bcdata workshop CloudPBX project code

asn_pairings.ipynb

{
  "cells": [
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "# ASN Pairings Mean MOS, Mean Qualfun, and Percentage \"Bad\" Calls",
      "execution_count": 1,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "import numpy as np\nimport pandas as pd\nimport matplotlib.pyplot as plt\n%matplotlib inline\nimport seaborn as sns\n#import geoip2.database\nimport ipaddress\nimport os\nimport dask.dataframe as dd",
      "execution_count": 2,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "# functions to get AS info\ndef getASobject(x):\n    ip = ipaddress.ip_address(x)\n    try: return readerASN.asn(str(ip))\n    except: return \"The address {} is not in the database.\".format(ip)\ndef getIP(x):\n    if type(x) == str: return x\n    return x.ip_address\ndef getASN(x):\n    if type(x) == str: return x\n    return x.autonomous_system_number\ndef getASorg(x):\n    if type(x) == str: return x\n    return x.autonomous_system_organization\ndef getInt(x):\n    return int(x)",
      "execution_count": 3,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "def getcount(row):\n    a=int(row['a_saddr'])\n    b=int(row['b_saddr'])\n    \n    df_asaddr=df['a_saddr']\n    df_new=df[df['a_saddr']==a][df['b_saddr']==b]\n\n    return len(df_new)\n\ndef getbadcount(row):\n    a=int(row['a_saddr'])\n    b=int(row['b_saddr'])\n    \n    df_asaddr=df_filtered['a_saddr']\n    df_new=df_filtered[df_filtered['a_saddr']==a][df_filtered['b_saddr']==b]\n\n    return len(df_new)\n\ndef getbqualfunmean(row):\n    a=int(row['a_saddr'])\n    b=int(row['b_saddr'])\n    \n    df_asaddr=df['a_saddr']\n    df_new=df[df['a_saddr']==a][df['b_saddr']==b]\n\n    return df_new['b_qualfun'].mean()\n\ndef getmosmean(row):\n    a=int(row['a_saddr'])\n    b=int(row['b_saddr'])\n    \n    df_asaddr=df['a_saddr']\n    df_new=df[df['a_saddr']==a][df['b_saddr']==b]\n\n    return df_new['b_mos_adapt_mult10'].mean()",
      "execution_count": 4,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "CSV_FILE_PATH = os.path.join('cdr_from_2018-05-01-order-calldate-asc.csv')\n\nHEADER = [\"ID\",\"calldate\",\"callend\",\"duration\",\"connect_duration\",\"progress_time\",\"first_rtp_time\",\"caller\",\n          \"caller_domain\",\"caller_reverse\",\"callername\",\"callername_reverse\",\"called\",\"called_domain\",\"called_reverse\",\n          \"sipcallerip\",\"sipcallerport\",\"sipcalledip\",\"sipcalledport\",\"whohanged\",\"bye\",\"lastSIPresponse_id\",\n          \"lastSIPresponseNum\",\"sighup\",\"dscp\",\"a_index\",\"b_index\",\"a_payload\",\"b_payload\",\"a_saddr\",\"b_saddr\",\n          \"a_received\",\"b_received\",\"a_lost\",\"b_lost\",\"a_ua_id\",\"b_ua_id\",\"a_avgjitter_mult10\",\"b_avgjitter_mult10\",\n          \"a_maxjitter\",\"b_maxjitter\",\"a_sl1\",\"a_sl2\",\"a_sl3\",\"a_sl4\",\"a_sl5\",\"a_sl6\",\"a_sl7\",\"a_sl8\",\"a_sl9\",\"a_sl10\",\n          \"a_d50\",\"a_d70\",\"a_d90\",\"a_d120\",\"a_d150\",\"a_d200\",\"a_d300\",\"b_sl1\",\"b_sl2\",\"b_sl3\",\"b_sl4\",\"b_sl5\",\"b_sl6\",\"b_sl7\",\n          \"b_sl8\",\"b_sl9\",\"b_sl10\",\"b_d50\",\"b_d70\",\"b_d90\",\"b_d120\",\"b_d150\",\"b_d200\",\"b_d300\",\"a_mos_lqo_mult10\",\n          \"b_mos_lqo_mult10\",\"a_mos_f1_mult10\",\"a_mos_f2_mult10\",\"a_mos_adapt_mult10\",\"b_mos_f1_mult10\",\"b_mos_f2_mult10\",\n          \"b_mos_adapt_mult10\",\"a_rtcp_loss\",\"a_rtcp_maxfr\",\"a_rtcp_avgfr_mult10\",\"a_rtcp_maxjitter\",\"a_rtcp_avgjitter_mult10\",\n          \"b_rtcp_loss\",\"b_rtcp_maxfr\",\"b_rtcp_avgfr_mult10\",\"b_rtcp_maxjitter\",\"b_rtcp_avgjitter_mult10\",\"a_last_rtp_from_end\",\n          \"b_last_rtp_from_end\",\"payload\",\"jitter_mult10\",\"mos_min_mult10\",\"a_mos_min_mult10\",\"b_mos_min_mult10\",\n          \"packet_loss_perc_mult1000\",\"a_packet_loss_perc_mult1000\",\"b_packet_loss_perc_mult1000\",\"delay_sum\",\"a_delay_sum\",\n          \"b_delay_sum\",\"delay_avg_mult100\",\"a_delay_avg_mult100\",\"b_delay_avg_mult100\",\"delay_cnt\",\"a_delay_cnt\",\n          \"b_delay_cnt\",\"rtcp_avgfr_mult10\",\"rtcp_avgjitter_mult10\",\"lost\",\"id_sensor\",\"price_operator_mult100\",\n          \"price_operator_currency_id\",\"price_customer_mult100\",\"price_customer_currency_id\",\"reason_sip_cause\",\n          \"reason_sip_text_id\",\"reason_q850_cause\",\"reason_q850_text_id\",\"caller_silence\",\"called_silence\",\n          \"caller_silence_end\",\"called_silence_end\",\"a_mos_xr_min_mult10\",\"b_mos_xr_min_mult10\",\"a_mos_xr_mult10\",\n          \"b_mos_xr_mult10\",\"response_time_100\",\"response_time_xxx\",\"a_mos_f1_min_mult10\",\"a_mos_f2_min_mult10\",\n          \"a_mos_adapt_min_mult10\",\"b_mos_f1_min_mult10\",\"b_mos_f2_min_mult10\",\"b_mos_adapt_min_mult10\",\"a_rtp_ptime\",\n          \"b_rtp_ptime\",\"flags\"]\n\ndf = pd.read_csv(CSV_FILE_PATH,low_memory=False,names=HEADER)",
      "execution_count": 5,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "def evaluaterow(a1, a2, a3, a4, a5, a6, a7, a8, a9, a10, delay_sum, connect_duration):\n\n    calc_sum = get_sum(a1, a2, a3, a4, a5, a6, a7, a8, a9, a10)\n\n    final_sum = (delay_sum + calc_sum)/(connect_duration*1000) #+ calc_sum)\n\n    return final_sum\n\n\ndef get_sum(a1, a2, a3, a4, a5, a6, a7, a8, a9, a10):\n    top_sum = 0\n    n = 1\n    w1 = 1\n    w2 = 2\n    w3 = 3\n    w4 = 4\n    w5 = 5\n    w6 = 6\n    w7 = 7\n    w8 = 8\n    w9 = 9\n    w10 = 10\n\n    top_sum += 20 * n * (a1) * (w1 - 1)\n    n += 1\n    top_sum += 20 * n * (a2) * (w2 - 1)\n    n += 1\n    top_sum += 20 * n * (a3) * (w3 - 1)\n    n += 1\n    top_sum += 20 * n * (a4) * (w4 - 1)\n    n += 1\n    top_sum += 20 * n * (a5) * (w5 - 1)\n    n += 1\n    top_sum += 20 * n * (a6) * (w6 - 1)\n    n += 1\n    top_sum += 20 * n * (a7) * (w7 - 1)\n    n += 1\n    top_sum += 20 * n * (a8) * (w8 - 1)\n    n += 1\n    top_sum += 20 * n * (a9) * (w9 - 1)\n    n += 1\n    top_sum += 20 * n * (a10) * (w10 - 1)\n\n    return (top_sum)\n\n\ndf_qualfun_a = df.apply(lambda df2: evaluaterow(df2['a_sl1'],df2['a_sl2'],df2['a_sl3'],df2['a_sl4'],df2['a_sl5'],df2['a_sl6'],df2['a_sl7'],df2['a_sl8'],df2['a_sl9'],df2['a_sl10'],df2['a_delay_sum'],df2['connect_duration']), axis=1)\n\n\ndf_qualfun_b = df.apply(lambda df2: evaluaterow(df2['b_sl1'],df2['b_sl2'],df2['b_sl3'],df2['b_sl4'],df2['b_sl5'],df2['b_sl6'],df2['b_sl7'],df2['b_sl8'],df2['b_sl9'],df2['b_sl10'],df2['b_delay_sum'],df2['connect_duration']), axis=1)\n\n\n\n#df = df.assign(qualfun=df_qualfun)\n#df.compute()\n#dfqf = df_qualfun.compute()",
      "execution_count": 6,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "df['a_qualfun'] = df_qualfun_a\ndf['b_qualfun'] = df_qualfun_b",
