Channels_first_analysis_1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "644f9627",
   "metadata": {},
   "source": [
    "# Forwards data analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "335eea86",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f5e6b6f4",
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "af1ea2de",
   "metadata": {},
   "source": [
    "**Uploading forwards csv**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "9ce31f12",
   "metadata": {},
   "outputs": [],
   "source": [
    "forwards_path = '../Datos/forwards_2022-06-29.csv'\n",
    "df_forwards = pd.read_csv(forwards_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b781930b",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# df_forwards.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "d7f77c14",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "19561"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_forwards.shape[0]\n",
    "# this is the number of successful forwards your node has recorded"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "84ec488d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# changing the type of the channel ids to be displayed correctly\n",
    "df_forwards[['outgoing_channel_id', 'incoming_channel_id']] = df_forwards[['outgoing_channel_id', \n",
    "                                                                           'incoming_channel_id']].astype(np.int64)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e995b9e2",
   "metadata": {},
   "source": [
    "## Incoming/Outgoing/Balanced tagging task"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f76750fa",
   "metadata": {},
   "source": [
    "**Creating df for this task**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "06262754",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(185, 4)"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "task1_df = pd.DataFrame(index=df_forwards['outgoing_channel_id'].value_counts().index, \n",
    "                        columns=['times_incoming', \n",
    "                                 'times_outgoing', \n",
    "                                 'total_incoming_amt',\n",
    "                                 'total_outgoing_amt'])\n",
    "task1_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "b3875727",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# task1_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "3178c9b3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# group by incoming/outgoing chhanels\n",
    "group_incoming = df_forwards.groupby('incoming_channel_id')['incoming_amount_msat']\n",
    "group_outgoing = df_forwards.groupby('outgoing_channel_id')['outgoing_amount_msat']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "3b3ab70e",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Fill task1_df \n",
    "task1_df['times_incoming'] = df_forwards['incoming_channel_id'].value_counts()\n",
    "task1_df['times_outgoing'] = df_forwards['outgoing_channel_id'].value_counts()\n",
    "task1_df['total_incoming_amt'] = group_incoming.aggregate('sum')\n",
    "task1_df['total_outgoing_amt'] = group_outgoing.aggregate('sum')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "4d798198",
   "metadata": {},
   "outputs": [],
   "source": [
    "# task1_df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a0e997a3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Searching for nan values\n",
    "task1_df.isnull().values.any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "15d075ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "# replace nan values for 0s\n",
    "task1_df.fillna(0, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "02a59216",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Searching for nan values\n",
    "task1_df.isnull().values.any()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e66edab",
   "metadata": {},
   "source": [
    "## Balanced metric\n",
    "The **Balanced metric** shows whether a channel is balanced. Here, we say a channel is balanced when the forwarding activity is going both ways. It is calculated by:<br>\n",
    "<center><i>balanced_rate = min(inbound, outbound) / max(inbound, outbound)</i></center>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "4ab040fb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    185.000000\n",
       "mean       0.593908\n",
       "std        0.354774\n",
       "min        0.000000\n",
       "25%        0.284634\n",
       "50%        0.710571\n",
       "75%        0.905679\n",
       "max        0.998764\n",
       "dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Total amount routed\n",
    "total = task1_df['total_incoming_amt'] + task1_df['total_outgoing_amt']\n",
    "# Compute balance rate of the channels\n",
    "balanced_metric = np.minimum(task1_df['total_incoming_amt'], task1_df['total_outgoing_amt']) / np.maximum(task1_df['total_incoming_amt'], task1_df['total_outgoing_amt'])\n",
    "balanced_metric.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "499bccc6",
   "metadata": {},
   "outputs": [],
   "source": [
    "# balanced_metric.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fb9c6e5b",
   "metadata": {},
   "source": [
    "The `balanced_metric` is in a scale between 0-1. <br><br>\n",
    "Now let's plot the following: *(inbound/total_amount_routed) vs (outbound/total_amount_routed)*\n",
    "\n",
    "This will show a line where in the middle the balanced channels will be placed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "199337b5",
   "metadata": {},
   "outputs": [],
   "source": [
    "# This is just for style to the charts\n",
    "import seaborn as sns\n",
    "\n",
    "sns.set_style('darkgrid') # darkgrid, white grid, dark, white and ticks\n",
    "plt.rc('axes', titlesize=16)     # fontsize of the axes title\n",
    "plt.rc('axes', labelsize=13)    # fontsize of the x and y labels\n",
    "plt.rc('xtick', labelsize=12)    # fontsize of the tick labels\n",
    "plt.rc('ytick', labelsize=12)    # fontsize of the tick labels\n",
    "plt.rc('legend', fontsize=12)    # legend fontsize\n",
    "plt.rc('font', size=12)          # controls default text sizes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "ad7f1ac0",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8, 6))\n",
