{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Licensed to the Apache Software Foundation (ASF) under one\nor more contributor license agreements.  See the NOTICE file\ndistributed with this work for additional information\nregarding copyright ownership.  The ASF licenses this file\nto you under the Apache License, Version 2.0 (the\n\"License\"); you may not use this file except in compliance\nwith the License.  You may obtain a copy of the License at\n\n  http://www.apache.org/licenses/LICENSE-2.0\n\nUnless required by applicable law or agreed to in writing,\nsoftware distributed under the License is distributed on an\n\"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\nKIND, either express or implied.  See the License for the\nspecific language governing permissions and limitations\nunder the License."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Execute this cell to install dependencies\n",
    "%pip install sf-hamilton[visualization]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Vaex + Hamilton integration [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/dagworks-inc/hamilton/blob/main/examples/vaex/notebook.ipynb) [![GitHub badge](https://img.shields.io/badge/github-view_source-2b3137?logo=github)](https://github.com/apache/hamilton/blob/main/examples/vaex/notebook.ipynb)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/konstantin/.miniconda3/envs/hamilton-vaex/lib/python3.9/site-packages/dask/dataframe/_pyarrow_compat.py:17: FutureWarning: Minimal version of pyarrow will soon be increased to 14.0.1. You are using 11.0.0. Please consider upgrading.\n",
      "  warnings.warn(\n",
      "/Users/konstantin/.miniconda3/envs/hamilton-vaex/lib/python3.9/site-packages/scipy/__init__.py:155: UserWarning: A NumPy version >=1.18.5 and <1.25.0 is required for this version of SciPy (detected version 1.26.3\n",
      "  warnings.warn(f\"A NumPy version >={np_minversion} and <{np_maxversion}\"\n"
     ]
    }
   ],
   "source": [
    "from hamilton import base, driver\n",
    "from hamilton.plugins import h_vaex"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import my_functions\n",
    "\n",
    "config = {\n",
    "    \"base_df_location\": \"dummy_value\",\n",
    "}\n",
    "adapter = base.SimplePythonGraphAdapter(result_builder=h_vaex.VaexDataFrameResult())\n",
    "dr = driver.Driver(config, my_functions, adapter=adapter)\n",
    "output_columns = [\n",
    "    \"spend\",\n",
    "    \"signups\",\n",
    "    \"spend_per_signup\",\n",
    "    \"spend_std_dev\",\n",
    "    \"spend_mean\",\n",
    "    \"spend_zero_mean_unit_variance\",\n",
    "]\n",
    "df = dr.execute(output_columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  #    spend    signups    spend_per_signup    spend_zero_mean_unit_variance    spend_std_dev    spend_mean\n",
      "  0       10          1              10                            -1.166             15.7233       28.3333\n",
      "  1       10         10               1                            -1.166             15.7233       28.3333\n",
      "  2       20         50               0.4                          -0.529999          15.7233       28.3333\n",
      "  3       40        100               0.4                           0.741999          15.7233       28.3333\n",
      "  4       40        200               0.2                           0.741999          15.7233       28.3333\n",
      "  5       50        400               0.125                         1.378             15.7233       28.3333\n"
     ]
    }
   ],
   "source": [
    "print(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.50.0 (0)\n",
