estimator {\n",
       "    margin-top: 1em;\n",
       "}\n",
       "\n",
       "/* Vertical spacing between steps */\n",
       "\n",
       ".component + .component {\n",
       "    margin-top: 2em;\n",
       "}\n",
       "\n",
       ".union > .estimator {\n",
       "    margin-top: 0;\n",
       "}\n",
       "\n",
       ".union > .pipeline {\n",
       "    margin-top: 0;\n",
       "}\n",
       "\n",
       "/* Spacing within a union of estimators */\n",
       "\n",
       ".union > .component + .component {\n",
       "    margin-left: 1em;\n",
       "}\n",
       "\n",
       "/* Typography */\n",
       "\n",
       ".estimator-params {\n",
       "    display: block;\n",
       "    white-space: pre-wrap;\n",
       "    font-size: 120%;\n",
       "    margin-bottom: -1em;\n",
       "}\n",
       "\n",
       ".estimator > code,\n",
       ".wrapper > details > code {\n",
       "    background-color: white !important;\n",
       "}\n",
       "\n",
       ".estimator-name {\n",
       "    display: inline;\n",
       "    margin: 0;\n",
       "    font-size: 130%;\n",
       "}\n",
       "\n",
       "/* Toggle */\n",
       "\n",
       "summary {\n",
       "    display: flex;\n",
       "    align-items:center;\n",
       "    cursor: pointer;\n",
       "}\n",
       "\n",
       "summary > div {\n",
       "    width: 100%;\n",
       "}\n",
       "</style></div>"
      ],
      "text/plain": [
       "Pipeline (\n",
       "  FuncTransformer (\n",
       "    func=\"get_ordinal_date\"\n",
       "  ),\n",
       "  StandardScaler (\n",
       "    with_std=True\n",
       "  ),\n",
       "  LinearRegression (\n",
       "    optimizer=SGD (\n",
       "      lr=Constant (\n",
       "        learning_rate=0.01\n",
       "      )\n",
       "    )\n",
       "    loss=Squared ()\n",
       "    l2=0.\n",
       "    intercept_init=0.\n",
       "    intercept_lr=Constant (\n",
       "      learning_rate=0.01\n",
       "    )\n",
       "    clip_gradient=1e+12\n",
       "    initializer=Zeros ()\n",
       "  )\n",
       ")"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def get_ordinal_date(x):\n",
    "    return {'ordinal_date': x['month'].toordinal()}\n",
    "\n",
    "model = compose.Pipeline(\n",
    "    ('ordinal_date', compose.FuncTransformer(get_ordinal_date)),\n",
    "    ('scale', preprocessing.StandardScaler()),\n",
    "    ('lin_reg', linear_model.LinearRegression())\n",
    ")\n",
    "\n",
    "model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "f1b7bd77",
   "metadata": {},
   "outputs": [],
   "source": [
    "def evaluate_model(model): \n",
    "\n",
    "    metric = metrics.Rolling(metrics.MAE(), 12)\n",
    "\n",
    "    dates = []\n",
    "    y_trues = []\n",
    "    y_preds = []\n",
    "\n",
    "    for x, y in datasets.AirlinePassengers():\n",
    "\n",
    "        # Obtain the prior prediction and update the model in one go\n",
    "        y_pred = model.predict_one(x)\n",
    "        model.learn_one(x, y)\n",
    "\n",
    "        # Update the error metric\n",
    "        metric.update(y, y_pred)\n",
    "\n",
    "        # Store the true value and the prediction\n",
    "        dates.append(x['month'])\n",
    "        y_trues.append(y)\n",
    "        y_preds.append(y_pred)\n",
    "\n",
    "    # Plot the results\n",
    "    fig, ax = plt.subplots(figsize=(10, 6))\n",
    "    ax.grid(alpha=0.75)\n",
    "    ax.plot(dates, y_trues, lw=3, color='#2ecc71', alpha=0.8, label='Ground truth')\n",
    "    ax.plot(dates, y_preds, lw=3, color='#e74c3c', alpha=0.8, label='Prediction')\n",
    "    ax.legend()\n",
    "    ax.set_title(metric)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "bcfb5a68",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.9/dist-packages/IPython/core/pylabtools.py:151: UserWarning: Glyph 9 (\t) missing from current font.\n",
      "  fig.canvas.print_figure(bytes_io, **kw)\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "evaluate_model(model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "ef83fe94",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "model = compose.Pipeline(\n",
    "    ('ordinal_date', compose.FuncTransformer(get_ordinal_date)),\n",
    "    ('scale', preprocessing.StandardScaler()),\n",
    "    ('lin_reg', preprocessing.TargetStandardScaler(regressor=linear_model.LinearRegression(intercept_lr=0))),\n",
    ")\n",
    "\n",
    "evaluate_model(model)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c5cb6db9",
   "metadata": {},
   "source": [
    "See also https://github.com/online-ml/river/issues/574\n",
    "for median, see below\n",
    "* https://gist.github.com/ashelly/5665911\n",
    "* https://stackoverflow.com/questions/5527437/rolling-median-in-c-turlach-implementation\n",
    "* https://www.philippe-fournier-viger.com/spmf/MedianSmoothing.php"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "dfbd6996",
   "metadata": {},
   "outputs": [],
   "source": [
    "median = stats.Quantile(0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "59bf5bf7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Quantile: 5."
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "median.update(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "de8ced48",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Quantile: 15."
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "median.update(15)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "cda163e9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Quantile: 5."
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "median.update(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "71d1fd2e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Quantile: 5."
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "median.update(3)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}