    "x = task1_df['total_incoming_amt']/total\n",
    "y = task1_df['total_outgoing_amt']/total\n",
    "plt.scatter(x, y, label='all channels')\n",
    "plt.xlabel('inbound amount msats / total')\n",
    "plt.ylabel('outbound amount msats / total')\n",
    "plt.legend()\n",
    "_ = plt.title('Channels inbound & outbound activity comparison')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0afcbc90",
   "metadata": {},
   "source": [
    "We clearly observe a big cluster in the middle, which represents the balanced channels based on their forwarding activity. We can make sure of this if we plot the points with a balanced rate close to 1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "a00abceb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Top 60 channels with best balanced rate\n",
    "top60 = balanced_metric.sort_values(ascending=False)[:60]\n",
    "t60_x = task1_df.loc[top60.index, 'total_incoming_amt']\n",
    "t60_y = task1_df.loc[top60.index, 'total_outgoing_amt']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "d940973a",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,6))\n",
    "x = task1_df['total_incoming_amt']/total\n",
    "y = task1_df['total_outgoing_amt']/total\n",
    "plt.scatter(x, y)\n",
    "plt.scatter(t60_x/total, t60_y/total)\n",
    "plt.xlabel('inbound amount msats / total')\n",
    "plt.ylabel('outbound amount msats / total')\n",
    "_ = plt.title('Channels inbound & outbound activity comparison')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d4d1b578",
   "metadata": {},
   "source": [
    "We can confirm that this balanced channels analysis is correct"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5d40e81c",
   "metadata": {},
   "source": [
    "## Balanced rate vs normalized volume\n",
    "\n",
    "We've already stablished a balanced metric to show which channels are balanced. If we plot it against a normalized version of the total volume, i.e. inbound + outbound, we'll see which channels are both balanced and highly active."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "af3b8aae",
   "metadata": {},
   "outputs": [],
   "source": [
    "# normalizing total volume\n",
    "total_volume = total.values\n",
    "norm = total_volume/np.linalg.norm(total_volume)\n",
    "norm_volume = pd.Series(norm, index=total.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "080fe545",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,6))\n",
    "\n",
    "plt.scatter(balanced_metric, norm_volume)\n",
    "plt.xlabel('Balanced rate')\n",
    "plt.ylabel('Normalized volume')\n",
    "\n",
    "x_avg = (balanced_metric.min() + balanced_metric.max()) / 2\n",
    "y_avg = (norm_volume.min() + norm_volume.max())/2\n",
    "plt.axvline(x_avg, c='black', lw=1)\n",
    "plt.axhline(y_avg, c='black', lw=1)\n",
    "_ = plt.title('Balanced rate vs Normalized volume')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6dc3fe47",
   "metadata": {},
   "source": [
    "We observe a 4 tile grid, where in Q1 are located the most balanced and highly active channels"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a0e6e7f3",
   "metadata": {},
   "source": [
    "We can see who these channels are:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "b04f5a50",
   "metadata": {},
   "outputs": [],
   "source": [
    "balanced_metric.name, norm_volume.name = \"balanced_rate\", \"normalized_volume\"\n",
    "channels_comparison = pd.merge(balanced_metric, norm_volume, left_index=True, right_index=True)\n",
    "# query for channels in Q1\n",
    "mask = (channels_comparison['balanced_rate'] > 0.5) & (channels_comparison['normalized_volume'] > y_avg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "81657f90",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Uncomment the cell below to see the channels that are the most balanced and highly active"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "5c800091",
   "metadata": {},
   "outputs": [],
   "source": [
    "# channels_comparison[mask]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1c642730",
   "metadata": {},
   "source": [
    "## Balanced rate vs normalized revenue\n",
    "\n",
    "We can do a similar this as above but with a normalized version of the total revenue earned per channels instead of the volume. This will show what channels are both balanced and generate value overall."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "d9d61ccb",
   "metadata": {},
   "outputs": [],
   "source": [
    "# total revenue\n",
    "total_revenue_df = df_forwards.groupby('outgoing_channel_id')['fee_msat'].aggregate('sum')\n",
    "# normalizing revenue\n",
    "total_revenue = total_revenue_df.values\n",
    "norm = total_revenue/np.linalg.norm(total_revenue)\n",
    "norm_revenue = pd.Series(norm, index=total.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "1a69cf14",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,6))\n",
    "plt.scatter(balanced_metric, norm_revenue)\n",
    "\n",
    "plt.xlabel('Balanced metric')\n",
    "plt.ylabel('Normalized total revenue')\n",
    "# 4 tile grid\n",
    "x_avg = (balanced_metric.min() + balanced_metric.max()) / 2\n",
    "y_avg = (norm_revenue.min() + norm_revenue.max())/2\n",
    "plt.axvline(x_avg, c='black', lw=1)\n",
    "plt.axhline(y_avg, c='black', lw=1)\n",
    "_ = plt.title('Normalized total revenue vs balanced metric')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ecf34611",
   "metadata": {},
   "source": [
    "We also observe a 4 tile grid where in Q1 are the channels that are the most balanced and generate the most value."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d08b21d6",
   "metadata": {},
   "source": [
    "We can see who these channels are:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "3a5635a0",
   "metadata": {},
   "outputs": [],
   "source": [
    "norm_revenue.name = \"normalized_revenue\"\n",
    "channels_comparison = pd.merge(channels_comparison, norm_revenue, left_index=True, right_index=True)\n",
    "# query for channels in Q1\n",