       " -->\n",
       "<!-- Pages: 1 -->\n",
       "<svg width=\"1008pt\" height=\"428pt\"\n",
       " viewBox=\"0.00 0.00 1008.00 428.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 424)\">\n",
       "<polygon fill=\"white\" stroke=\"transparent\" points=\"-4,4 -4,-424 1004,-424 1004,4 -4,4\"/>\n",
       "<g id=\"clust1\" class=\"cluster\">\n",
       "<title>cluster__legend</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"34.5,-225 34.5,-412 120.5,-412 120.5,-225 34.5,-225\"/>\n",
       "<text text-anchor=\"middle\" x=\"77.5\" y=\"-396.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Legend</text>\n",
       "</g>\n",
       "<!-- signups -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>signups</title>\n",
       "<path fill=\"#ffc857\" stroke=\"black\" d=\"M381,-266C381,-266 316,-266 316,-266 310,-266 304,-260 304,-254 304,-254 304,-214 304,-214 304,-208 310,-202 316,-202 316,-202 381,-202 381,-202 387,-202 393,-208 393,-214 393,-214 393,-254 393,-254 393,-260 387,-266 381,-266\"/>\n",
       "<text text-anchor=\"start\" x=\"322.5\" y=\"-244.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">signups</text>\n",
       "<text text-anchor=\"start\" x=\"315\" y=\"-216.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">Expression</text>\n",
       "</g>\n",
       "<!-- spend_per_signup -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>spend_per_signup</title>\n",
       "<path fill=\"#ffc857\" stroke=\"black\" d=\"M553,-266C553,-266 434,-266 434,-266 428,-266 422,-260 422,-254 422,-254 422,-214 422,-214 422,-208 428,-202 434,-202 434,-202 553,-202 553,-202 559,-202 565,-208 565,-214 565,-214 565,-254 565,-254 565,-260 559,-266 553,-266\"/>\n",
       "<text text-anchor=\"start\" x=\"433\" y=\"-244.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">spend_per_signup</text>\n",
       "<text text-anchor=\"start\" x=\"460\" y=\"-216.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">Expression</text>\n",
       "</g>\n",
       "<!-- signups&#45;&gt;spend_per_signup -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>signups&#45;&gt;spend_per_signup</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M393.18,-234C399.16,-234 405.46,-234 411.85,-234\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"411.94,-237.5 421.94,-234 411.94,-230.5 411.94,-237.5\"/>\n",
       "</g>\n",
       "<!-- base_df -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>base_df</title>\n",
       "<path fill=\"#b4d8e4\" stroke=\"black\" d=\"M263,-225C263,-225 196,-225 196,-225 190,-225 184,-219 184,-213 184,-213 184,-173 184,-173 184,-167 190,-161 196,-161 196,-161 263,-161 263,-161 269,-161 275,-167 275,-173 275,-173 275,-213 275,-213 275,-219 269,-225 263,-225\"/>\n",
       "<text text-anchor=\"start\" x=\"203.5\" y=\"-203.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">base_df</text>\n",
       "<text text-anchor=\"start\" x=\"195\" y=\"-175.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">DataFrame</text>\n",
       "</g>\n",
       "<!-- base_df&#45;&gt;signups -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>base_df&#45;&gt;signups</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M275.07,-208.62C281.25,-210.79 287.66,-213.03 293.99,-215.25\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"293.24,-218.69 303.83,-218.7 295.55,-212.09 293.24,-218.69\"/>\n",
       "</g>\n",
       "<!-- spend -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>spend</title>\n",
       "<path fill=\"#ffc857\" stroke=\"black\" d=\"M381,-184C381,-184 316,-184 316,-184 310,-184 304,-178 304,-172 304,-172 304,-132 304,-132 304,-126 310,-120 316,-120 316,-120 381,-120 381,-120 387,-120 393,-126 393,-132 393,-132 393,-172 393,-172 393,-178 387,-184 381,-184\"/>\n",