      "execution_count": 8,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "filename='van_rtp_unique_ip'\n\nDATA_PATH = (filename+'.csv')\nDESCRIBED_COLUMNS = ['a_saddr','b_saddr','a_saddr_asn', 'a_saddr_asorg', 'a_saddr_lat', 'a_saddr_long','b_saddr_asn', 'b_saddr_asorg', 'b_saddr_lat', 'b_saddr_long','send_A']\n\n# read dataframe. calldate and callend are the only date columns\ndf_pairings_van = pd.read_csv(DATA_PATH)\ndf_pairings_van = df_pairings_van[DESCRIBED_COLUMNS]\n\ndf_pairings_van['count']=df_pairings_van.apply(getcount,axis=1)\n#df_pairings_van['mean_qualfun']=df_pairings_van.apply(getbqualfunmean,axis=1)\n#df_pairings_van['mean_mos']=df_pairings_van.apply(getmosmean,axis=1)\n\n#df_pairings_van.to_csv(filename+'_wextra.csv')",
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:6: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n  \n"
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "filename='tor_rtp_unique_ip'\n\nDATA_PATH = (filename+'.csv')\nDESCRIBED_COLUMNS = ['a_saddr','b_saddr','a_saddr_asn', 'a_saddr_asorg', 'a_saddr_lat', 'a_saddr_long','b_saddr_asn', 'b_saddr_asorg', 'b_saddr_lat', 'b_saddr_long','send_A']\n\n# read dataframe. calldate and callend are the only date columns\ndf_pairings_tor = pd.read_csv(DATA_PATH)\ndf_pairings_tor = df_pairings_tor[DESCRIBED_COLUMNS]\n\ndf_pairings_tor['count']=df_pairings_tor.apply(getcount,axis=1)\n#df_pairings_tor['mean_qualfun']=df_pairings_tor.apply(getbqualfunmean,axis=1)\n#df_pairings_tor['mean_mos']=df_pairings_tor.apply(getmosmean,axis=1)\n\n#df_pairings_tor.to_csv(filename+'wcount.csv')",
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:6: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n  \n"
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "filename='mtl_rtp_unique_ip'\n\nDATA_PATH = (filename+'.csv')\nDESCRIBED_COLUMNS = ['a_saddr','b_saddr','a_saddr_asn', 'a_saddr_asorg', 'a_saddr_lat', 'a_saddr_long','b_saddr_asn', 'b_saddr_asorg', 'b_saddr_lat', 'b_saddr_long','send_A']\n\n# read dataframe. calldate and callend are the only date columns\ndf_pairings_mtl = pd.read_csv(DATA_PATH)\ndf_pairings_mtl = df_pairings_mtl[DESCRIBED_COLUMNS]\n\ndf_pairings_mtl['count']=df_pairings_mtl.apply(getcount,axis=1)\n#df_pairings_mtl['mean_qualfun']=df_pairings_mtl.apply(getbqualfunmean,axis=1)\n#df_pairings_mtl['mean_mos']=df_pairings_mtl.apply(getmosmean,axis=1)\n\n#df_pairings_mtl.to_csv(filename+'wcount.csv')",
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:6: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n  \n"
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "def getcount(row):\n    a=int(row['a_saddr'])\n    b=int(row['b_saddr'])\n    \n    df_asaddr=df['a_saddr']\n    df_new=df[df['a_saddr']==a][df['b_saddr']==b]\n\n    return len(df_new)\n\ndef getbqualfunmean(row):\n    \n    a=int(row['a_saddr'])\n    b=int(row['b_saddr'])\n    \n    df_asaddr=df['a_saddr']\n    df_new=df[df['a_saddr']==a][df['b_saddr']==b]\n\n    if row['send_A']:\n        return df_new['a_qualfun'].mean()\n    else:\n        return df_new['b_qualfun'].mean()\n\ndef getmosmean(row):\n    a=int(row['a_saddr'])\n    b=int(row['b_saddr'])\n    \n    df_asaddr=df['a_saddr']\n    df_new=df[df['a_saddr']==a][df['b_saddr']==b]\n    \n    if row['send_A']:\n        return df_new['a_mos_adapt_mult10'].mean()\n    else:\n        return df_new['b_mos_adapt_mult10'].mean()\n    \ndef getbadratio(row):\n    total_count=float(row['count'])\n    bad_count=float(row['bad_count'])\n\n    return bad_count/total_count",
      "execution_count": 25,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "thresh=0.02\n\ndf_filtered_a=df[df['a_qualfun']>thresh]\ndf_filtered_b=df[df['b_qualfun']>thresh]\n\ndef getbadcount(row):\n    a=int(row['a_saddr'])\n    b=int(row['b_saddr'])\n    \n    if row['send_A']:\n        df_new=df_filtered_a[df_filtered_a['a_saddr']==a][df_filtered_a['b_saddr']==b]\n    else:\n        df_new=df_filtered_b[df_filtered_b['a_saddr']==a][df_filtered_b['b_saddr']==b]\n                \n    return len(df_new)",