    "mask = (channels_comparison['balanced_rate'] > 0.5) & (channels_comparison['normalized_revenue'] > y_avg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "1d7fe907",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Uncomment the cell below to see the channels that are the most balanced and generate the most value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "43d01365",
   "metadata": {},
   "outputs": [],
   "source": [
    "# channels_comparison[mask]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d3ee2522",
   "metadata": {},
   "source": [
    "# HTLC data analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62f91f5a",
   "metadata": {},
   "source": [
    "**Uploading htlc dataset**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "a3d18599",
   "metadata": {},
   "outputs": [],
   "source": [
    "htlc_path = '../Datos/htlc_2022-06-29.csv'\n",
    "df_htlc = pd.read_csv(htlc_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "763e7d06",
   "metadata": {},
   "outputs": [],
   "source": [
    "# df_htlc.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "92077cf7",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_htlc[['outgoing_channel_id', 'incoming_channel_id']] = df_htlc[['outgoing_channel_id', \n",
    "                                                                   'incoming_channel_id']].astype(np.int64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "e14f0f49",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['FORWARD', 'RECEIVE', 'SEND', '3', '2', '1', 3, 2, 1], dtype=object)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_htlc['event_origin'].unique()\n",
    "# Forward events are labeld as 'FORWARD' & '3'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "52316cb7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Filtering df to have only forward events\n",
    "mask = ((df_htlc['event_origin'] == 'FORWARD') | (df_htlc['event_origin'] == '3'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "be66050a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['FORWARD', '3'], dtype=object)"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_htlc[mask]['event_origin'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "0f77dadc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1052356 htlcs\n"
     ]
    }
   ],
   "source": [
    "# Apply filter\n",
    "htlc_forwards = df_htlc[mask]\n",
    "print(f'{htlc_forwards.shape[0]} htlcs')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "5dcb6d79",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fails\n",
    "ForwardFailEvents = htlc_forwards[htlc_forwards['event_type'] == 'ForwardFailEvent']\n",
    "df_failures = ForwardFailEvents.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "d8e47d3e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This node has recorded 497639 htlcs failures\n"
     ]
    }
   ],
   "source": [
    "print(f'This node has recorded {ForwardFailEvents.shape[0]} htlcs failures')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "51999e53",
   "metadata": {},
   "source": [
    "## Success rate"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dfad6f38",
   "metadata": {},
   "source": [
    "This measures how successful a channel is, in a scale from 0-1, where a 1 means the channel has successfully forwarded all payments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "e0b2852c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# failures\n",
    "failures = df_failures.groupby('outgoing_channel_id').size()\n",
    "# forward attempts\n",
    "mask = htlc_forwards['event_type'] == 'ForwardEvent'\n",
    "fwd_att = htlc_forwards[mask].groupby('outgoing_channel_id').size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "d70625a7",
   "metadata": {},
   "outputs": [],
   "source": [
    "success_rate = (1-(failures/fwd_att))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "8b8ef48f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# success_rate.sort_values()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2108cfb3",
   "metadata": {
    "scrolled": true
   },
   "source": [
    "## See failure rate of the nodes our channels are linked to"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "9b5a4443",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "channels_path = '../Datos/channel_events_2022-06-29.csv'\n",
    "df_channels = pd.read_csv(channels_path)\n",
    "df_channels['chan_id'] = df_channels['chan_id'].astype(np.int64)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "eb6d5054",
   "metadata": {},
   "outputs": [],
   "source": [
    "# df_channels.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "442a5043",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Node events dataset (node alias)\n",
    "nodes_path = '../Datos/node_events_2022-06-29.csv'\n",
    "df_nodes = pd.read_csv(nodes_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "d0f63f5d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# df_nodes.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "75bb7910",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fails\n",
    "ForwardFailEvents = htlc_forwards[htlc_forwards['event_type'] == 'ForwardFailEvent']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "94cb6b37",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ForwardFailEvents.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "d7fa4794",
   "metadata": {},
   "outputs": [],
   "source": [
    "# total\n",
    "total_fws = htlc_forwards.copy()\n",
    "total_fws.time = pd.to_datetime(total_fws.time)\n",
    "total_df = total_fws.set_index('time')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "c7061916",
   "metadata": {},
   "outputs": [],
   "source": [
    "ForwardFailEvents.time = pd.to_datetime(ForwardFailEvents.time)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "2640dfd6",
   "metadata": {},
   "outputs": [],
   "source": [
    "FFE_df = ForwardFailEvents.set_index('time')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "d535fe8b",
   "metadata": {},
   "outputs": [],
   "source": [
    "def add_pubkey(fails, channels):\n",
    "    channels = channels.drop_duplicates(['chan_id'], keep='last')\n",
    "    channels['chan_id'] = channels['chan_id'].astype(np.int64)\n",
    "    fails['pubkey'] = ''\n",
    "    out_chan_ids = fails['outgoing_channel_id'].unique()\n",
    "    \n",