       "<text text-anchor=\"start\" x=\"328.5\" y=\"-162.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">spend</text>\n",
       "<text text-anchor=\"start\" x=\"315\" y=\"-134.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">Expression</text>\n",
       "</g>\n",
       "<!-- base_df&#45;&gt;spend -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>base_df&#45;&gt;spend</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M275.07,-177.38C281.25,-175.21 287.66,-172.97 293.99,-170.75\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"295.55,-173.91 303.83,-167.3 293.24,-167.31 295.55,-173.91\"/>\n",
       "</g>\n",
       "<!-- spend_zero_mean -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>spend_zero_mean</title>\n",
       "<path fill=\"#b4d8e4\" stroke=\"black\" d=\"M723,-146C723,-146 606,-146 606,-146 600,-146 594,-140 594,-134 594,-134 594,-94 594,-94 594,-88 600,-82 606,-82 606,-82 723,-82 723,-82 729,-82 735,-88 735,-94 735,-94 735,-134 735,-134 735,-140 729,-146 723,-146\"/>\n",
       "<text text-anchor=\"start\" x=\"605\" y=\"-124.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">spend_zero_mean</text>\n",
       "<text text-anchor=\"start\" x=\"641.5\" y=\"-96.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">ndarray</text>\n",
       "</g>\n",
       "<!-- spend_zero_mean_unit_variance -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>spend_zero_mean_unit_variance</title>\n",
       "<path fill=\"#ffc857\" stroke=\"black\" d=\"M988,-105C988,-105 776,-105 776,-105 770,-105 764,-99 764,-93 764,-93 764,-53 764,-53 764,-47 770,-41 776,-41 776,-41 988,-41 988,-41 994,-41 1000,-47 1000,-53 1000,-53 1000,-93 1000,-93 1000,-99 994,-105 988,-105\"/>\n",
       "<text text-anchor=\"start\" x=\"775\" y=\"-83.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">spend_zero_mean_unit_variance</text>\n",
       "<text text-anchor=\"start\" x=\"859\" y=\"-55.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">ndarray</text>\n",
       "</g>\n",
       "<!-- spend_zero_mean&#45;&gt;spend_zero_mean_unit_variance -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>spend_zero_mean&#45;&gt;spend_zero_mean_unit_variance</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M735.17,-100.75C741.22,-99.59 747.46,-98.41 753.79,-97.2\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"754.63,-100.61 763.8,-95.3 753.32,-93.73 754.63,-100.61\"/>\n",
       "</g>\n",
       "<!-- spend&#45;&gt;spend_zero_mean -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>spend&#45;&gt;spend_zero_mean</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M393.33,-157.94C437.22,-162.65 506.48,-166.69 565,-155 572.12,-153.58 579.36,-151.63 586.53,-149.35\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"587.9,-152.58 596.24,-146.06 585.65,-145.96 587.9,-152.58\"/>\n",
       "</g>\n",
       "<!-- spend&#45;&gt;spend_per_signup -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>spend&#45;&gt;spend_per_signup</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M393.18,-177.05C404.11,-183.32 416.07,-190.17 427.78,-196.89\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"426.26,-200.05 436.67,-201.99 429.74,-193.98 426.26,-200.05\"/>\n",
       "</g>\n",
       "<!-- spend_std_dev -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>spend_std_dev</title>\n",
       "<path fill=\"#ffc857\" stroke=\"black\" d=\"M542,-64C542,-64 445,-64 445,-64 439,-64 433,-58 433,-52 433,-52 433,-12 433,-12 433,-6 439,0 445,0 445,0 542,0 542,0 548,0 554,-6 554,-12 554,-12 554,-52 554,-52 554,-58 548,-64 542,-64\"/>\n",