      "execution_count": 55,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "# Create matrix\n\ndf_pairings_raw=df_pairings_van\n\ndf_pairings_raw['bad_count']=df_pairings_raw.apply(getbadcount,axis=1)\ndf_pairings_raw['bad_ratio']=df_pairings_raw.apply(getbadratio,axis=1)\n\ndf_pairings=df_pairings_raw[df_pairings_raw['a_saddr_asorg']=='CloudPBX']\n\ndf_asnpairings=df_pairings.groupby(['a_saddr_asn','b_saddr_asn'])[['count','bad_count']].sum()\n\naasn_array = []\nbasn_array = []\nparameter_array= []\n\nfor row in df_asnpairings.iterrows():\n    if float(row[1]['bad_count'])>30:\n        aasn_array.append(int(row[0][0]))\n        basn_array.append(int(row[0][1]))\n        parameter_array.append(float(row[1]['bad_count'])/float(row[1]['count']))\n    \n\n#asnlist=sorted(list(set(df['a_saddr_asn'].unique()).union(set(df['b_saddr_asn'].unique()))))\n\nno_entries=len(aasn_array)\nasnlist=sorted(list(set(aasn_array).union(set(basn_array))))\n\naasnlist=sorted(list(set(aasn_array)))\nbasnlist=sorted(list(set(basn_array)))\nno_aasn=len(aasnlist)\nno_basn=len(basnlist)\n\n\n#no_asn=len(asnlist)\n#print(no_aasn)\n#print(no_basn)\n\nasn_par=np.zeros([no_aasn,no_basn])\nfor i in range(0,no_entries):\n    row=aasnlist.index(aasn_array[i])\n    column=basnlist.index(basn_array[i])\n    asn_par[row,column]=parameter_array[i]\n    \n\n\naasorglist=list()\nbasorglist=list()\n\nfor asn in iter(aasnlist):\n    aasorglist.append(df_pairings[df_pairings['a_saddr_asn']==asn]['a_saddr_asorg'][0:1].values[0])\n\nfor asn in iter(basnlist):\n    basorglist.append(df_pairings[df_pairings['b_saddr_asn']==asn]['b_saddr_asorg'][0:1].values[0])",
      "execution_count": 65,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:13: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n  del sys.path[0]\n/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:11: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n  # This is added back by InteractiveShellApp.init_path()\n"
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "plt.rcParams['axes.labelsize'] = 10\n# Set up the matplotlib figure\nf, ax = plt.subplots(figsize=(20, 16))\n\nplt.title('ASN Pairings with High Bad Count to Good Count Ratio. A call is bad if Qualfun>0.02. ASN pairings with less than 30 bad calls are ignored.')\n\n# We want to show all ticks...\n#ax.set_xticks(np.arange(len(farmers)))\n#ax.set_yticks(np.arange(len(vegetables)))\n# ... and label them with the respective list entries\n#ax.set_xticklabels(basnlist)\n#ax.set_yticklabels(aasnlist)\n\n# Generate a custom diverging colormap\ncmap = sns.diverging_palette(10, 220, as_cmap=True)\n\n\n# Draw the heatmap with the mask and correct aspect ratio\nsns.heatmap(asn_par, cmap=cmap, center=0,\n            square=True, linewidths=.5, cbar_kws={\"shrink\": .5},yticklabels=aasorglist,xticklabels=basorglist);\nplt.tight_layout()\nplt.savefig('./img/asorg_badcallsratio_byQualfun_1millionstar_aCloudPBX_van.png', dpi=600)",
      "execution_count": 66,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<matplotlib.figure.Figure at 0x7fa7b7b6a588>",