    "    for i in out_chan_ids:\n",
    "        \n",
    "        for z, j in channels[['chan_id', 'pub_key']].iterrows():\n",
    "            \n",
    "            if i == j['chan_id']:\n",
    "                fails.loc[fails['outgoing_channel_id'] == i, 'pubkey'] = j['pub_key']\n",
    "                \n",
    "                break\n",
    "                \n",
    "    return fails['pubkey']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "3801d4c2",
   "metadata": {},
   "outputs": [],
   "source": [
    "total_df['pubkey'] = add_pubkey(total_df, df_channels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "123bb6c9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1052356, 15)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "7dad2022",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "FFE_df['pubkey'] = add_pubkey(FFE_df, df_channels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "9c4a00db",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ForwardFailEvents.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "877302e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# FFE_df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "93a13afe",
   "metadata": {},
   "outputs": [],
   "source": [
    "failures = FFE_df.groupby('pubkey').size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "9da29065",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pubkey\n",
       "0203f3e0fdaa52ca5c804d145b6a47e5a620c02123106578230061deb212e720bd    3097\n",
       "0205306aa071fc026a4884c210d99d886f199068a6ccc65ec1bde085f846be3641    4098\n",
       "dtype: int64"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "failures.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "f1759443",
   "metadata": {},
   "outputs": [],
   "source": [
    "total = total_df[total_df['event_type'] == 'ForwardEvent'].groupby('pubkey').size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "0bd39f4f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pubkey\n",
       "0203f3e0fdaa52ca5c804d145b6a47e5a620c02123106578230061deb212e720bd    3104\n",
       "0205306aa071fc026a4884c210d99d886f199068a6ccc65ec1bde085f846be3641    4113\n",
       "dtype: int64"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "3be12a66",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pubkey\n",
       "037659a0ac8eb3b8d0a720114efc861d3a940382dcfa1403746b4f8f6b2e8810ba    0.698182\n",
       "0253064653c3c2b52ea76201a24771580ac562902fb603b7b077c347a2420d7190    0.694900\n",
       "021c97a90a411ff2b10dc2a8e32de2f29d2fa49d41bfbb52bd416e460db0747d0d    0.423077\n",
       "030ed0e8c35b473e0cdf3b384262688be36a2a7bb61ecf34c836df5b29c81dcd79    0.374199\n",
       "022ba42f1a9a7b5ee9f3a783b02ce401c06930414fd1386c714a9987d51f3b475a    0.266667\n",
       "dtype: float64"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "success_rate_failures = (1-(failures/total))\n",
    "success_rate = success_rate_failures.copy()\n",
    "success_rate.sort_values(ascending=False).iloc[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "54835dc4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "118"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "success_rate.reset_index()['pubkey'].nunique()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6b0b7881",
   "metadata": {},
   "source": [
    "This are some nodes of interest in our special case"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "b76bedd5",
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "interesting_nodes = ['c-otto.de', 'SWISS.SOVEREIGN', \n",
    "                     'zero fee routing | CLN', 'WalletOfSatoshi.com', \n",
    "                     'bfx-lnd0', 'bfx-lnd1', 'okex', \n",
    "                    'BCash_Is_Trash']\n",
    "pub_keys=['027ce055380348d7812d2ae7745701c9f93e70c1adeb2657f053f91df4f2843c71',\n",
    "          '03d4e028a0d4a90868ec202ab684fb0085779defea9ca7553e06146557631eec20',\n",
    "          '038fe1bd966b5cb0545963490c631eaa1924e2c4c0ea4e7dcb5d4582a1e7f2f1a5',\n",
    "          '035e4ff418fc8b5554c5d9eea66396c227bd429a3251c8cbc711002ba215bfc226',\n",
    "          '033d8656219478701227199cbd6f670335c8d408a92ae88b962c49d4dc0e83e025',\n",
    "          '03cde60a6323f7122d5178255766e38114b4722ede08f7c9e0c5df9b912cc201d6',\n",
    "          '0294ac3e099def03c12a37e30fe5364b1223fd60069869142ef96580c8439c2e0a',\n",
    "          '0298f6074a454a1f5345cb2a7c6f9fce206cd0bf675d177cdbf0ca7508dd28852f']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a87fcf87",
   "metadata": {},
   "source": [
    "## SEE ALL NODES BEHAVIOUR IN OUR GRAPH\n",
    "\n",
    "See the metrics obtained so far for the nodes our channels are linked to"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "3a6897de",
   "metadata": {},
   "outputs": [],
   "source": [
    "def add_node_alias(rate_df, nodes):\n",
    "\n",
    "    rate_df = rate_df.reset_index()\n",
    "    rate_df['node_alias'] = ''\n",
    "#     print(rate_df)\n",
    "    nodes_pubkeys = rate_df['pubkey'].unique()\n",
    "    \n",
    "    for i in nodes_pubkeys:\n",
    "        \n",
    "        for ix, j in nodes[['pub_key', 'alias']].iterrows():\n",
    "            \n",
    "            if i == j['pub_key']:\n",
    "                rate_df.loc[rate_df['pubkey'] == i, 'node_alias'] = j['alias']\n",
    "                \n",
    "                break\n",
    "    rate_df = rate_df.set_index('pubkey')\n",
    "    \n",
    "    return rate_df['node_alias']\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "4c8d2989",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "pubkey\n",
       "0203f3e0fdaa52ca5c804d145b6a47e5a620c02123106578230061deb212e720bd    0.002255\n",
       "0205306aa071fc026a4884c210d99d886f199068a6ccc65ec1bde085f846be3641    0.003647\n",
       "dtype: float64"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "success_rate.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "a5ee88be",
   "metadata": {},
   "outputs": [],
   "source": [
    "s_r_df = success_rate.copy()\n",
    "s_r_df = s_r_df.reset_index()\n",
    "s_r_df['alias'] = ''\n",
    "s_r_df.set_index('pubkey', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "2773f8af",
   "metadata": {},
   "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>0</th>\n",