       "<text text-anchor=\"start\" x=\"444\" y=\"-42.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">spend_std_dev</text>\n",
       "<text text-anchor=\"start\" x=\"480.5\" y=\"-14.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">float</text>\n",
       "</g>\n",
       "<!-- spend&#45;&gt;spend_std_dev -->\n",
       "<g id=\"edge10\" class=\"edge\">\n",
       "<title>spend&#45;&gt;spend_std_dev</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M375.07,-119.88C388.14,-104.63 404.89,-86.77 422,-73 423.17,-72.06 424.37,-71.12 425.58,-70.19\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"427.81,-72.91 433.86,-64.21 423.71,-67.23 427.81,-72.91\"/>\n",
       "</g>\n",
       "<!-- spend_mean -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>spend_mean</title>\n",
       "<path fill=\"#ffc857\" stroke=\"black\" d=\"M534.5,-146C534.5,-146 452.5,-146 452.5,-146 446.5,-146 440.5,-140 440.5,-134 440.5,-134 440.5,-94 440.5,-94 440.5,-88 446.5,-82 452.5,-82 452.5,-82 534.5,-82 534.5,-82 540.5,-82 546.5,-88 546.5,-94 546.5,-94 546.5,-134 546.5,-134 546.5,-140 540.5,-146 534.5,-146\"/>\n",
       "<text text-anchor=\"start\" x=\"451.5\" y=\"-124.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">spend_mean</text>\n",
       "<text text-anchor=\"start\" x=\"480.5\" y=\"-96.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">float</text>\n",
       "</g>\n",
       "<!-- spend&#45;&gt;spend_mean -->\n",
       "<g id=\"edge11\" class=\"edge\">\n",
       "<title>spend&#45;&gt;spend_mean</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M393.18,-140.39C404.91,-137.28 417.83,-133.84 430.35,-130.51\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"431.52,-133.83 440.28,-127.88 429.72,-127.06 431.52,-133.83\"/>\n",
       "</g>\n",
       "<!-- spend_std_dev&#45;&gt;spend_zero_mean_unit_variance -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>spend_std_dev&#45;&gt;spend_zero_mean_unit_variance</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M554.04,-38.32C607.01,-43.94 686.27,-52.34 753.49,-59.47\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"753.43,-62.99 763.74,-60.56 754.17,-56.03 753.43,-62.99\"/>\n",
       "</g>\n",
       "<!-- spend_mean&#45;&gt;spend_zero_mean -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>spend_mean&#45;&gt;spend_zero_mean</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M546.58,-114C558.37,-114 571.15,-114 583.78,-114\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"583.86,-117.5 593.86,-114 583.86,-110.5 583.86,-117.5\"/>\n",
       "</g>\n",
       "<!-- _base_df_inputs -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>_base_df_inputs</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" stroke-dasharray=\"5,2\" points=\"155,-215.5 0,-215.5 0,-170.5 155,-170.5 155,-215.5\"/>\n",
       "<text text-anchor=\"start\" x=\"15.5\" y=\"-188.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">base_df_location</text>\n",
       "<text text-anchor=\"start\" x=\"124.5\" y=\"-188.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">str</text>\n",
       "</g>\n",
       "<!-- _base_df_inputs&#45;&gt;base_df -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>_base_df_inputs&#45;&gt;base_df</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M155.26,-193C161.54,-193 167.8,-193 173.87,-193\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"173.88,-196.5 183.88,-193 173.88,-189.5 173.88,-196.5\"/>\n",
       "</g>\n",