       "      <th>alias</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pubkey</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0203f3e0fdaa52ca5c804d145b6a47e5a620c02123106578230061deb212e720bd</th>\n",
       "      <td>0.002255</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0205306aa071fc026a4884c210d99d886f199068a6ccc65ec1bde085f846be3641</th>\n",
       "      <td>0.003647</td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                           0 alias\n",
       "pubkey                                                            \n",
       "0203f3e0fdaa52ca5c804d145b6a47e5a620c0212310657...  0.002255      \n",
       "0205306aa071fc026a4884c210d99d886f199068a6ccc65...  0.003647      "
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_r_df.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "002c7e76",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# Adding node alias \n",
    "s_r_df['alias'] = add_node_alias(success_rate, df_nodes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "2bcbe06c",
   "metadata": {},
   "outputs": [],
   "source": [
    "s_r_df.rename({0:'success_rate'}, axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "92342f48",
   "metadata": {},
   "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>success_rate</th>\n",
       "      <th>alias</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pubkey</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>037659a0ac8eb3b8d0a720114efc861d3a940382dcfa1403746b4f8f6b2e8810ba</th>\n",
       "      <td>0.698182</td>\n",
       "      <td>ln.nicehash.com [Nicehash]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0253064653c3c2b52ea76201a24771580ac562902fb603b7b077c347a2420d7190</th>\n",
       "      <td>0.694900</td>\n",
       "      <td>noserver4u</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>021c97a90a411ff2b10dc2a8e32de2f29d2fa49d41bfbb52bd416e460db0747d0d</th>\n",
       "      <td>0.423077</td>\n",
       "      <td>LOOP</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    success_rate  \\\n",
       "pubkey                                                             \n",
       "037659a0ac8eb3b8d0a720114efc861d3a940382dcfa140...      0.698182   \n",
       "0253064653c3c2b52ea76201a24771580ac562902fb603b...      0.694900   \n",
       "021c97a90a411ff2b10dc2a8e32de2f29d2fa49d41bfbb5...      0.423077   \n",
       "\n",
       "                                                                         alias  \n",
       "pubkey                                                                          \n",
       "037659a0ac8eb3b8d0a720114efc861d3a940382dcfa140...  ln.nicehash.com [Nicehash]  \n",
       "0253064653c3c2b52ea76201a24771580ac562902fb603b...                  noserver4u  \n",
       "021c97a90a411ff2b10dc2a8e32de2f29d2fa49d41bfbb5...                        LOOP  "
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s_r_df.sort_values(by='success_rate', ascending=False).head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ba6fe99",
   "metadata": {},
   "source": [
    "**Adding balanced metric**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "13a2194f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def add_pubkey2(series, channels):\n",
    "    channels = channels.drop_duplicates(['chan_id'], keep='last')\n",
    "    channels['chan_id'] = channels['chan_id'].astype(np.int64)\n",
    "    series = series.reset_index().rename({'index':'outgoing_channel_id',\n",
    "                                      0:'balanced_metric'}, axis=1)\n",
    "    series['pubkey'] = ''\n",
    "    out_chan_ids = series['outgoing_channel_id'].unique()\n",
    "    \n",
    "    for i in out_chan_ids:\n",
    "        \n",
    "        for z, j in channels[['chan_id', 'pub_key']].iterrows():\n",
    "            \n",
    "            if i == j['chan_id']:\n",
    "                series.loc[series['outgoing_channel_id'] == i, 'pubkey'] = j['pub_key']\n",
    "                \n",
    "                break\n",
    "#     print(series.index)     \n",
    "    return series['pubkey']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "4cb13ded",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "b_m_df = balanced_metric.copy()\n",
    "b_m_df = b_m_df.reset_index()\n",
    "b_m_df['pubkey'] = add_pubkey2(balanced_metric, df_channels)\n",
    "\n",
    "b_m_df.set_index('pubkey', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "f21d8cb1",
   "metadata": {},
   "outputs": [],
   "source": [
    "# droping duplicates\n",
    "# duplicates occur because sometimes you close a channel and open another one with the same peer.\n",
    "b_m_df = b_m_df.reset_index().sort_values(by='balanced_rate').drop_duplicates('pubkey', keep='last')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "50e1f39d",
   "metadata": {},
   "outputs": [],
   "source": [
    "b_m_df.set_index('pubkey', inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15aae280",
   "metadata": {},
   "source": [
    "**Join balanced metric and success rate in the same df**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "ad7cb699",
   "metadata": {},
   "outputs": [],
   "source": [
    "comparing_node_df = pd.merge(s_r_df, b_m_df['balanced_rate'], left_index=True, right_index=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "3f925f12",
   "metadata": {
    "scrolled": false
   },
   "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>success_rate</th>\n",
       "      <th>alias</th>\n",
       "      <th>balanced_rate</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pubkey</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>037659a0ac8eb3b8d0a720114efc861d3a940382dcfa1403746b4f8f6b2e8810ba</th>\n",
       "      <td>0.698182</td>\n",
       "      <td>ln.nicehash.com [Nicehash]</td>\n",
       "      <td>0.730948</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0253064653c3c2b52ea76201a24771580ac562902fb603b7b077c347a2420d7190</th>\n",
       "      <td>0.694900</td>\n",
       "      <td>noserver4u</td>\n",
       "      <td>0.998764</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>021c97a90a411ff2b10dc2a8e32de2f29d2fa49d41bfbb52bd416e460db0747d0d</th>\n",
       "      <td>0.423077</td>\n",
       "      <td>LOOP</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>030ed0e8c35b473e0cdf3b384262688be36a2a7bb61ecf34c836df5b29c81dcd79</th>\n",
       "      <td>0.374199</td>\n",
       "      <td>FriendsWithBenefits</td>\n",
       "      <td>0.959066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>022ba42f1a9a7b5ee9f3a783b02ce401c06930414fd1386c714a9987d51f3b475a</th>\n",
       "      <td>0.266667</td>\n",
       "      <td>Routing24/7Bitcoin</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>026db2cbf3d8ab4a4c01eed2df432d20cf0a13136402097574209d2595cb9e9d93</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>⚡️ bCyber ⚡️</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>02fc63f7b07cc69c756d432f77ce36ce4e5dbe24f43f3638f38f726d7ac1d9ecc4</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>JDs.Connection</td>\n",
       "      <td>0.981045</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0296b46141cd8baf13f3eff9bb217c5f62ce0a871886559d661af0ef422c042d4b</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>EDON</td>\n",
       "      <td>0.874059</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0382b31dcff337311bf919411c5073c9c9a129890993f94f4a16eaaeffd91c7788</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>lnnodeinsight.com</td>\n",
       "      <td>0.746872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>035a95bdafeaec089ded020248939b636296ea32e8cf9a0b6af70e9e517f703fce</th>\n",
       "      <td>NaN</td>\n",
       "      <td>otech</td>\n",
       "      <td>0.000051</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>114 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    success_rate  \\\n",
       "pubkey                                                             \n",
       "037659a0ac8eb3b8d0a720114efc861d3a940382dcfa140...      0.698182   \n",
       "0253064653c3c2b52ea76201a24771580ac562902fb603b...      0.694900   \n",
       "021c97a90a411ff2b10dc2a8e32de2f29d2fa49d41bfbb5...      0.423077   \n",
       "030ed0e8c35b473e0cdf3b384262688be36a2a7bb61ecf3...      0.374199   \n",
       "022ba42f1a9a7b5ee9f3a783b02ce401c06930414fd1386...      0.266667   \n",
       "...                                                          ...   \n",
       "026db2cbf3d8ab4a4c01eed2df432d20cf0a13136402097...      0.000000   \n",
       "02fc63f7b07cc69c756d432f77ce36ce4e5dbe24f43f363...      0.000000   \n",
       "0296b46141cd8baf13f3eff9bb217c5f62ce0a871886559...      0.000000   \n",
       "0382b31dcff337311bf919411c5073c9c9a129890993f94...      0.000000   \n",
       "035a95bdafeaec089ded020248939b636296ea32e8cf9a0...           NaN   \n",
       "\n",
       "                                                                         alias  \\\n",
       "pubkey                                                                           \n",
       "037659a0ac8eb3b8d0a720114efc861d3a940382dcfa140...  ln.nicehash.com [Nicehash]   \n",
       "0253064653c3c2b52ea76201a24771580ac562902fb603b...                  noserver4u   \n",
       "021c97a90a411ff2b10dc2a8e32de2f29d2fa49d41bfbb5...                        LOOP   \n",
       "030ed0e8c35b473e0cdf3b384262688be36a2a7bb61ecf3...         FriendsWithBenefits   \n",
       "022ba42f1a9a7b5ee9f3a783b02ce401c06930414fd1386...          Routing24/7Bitcoin   \n",
       "...                                                                        ...   \n",
       "026db2cbf3d8ab4a4c01eed2df432d20cf0a13136402097...                ⚡️ bCyber ⚡️   \n",
       "02fc63f7b07cc69c756d432f77ce36ce4e5dbe24f43f363...              JDs.Connection   \n",
       "0296b46141cd8baf13f3eff9bb217c5f62ce0a871886559...                        EDON   \n",
       "0382b31dcff337311bf919411c5073c9c9a129890993f94...           lnnodeinsight.com   \n",
       "035a95bdafeaec089ded020248939b636296ea32e8cf9a0...                       otech   \n",
       "\n",
       "                                                    balanced_rate  \n",
       "pubkey                                                             \n",
       "037659a0ac8eb3b8d0a720114efc861d3a940382dcfa140...       0.730948  \n",
       "0253064653c3c2b52ea76201a24771580ac562902fb603b...       0.998764  \n",
       "021c97a90a411ff2b10dc2a8e32de2f29d2fa49d41bfbb5...       0.000000  \n",
       "030ed0e8c35b473e0cdf3b384262688be36a2a7bb61ecf3...       0.959066  \n",
       "022ba42f1a9a7b5ee9f3a783b02ce401c06930414fd1386...       0.000000  \n",
       "...                                                           ...  \n",
       "026db2cbf3d8ab4a4c01eed2df432d20cf0a13136402097...       0.000000  \n",
       "02fc63f7b07cc69c756d432f77ce36ce4e5dbe24f43f363...       0.981045  \n",
       "0296b46141cd8baf13f3eff9bb217c5f62ce0a871886559...       0.874059  \n",
       "0382b31dcff337311bf919411c5073c9c9a129890993f94...       0.746872  \n",
       "035a95bdafeaec089ded020248939b636296ea32e8cf9a0...       0.000051  \n",
       "\n",
       "[114 rows x 3 columns]"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "comparing_node_df.sort_values(by='success_rate', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "59602ee5",
   "metadata": {},
   "outputs": [],
   "source": [
    "best_success_rate_nodes = success_rate.sort_values(ascending=False).iloc[:30]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "420eec91",
   "metadata": {},
   "outputs": [],
   "source": [
    "comparing_node_df['total_Forwards'] = total_df.groupby('pubkey')['event_type'].value_counts()[:,'ForwardEvent']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "5c01c172",
   "metadata": {},
   "outputs": [],
   "source": [
    "# comparing_node_df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "5dec8a2c",
   "metadata": {},
   "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>success_rate</th>\n",
       "      <th>alias</th>\n",
       "      <th>balanced_rate</th>\n",
       "      <th>total_Forwards</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pubkey</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0298f6074a454a1f5345cb2a7c6f9fce206cd0bf675d177cdbf0ca7508dd28852f</th>\n",
       "      <td>0.009389</td>\n",
       "      <td>BCash_Is_Trash</td>\n",
       "      <td>0.889287</td>\n",