       "<!-- input -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>input</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" stroke-dasharray=\"5,2\" points=\"104.5,-380.5 50.5,-380.5 50.5,-343.5 104.5,-343.5 104.5,-380.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"77.5\" y=\"-358.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">input</text>\n",
       "</g>\n",
       "<!-- function -->\n",
       "<g id=\"node11\" class=\"node\">\n",
       "<title>function</title>\n",
       "<path fill=\"#b4d8e4\" stroke=\"black\" d=\"M100.5,-325.5C100.5,-325.5 54.5,-325.5 54.5,-325.5 48.5,-325.5 42.5,-319.5 42.5,-313.5 42.5,-313.5 42.5,-300.5 42.5,-300.5 42.5,-294.5 48.5,-288.5 54.5,-288.5 54.5,-288.5 100.5,-288.5 100.5,-288.5 106.5,-288.5 112.5,-294.5 112.5,-300.5 112.5,-300.5 112.5,-313.5 112.5,-313.5 112.5,-319.5 106.5,-325.5 100.5,-325.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"77.5\" y=\"-303.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">function</text>\n",
       "</g>\n",
       "<!-- output -->\n",
       "<g id=\"node12\" class=\"node\">\n",
       "<title>output</title>\n",
       "<path fill=\"#ffc857\" stroke=\"black\" d=\"M95.5,-270.5C95.5,-270.5 59.5,-270.5 59.5,-270.5 53.5,-270.5 47.5,-264.5 47.5,-258.5 47.5,-258.5 47.5,-245.5 47.5,-245.5 47.5,-239.5 53.5,-233.5 59.5,-233.5 59.5,-233.5 95.5,-233.5 95.5,-233.5 101.5,-233.5 107.5,-239.5 107.5,-245.5 107.5,-245.5 107.5,-258.5 107.5,-258.5 107.5,-264.5 101.5,-270.5 95.5,-270.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"77.5\" y=\"-248.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">output</text>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.graphs.Digraph at 0x1664d4d90>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dr.visualize_execution(output_columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/svg+xml": [
       "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>\n",
       "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n",
       " \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n",
       "<!-- Generated by graphviz version 2.50.0 (0)\n",
       " -->\n",
       "<!-- Pages: 1 -->\n",
       "<svg width=\"1008pt\" height=\"373pt\"\n",
       " viewBox=\"0.00 0.00 1008.00 373.00\" xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\">\n",
       "<g id=\"graph0\" class=\"graph\" transform=\"scale(1 1) rotate(0) translate(4 369)\">\n",
       "<polygon fill=\"white\" stroke=\"transparent\" points=\"-4,4 -4,-369 1004,-369 1004,4 -4,4\"/>\n",
       "<g id=\"clust1\" class=\"cluster\">\n",
       "<title>cluster__legend</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" points=\"34.5,-225 34.5,-357 120.5,-357 120.5,-225 34.5,-225\"/>\n",
       "<text text-anchor=\"middle\" x=\"77.5\" y=\"-341.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">Legend</text>\n",
       "</g>\n",
       "<!-- signups -->\n",
       "<g id=\"node1\" class=\"node\">\n",
       "<title>signups</title>\n",
       "<path fill=\"#b4d8e4\" stroke=\"black\" d=\"M381,-266C381,-266 316,-266 316,-266 310,-266 304,-260 304,-254 304,-254 304,-214 304,-214 304,-208 310,-202 316,-202 316,-202 381,-202 381,-202 387,-202 393,-208 393,-214 393,-214 393,-254 393,-254 393,-260 387,-266 381,-266\"/>\n",
       "<text text-anchor=\"start\" x=\"322.5\" y=\"-244.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">signups</text>\n",
       "<text text-anchor=\"start\" x=\"315\" y=\"-216.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">Expression</text>\n",
       "</g>\n",
       "<!-- spend_per_signup -->\n",
       "<g id=\"node6\" class=\"node\">\n",
       "<title>spend_per_signup</title>\n",