       "      <td>32486.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>027ce055380348d7812d2ae7745701c9f93e70c1adeb2657f053f91df4f2843c71</th>\n",
       "      <td>0.026347</td>\n",
       "      <td>c-otto.de</td>\n",
       "      <td>0.983971</td>\n",
       "      <td>29225.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>038fe1bd966b5cb0545963490c631eaa1924e2c4c0ea4e7dcb5d4582a1e7f2f1a5</th>\n",
       "      <td>0.035517</td>\n",
       "      <td>zero fee routing | c-lightning</td>\n",
       "      <td>0.992703</td>\n",
       "      <td>20863.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>033d8656219478701227199cbd6f670335c8d408a92ae88b962c49d4dc0e83e025</th>\n",
       "      <td>0.015461</td>\n",
       "      <td>bfx-lnd0</td>\n",
       "      <td>0.484106</td>\n",
       "      <td>3040.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>03d4e028a0d4a90868ec202ab684fb0085779defea9ca7553e06146557631eec20</th>\n",
       "      <td>0.032591</td>\n",
       "      <td>SWISS.SOVEREIGN</td>\n",
       "      <td>0.927653</td>\n",
       "      <td>1258.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0294ac3e099def03c12a37e30fe5364b1223fd60069869142ef96580c8439c2e0a</th>\n",
       "      <td>0.084677</td>\n",
       "      <td>okex</td>\n",
       "      <td>0.791212</td>\n",
       "      <td>744.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>035e4ff418fc8b5554c5d9eea66396c227bd429a3251c8cbc711002ba215bfc226</th>\n",
       "      <td>0.225564</td>\n",
       "      <td>WalletOfSatoshi.com</td>\n",
       "      <td>0.792625</td>\n",
       "      <td>133.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>03cde60a6323f7122d5178255766e38114b4722ede08f7c9e0c5df9b912cc201d6</th>\n",
       "      <td>0.166667</td>\n",
       "      <td>bfx-lnd1</td>\n",
       "      <td>0.040178</td>\n",
       "      <td>90.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    success_rate  \\\n",
       "pubkey                                                             \n",
       "0298f6074a454a1f5345cb2a7c6f9fce206cd0bf675d177...      0.009389   \n",
       "027ce055380348d7812d2ae7745701c9f93e70c1adeb265...      0.026347   \n",
       "038fe1bd966b5cb0545963490c631eaa1924e2c4c0ea4e7...      0.035517   \n",
       "033d8656219478701227199cbd6f670335c8d408a92ae88...      0.015461   \n",
       "03d4e028a0d4a90868ec202ab684fb0085779defea9ca75...      0.032591   \n",
       "0294ac3e099def03c12a37e30fe5364b1223fd600698691...      0.084677   \n",
       "035e4ff418fc8b5554c5d9eea66396c227bd429a3251c8c...      0.225564   \n",
       "03cde60a6323f7122d5178255766e38114b4722ede08f7c...      0.166667   \n",
       "\n",
       "                                                                             alias  \\\n",
       "pubkey                                                                               \n",
       "0298f6074a454a1f5345cb2a7c6f9fce206cd0bf675d177...                  BCash_Is_Trash   \n",
       "027ce055380348d7812d2ae7745701c9f93e70c1adeb265...                       c-otto.de   \n",
       "038fe1bd966b5cb0545963490c631eaa1924e2c4c0ea4e7...  zero fee routing | c-lightning   \n",
       "033d8656219478701227199cbd6f670335c8d408a92ae88...                        bfx-lnd0   \n",
       "03d4e028a0d4a90868ec202ab684fb0085779defea9ca75...                 SWISS.SOVEREIGN   \n",
       "0294ac3e099def03c12a37e30fe5364b1223fd600698691...                            okex   \n",
       "035e4ff418fc8b5554c5d9eea66396c227bd429a3251c8c...             WalletOfSatoshi.com   \n",
       "03cde60a6323f7122d5178255766e38114b4722ede08f7c...                        bfx-lnd1   \n",
       "\n",
       "                                                    balanced_rate  \\\n",
       "pubkey                                                              \n",
       "0298f6074a454a1f5345cb2a7c6f9fce206cd0bf675d177...       0.889287   \n",
       "027ce055380348d7812d2ae7745701c9f93e70c1adeb265...       0.983971   \n",
       "038fe1bd966b5cb0545963490c631eaa1924e2c4c0ea4e7...       0.992703   \n",
       "033d8656219478701227199cbd6f670335c8d408a92ae88...       0.484106   \n",
       "03d4e028a0d4a90868ec202ab684fb0085779defea9ca75...       0.927653   \n",
       "0294ac3e099def03c12a37e30fe5364b1223fd600698691...       0.791212   \n",
       "035e4ff418fc8b5554c5d9eea66396c227bd429a3251c8c...       0.792625   \n",
       "03cde60a6323f7122d5178255766e38114b4722ede08f7c...       0.040178   \n",
       "\n",
       "                                                    total_Forwards  \n",
       "pubkey                                                              \n",
       "0298f6074a454a1f5345cb2a7c6f9fce206cd0bf675d177...         32486.0  \n",
       "027ce055380348d7812d2ae7745701c9f93e70c1adeb265...         29225.0  \n",
       "038fe1bd966b5cb0545963490c631eaa1924e2c4c0ea4e7...         20863.0  \n",
       "033d8656219478701227199cbd6f670335c8d408a92ae88...          3040.0  \n",
       "03d4e028a0d4a90868ec202ab684fb0085779defea9ca75...          1258.0  \n",
       "0294ac3e099def03c12a37e30fe5364b1223fd600698691...           744.0  \n",
       "035e4ff418fc8b5554c5d9eea66396c227bd429a3251c8c...           133.0  \n",
       "03cde60a6323f7122d5178255766e38114b4722ede08f7c...            90.0  "
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Seeing the behaviour of the nodes of interest\n",
    "comparing_node_df.loc[pub_keys,:].sort_values(by='total_Forwards', ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21c3ca6e",
   "metadata": {},
   "source": [
    "We see that as the total number of forwards sent from a channel to that node increases, the success_rate decreases. This is only for the nodes of interest here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "c9cd820e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# We can see if the above idea generalizes with a scatter plot comparing those features\n",
    "plt.scatter(comparing_node_df['total_Forwards'], comparing_node_df['success_rate'])\n",
    "plt.xlabel('Total Forwards')\n",
    "plt.ylabel('Success rate')\n",
    "_ = plt.title('Forwards vs success rate')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e20b744b",
   "metadata": {},
   "source": [
    "We are only interested in routing nodes. So we might filter some data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "e762c8f5",
   "metadata": {},
   "outputs": [],
   "source": [
    "the_filter = comparing_node_df.describe().loc['25%', 'total_Forwards']\n",
    "# filter out channels whose total_forwards feature is less than the 25th quantile\n",
    "# and whose success_rate is considered as outlier (this can be subjective now)\n",
    "outlier_scc_rate = 0.35 # You can change this value based on the chart above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "446f21eb",
   "metadata": {},
   "outputs": [],
   "source": [
    "active_routing_nodes = comparing_node_df[(comparing_node_df['total_Forwards'] > the_filter) & (comparing_node_df['success_rate'] < outlier_scc_rate)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "240b64bb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# active_routing_nodes = comparing_node_df[comparing_node_df['total_Forwards'] > the_filter]\n",
    "\n",
    "plt.scatter(active_routing_nodes['total_Forwards'], active_routing_nodes['success_rate'])\n",
    "plt.xlabel('Total Forwards')\n",
    "plt.ylabel('Success rate')\n",
    "_ = plt.title('Forwards vs success rate')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5819101a",
   "metadata": {},
   "source": [
    "We observe that the channels with bigger success rate are also the ones with not that much total forwards. This might suggest that in order to increase the success rate of a channel, increasing the forward attempts of the channel has not a big weight on it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "40df050e",
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "f59b1175",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 648x504 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Compute the correlation matrix\n",
    "corr = active_routing_nodes[['success_rate','balanced_rate','total_Forwards']].corr()\n",
    "\n",
    "# Generate a mask for the upper triangle\n",
    "mask = np.triu(np.ones_like(corr, dtype=bool))\n",
    "\n",
    "# Set up the matplotlib figure\n",
    "f, ax = plt.subplots(figsize=(9, 7))\n",
    "\n",
    "# Generate a custom diverging colormap\n",
    "cmap = sns.diverging_palette(230, 20, as_cmap=True)\n",
    "\n",
    "# Draw the heatmap with the mask and correct aspect ratio\n",
    "sns.heatmap(corr, mask=mask, cmap=cmap, vmax=.3, center=0,\n",
    "            square=True, linewidths=.5, cbar_kws={\"shrink\": .5})\n",
    "\n",
    "_ = plt.title('Correlation Matrix')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "60d21d10",
   "metadata": {},
   "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>success_rate</th>\n",
       "      <th>balanced_rate</th>\n",
       "      <th>total_Forwards</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>success_rate</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.276779</td>\n",
       "      <td>-0.131584</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>balanced_rate</th>\n",
       "      <td>0.276779</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.213419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>total_Forwards</th>\n",
       "      <td>-0.131584</td>\n",
       "      <td>0.213419</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                success_rate  balanced_rate  total_Forwards\n",
       "success_rate        1.000000       0.276779       -0.131584\n",
       "balanced_rate       0.276779       1.000000        0.213419\n",
       "total_Forwards     -0.131584       0.213419        1.000000"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "corr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "id": "3503654e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,6))\n",
    "plt.scatter(active_routing_nodes['balanced_rate'], active_routing_nodes['success_rate'])\n",
    "plt.xlabel('balanced_rate')\n",
    "plt.ylabel('success_rate')\n",
    "_ = plt.title('relation between balanced and success metrics')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "e34d30ef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,6))\n",
    "plt.scatter(active_routing_nodes['balanced_rate'], active_routing_nodes['total_Forwards'])\n",
    "plt.xlabel('balanced rate')\n",
    "plt.ylabel('total forwards')\n",
    "_ = plt.title('relation between balanced and total forwards')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a3fe7de3",
   "metadata": {},
   "source": [
    "- We can see there's a positive correlation between the balanced rate and the success rate. When a channel is balanced, then  there's a chance it's success rate is better than other channels. It makes sense, since when you have a channel that, due to its outbound & inbound forwarding behaviour it balances naturally, then it can keep sending and receiving forwards, without them failing due to that channel.\n",
    "- We observe the total forwards and success rate have a weak negative correlation. Once again, we can say this might suggest that in order to increase the success rate of a channel, increasing the forward attempts of the channel has not a big weight on it.\n",
    "- We see in the correlation matrix a positive correlation between the balaced rate of a channel an its total forward attempts. The LND path-finding algorithm takes into account the degree to which former payment attempts have been successful. Then, if a channel has a good success rate, by the *relation between balanced and success metrics* described above, it might have a good balanced rate as well. If it performs well, then the forwards sent through that channel might increase, as the path-finding algorithm will take it into account more times."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "id": "cb4f8591",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Uncomment the cell below to save the active routing nodes df for future analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "fcefe25f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# active_routing_nodes.to_csv('active_routing_nodes.csv', encoding='utf-8')"
   ]
  }
 ],
 "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": "3.8.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}