       "<path fill=\"#b4d8e4\" stroke=\"black\" d=\"M553,-266C553,-266 434,-266 434,-266 428,-266 422,-260 422,-254 422,-254 422,-214 422,-214 422,-208 428,-202 434,-202 434,-202 553,-202 553,-202 559,-202 565,-208 565,-214 565,-214 565,-254 565,-254 565,-260 559,-266 553,-266\"/>\n",
       "<text text-anchor=\"start\" x=\"433\" y=\"-244.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">spend_per_signup</text>\n",
       "<text text-anchor=\"start\" x=\"460\" y=\"-216.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">Expression</text>\n",
       "</g>\n",
       "<!-- signups&#45;&gt;spend_per_signup -->\n",
       "<g id=\"edge9\" class=\"edge\">\n",
       "<title>signups&#45;&gt;spend_per_signup</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M393.18,-234C399.16,-234 405.46,-234 411.85,-234\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"411.94,-237.5 421.94,-234 411.94,-230.5 411.94,-237.5\"/>\n",
       "</g>\n",
       "<!-- base_df -->\n",
       "<g id=\"node2\" class=\"node\">\n",
       "<title>base_df</title>\n",
       "<path fill=\"#b4d8e4\" stroke=\"black\" d=\"M263,-225C263,-225 196,-225 196,-225 190,-225 184,-219 184,-213 184,-213 184,-173 184,-173 184,-167 190,-161 196,-161 196,-161 263,-161 263,-161 269,-161 275,-167 275,-173 275,-173 275,-213 275,-213 275,-219 269,-225 263,-225\"/>\n",
       "<text text-anchor=\"start\" x=\"203.5\" y=\"-203.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">base_df</text>\n",
       "<text text-anchor=\"start\" x=\"195\" y=\"-175.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">DataFrame</text>\n",
       "</g>\n",
       "<!-- base_df&#45;&gt;signups -->\n",
       "<g id=\"edge1\" class=\"edge\">\n",
       "<title>base_df&#45;&gt;signups</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M275.07,-208.62C281.25,-210.79 287.66,-213.03 293.99,-215.25\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"293.24,-218.69 303.83,-218.7 295.55,-212.09 293.24,-218.69\"/>\n",
       "</g>\n",
       "<!-- spend -->\n",
       "<g id=\"node5\" class=\"node\">\n",
       "<title>spend</title>\n",
       "<path fill=\"#b4d8e4\" stroke=\"black\" d=\"M381,-184C381,-184 316,-184 316,-184 310,-184 304,-178 304,-172 304,-172 304,-132 304,-132 304,-126 310,-120 316,-120 316,-120 381,-120 381,-120 387,-120 393,-126 393,-132 393,-132 393,-172 393,-172 393,-178 387,-184 381,-184\"/>\n",
       "<text text-anchor=\"start\" x=\"328.5\" y=\"-162.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">spend</text>\n",
       "<text text-anchor=\"start\" x=\"315\" y=\"-134.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">Expression</text>\n",
       "</g>\n",
       "<!-- base_df&#45;&gt;spend -->\n",
       "<g id=\"edge7\" class=\"edge\">\n",
       "<title>base_df&#45;&gt;spend</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M275.07,-177.38C281.25,-175.21 287.66,-172.97 293.99,-170.75\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"295.55,-173.91 303.83,-167.3 293.24,-167.31 295.55,-173.91\"/>\n",
       "</g>\n",
       "<!-- spend_zero_mean -->\n",
       "<g id=\"node3\" class=\"node\">\n",
       "<title>spend_zero_mean</title>\n",
       "<path fill=\"#b4d8e4\" stroke=\"black\" d=\"M723,-146C723,-146 606,-146 606,-146 600,-146 594,-140 594,-134 594,-134 594,-94 594,-94 594,-88 600,-82 606,-82 606,-82 723,-82 723,-82 729,-82 735,-88 735,-94 735,-94 735,-134 735,-134 735,-140 729,-146 723,-146\"/>\n",
       "<text text-anchor=\"start\" x=\"605\" y=\"-124.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">spend_zero_mean</text>\n",
       "<text text-anchor=\"start\" x=\"641.5\" y=\"-96.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">ndarray</text>\n",
       "</g>\n",
       "<!-- spend_zero_mean_unit_variance -->\n",
       "<g id=\"node4\" class=\"node\">\n",
       "<title>spend_zero_mean_unit_variance</title>\n",
       "<path fill=\"#b4d8e4\" stroke=\"black\" d=\"M988,-105C988,-105 776,-105 776,-105 770,-105 764,-99 764,-93 764,-93 764,-53 764,-53 764,-47 770,-41 776,-41 776,-41 988,-41 988,-41 994,-41 1000,-47 1000,-53 1000,-53 1000,-93 1000,-93 1000,-99 994,-105 988,-105\"/>\n",
       "<text text-anchor=\"start\" x=\"775\" y=\"-83.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">spend_zero_mean_unit_variance</text>\n",
       "<text text-anchor=\"start\" x=\"859\" y=\"-55.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">ndarray</text>\n",
       "</g>\n",
       "<!-- spend_zero_mean&#45;&gt;spend_zero_mean_unit_variance -->\n",
       "<g id=\"edge5\" class=\"edge\">\n",
       "<title>spend_zero_mean&#45;&gt;spend_zero_mean_unit_variance</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M735.17,-100.75C741.22,-99.59 747.46,-98.41 753.79,-97.2\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"754.63,-100.61 763.8,-95.3 753.32,-93.73 754.63,-100.61\"/>\n",
       "</g>\n",
       "<!-- spend&#45;&gt;spend_zero_mean -->\n",
       "<g id=\"edge3\" class=\"edge\">\n",
       "<title>spend&#45;&gt;spend_zero_mean</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M393.33,-157.94C437.22,-162.65 506.48,-166.69 565,-155 572.12,-153.58 579.36,-151.63 586.53,-149.35\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"587.9,-152.58 596.24,-146.06 585.65,-145.96 587.9,-152.58\"/>\n",
       "</g>\n",
       "<!-- spend&#45;&gt;spend_per_signup -->\n",
       "<g id=\"edge8\" class=\"edge\">\n",
       "<title>spend&#45;&gt;spend_per_signup</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M393.18,-177.05C404.11,-183.32 416.07,-190.17 427.78,-196.89\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"426.26,-200.05 436.67,-201.99 429.74,-193.98 426.26,-200.05\"/>\n",
       "</g>\n",
       "<!-- spend_std_dev -->\n",
       "<g id=\"node7\" class=\"node\">\n",
       "<title>spend_std_dev</title>\n",
       "<path fill=\"#b4d8e4\" stroke=\"black\" d=\"M542,-64C542,-64 445,-64 445,-64 439,-64 433,-58 433,-52 433,-52 433,-12 433,-12 433,-6 439,0 445,0 445,0 542,0 542,0 548,0 554,-6 554,-12 554,-12 554,-52 554,-52 554,-58 548,-64 542,-64\"/>\n",
       "<text text-anchor=\"start\" x=\"444\" y=\"-42.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">spend_std_dev</text>\n",
       "<text text-anchor=\"start\" x=\"480.5\" y=\"-14.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">float</text>\n",
       "</g>\n",
       "<!-- spend&#45;&gt;spend_std_dev -->\n",
       "<g id=\"edge10\" class=\"edge\">\n",
       "<title>spend&#45;&gt;spend_std_dev</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M375.07,-119.88C388.14,-104.63 404.89,-86.77 422,-73 423.17,-72.06 424.37,-71.12 425.58,-70.19\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"427.81,-72.91 433.86,-64.21 423.71,-67.23 427.81,-72.91\"/>\n",
       "</g>\n",
       "<!-- spend_mean -->\n",
       "<g id=\"node8\" class=\"node\">\n",
       "<title>spend_mean</title>\n",
       "<path fill=\"#b4d8e4\" stroke=\"black\" d=\"M534.5,-146C534.5,-146 452.5,-146 452.5,-146 446.5,-146 440.5,-140 440.5,-134 440.5,-134 440.5,-94 440.5,-94 440.5,-88 446.5,-82 452.5,-82 452.5,-82 534.5,-82 534.5,-82 540.5,-82 546.5,-88 546.5,-94 546.5,-94 546.5,-134 546.5,-134 546.5,-140 540.5,-146 534.5,-146\"/>\n",
       "<text text-anchor=\"start\" x=\"451.5\" y=\"-124.8\" font-family=\"Helvetica,sans-Serif\" font-weight=\"bold\" font-size=\"14.00\">spend_mean</text>\n",
       "<text text-anchor=\"start\" x=\"480.5\" y=\"-96.8\" font-family=\"Helvetica,sans-Serif\" font-style=\"italic\" font-size=\"14.00\">float</text>\n",
       "</g>\n",
       "<!-- spend&#45;&gt;spend_mean -->\n",
       "<g id=\"edge11\" class=\"edge\">\n",
       "<title>spend&#45;&gt;spend_mean</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M393.18,-140.39C404.91,-137.28 417.83,-133.84 430.35,-130.51\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"431.52,-133.83 440.28,-127.88 429.72,-127.06 431.52,-133.83\"/>\n",
       "</g>\n",
       "<!-- spend_std_dev&#45;&gt;spend_zero_mean_unit_variance -->\n",
       "<g id=\"edge6\" class=\"edge\">\n",
       "<title>spend_std_dev&#45;&gt;spend_zero_mean_unit_variance</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M554.04,-38.32C607.01,-43.94 686.27,-52.34 753.49,-59.47\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"753.43,-62.99 763.74,-60.56 754.17,-56.03 753.43,-62.99\"/>\n",
       "</g>\n",
       "<!-- spend_mean&#45;&gt;spend_zero_mean -->\n",
       "<g id=\"edge4\" class=\"edge\">\n",
       "<title>spend_mean&#45;&gt;spend_zero_mean</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M546.58,-114C558.37,-114 571.15,-114 583.78,-114\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"583.86,-117.5 593.86,-114 583.86,-110.5 583.86,-117.5\"/>\n",
       "</g>\n",
       "<!-- _base_df_inputs -->\n",
       "<g id=\"node9\" class=\"node\">\n",
       "<title>_base_df_inputs</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" stroke-dasharray=\"5,2\" points=\"155,-215.5 0,-215.5 0,-170.5 155,-170.5 155,-215.5\"/>\n",
       "<text text-anchor=\"start\" x=\"15.5\" y=\"-188.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">base_df_location</text>\n",
       "<text text-anchor=\"start\" x=\"124.5\" y=\"-188.8\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">str</text>\n",
       "</g>\n",
       "<!-- _base_df_inputs&#45;&gt;base_df -->\n",
       "<g id=\"edge2\" class=\"edge\">\n",
       "<title>_base_df_inputs&#45;&gt;base_df</title>\n",
       "<path fill=\"none\" stroke=\"black\" d=\"M155.26,-193C161.54,-193 167.8,-193 173.87,-193\"/>\n",
       "<polygon fill=\"black\" stroke=\"black\" points=\"173.88,-196.5 183.88,-193 173.88,-189.5 173.88,-196.5\"/>\n",
       "</g>\n",
       "<!-- input -->\n",
       "<g id=\"node10\" class=\"node\">\n",
       "<title>input</title>\n",
       "<polygon fill=\"none\" stroke=\"black\" stroke-dasharray=\"5,2\" points=\"104.5,-325.5 50.5,-325.5 50.5,-288.5 104.5,-288.5 104.5,-325.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"77.5\" y=\"-303.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">input</text>\n",
       "</g>\n",
       "<!-- function -->\n",
       "<g id=\"node11\" class=\"node\">\n",
       "<title>function</title>\n",
       "<path fill=\"#b4d8e4\" stroke=\"black\" d=\"M100.5,-270.5C100.5,-270.5 54.5,-270.5 54.5,-270.5 48.5,-270.5 42.5,-264.5 42.5,-258.5 42.5,-258.5 42.5,-245.5 42.5,-245.5 42.5,-239.5 48.5,-233.5 54.5,-233.5 54.5,-233.5 100.5,-233.5 100.5,-233.5 106.5,-233.5 112.5,-239.5 112.5,-245.5 112.5,-245.5 112.5,-258.5 112.5,-258.5 112.5,-264.5 106.5,-270.5 100.5,-270.5\"/>\n",
       "<text text-anchor=\"middle\" x=\"77.5\" y=\"-248.3\" font-family=\"Helvetica,sans-Serif\" font-size=\"14.00\">function</text>\n",
       "</g>\n",
       "</g>\n",
       "</svg>\n"
      ],
      "text/plain": [
       "<graphviz.graphs.Digraph at 0x1664d0790>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dr.display_all_functions()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "hamilton-vaex",
   "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": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}