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "# Create matrix\n\ndf_pairings_raw=df_pairings_van\n\ndf_pairings_raw['bad_count']=df_pairings_raw.apply(getbadcount,axis=1)\ndf_pairings_raw['bad_ratio']=df_pairings_raw.apply(getbadratio,axis=1)\n\ndf_pairings=df_pairings_raw[df_pairings_raw['b_saddr_asorg']=='CloudPBX']\n\ndf_asnpairings=df_pairings.groupby(['a_saddr_asn','b_saddr_asn'])[['count','bad_count']].sum()\n\naasn_array = []\nbasn_array = []\nparameter_array= []\n\nfor row in df_asnpairings.iterrows():\n    if float(row[1]['bad_count'])>30:\n        aasn_array.append(int(row[0][0]))\n        basn_array.append(int(row[0][1]))\n        parameter_array.append(float(row[1]['bad_count'])/float(row[1]['count']))\n    \n\n#asnlist=sorted(list(set(df['a_saddr_asn'].unique()).union(set(df['b_saddr_asn'].unique()))))\n\nno_entries=len(aasn_array)\nasnlist=sorted(list(set(aasn_array).union(set(basn_array))))\n\naasnlist=sorted(list(set(aasn_array)))\nbasnlist=sorted(list(set(basn_array)))\nno_aasn=len(aasnlist)\nno_basn=len(basnlist)\n\n\n#no_asn=len(asnlist)\n#print(no_aasn)\n#print(no_basn)\n\nasn_par=np.zeros([no_aasn,no_basn])\nfor i in range(0,no_entries):\n    row=aasnlist.index(aasn_array[i])\n    column=basnlist.index(basn_array[i])\n    asn_par[row,column]=parameter_array[i]\n    \nasn_par=asn_par.transpose()\n\naasorglist=list()\nbasorglist=list()\n\nfor asn in iter(aasnlist):\n    aasorglist.append(df_pairings[df_pairings['a_saddr_asn']==asn]['a_saddr_asorg'][0:1].values[0])\n\nfor asn in iter(basnlist):\n    basorglist.append(df_pairings[df_pairings['b_saddr_asn']==asn]['b_saddr_asorg'][0:1].values[0])",
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:13: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n  del sys.path[0]\n/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:11: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n  # This is added back by InteractiveShellApp.init_path()\n"
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "plt.rcParams['axes.labelsize'] = 10\n# Set up the matplotlib figure\nf, ax = plt.subplots(figsize=(20, 16))\n\nplt.title('ASN Pairings with High Bad Count to Good Count Ratio. A call is bad if Qualfun>0.02. ASN pairings with less than 30 bad calls are ignored.')\n\n# We want to show all ticks...\n#ax.set_xticks(np.arange(len(farmers)))\n#ax.set_yticks(np.arange(len(vegetables)))\n# ... and label them with the respective list entries\n#ax.set_xticklabels(basnlist)\n#ax.set_yticklabels(aasnlist)\n\n# Generate a custom diverging colormap\ncmap = sns.diverging_palette(10, 220, as_cmap=True)\n\n\n# Draw the heatmap with the mask and correct aspect ratio\nsns.heatmap(asn_par, cmap=cmap, center=0,\n            square=True, linewidths=.5, cbar_kws={\"shrink\": .5},yticklabels=basorglist,xticklabels=aasorglist);\nplt.tight_layout()\nplt.savefig('./img/asorg_badcallsratio_byQualfun_1millionstar_bCloudPBX_van.png', dpi=600)",
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<matplotlib.figure.Figure at 0x7fa7b7a37e48>",
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "df_pairings_comb=df_pairings_van.merge(df_pairings_mtl,how='outer')\ndf_pairings_comb=df_pairings_comb.merge(df_pairings_tor,how='outer')",
      "execution_count": 73,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "# Create matrix\n\ndf_pairings_raw=df_pairings_comb\n\ndf_pairings_raw['bad_count']=df_pairings_raw.apply(getbadcount,axis=1)\ndf_pairings_raw['bad_ratio']=df_pairings_raw.apply(getbadratio,axis=1)\n\ndf_pairings=df_pairings_raw[df_pairings_raw['a_saddr_asorg']=='CloudPBX']\n\ndf_asnpairings=df_pairings.groupby(['a_saddr_asn','b_saddr_asn'])[['count','bad_count']].sum()\n\naasn_array = []\nbasn_array = []\nparameter_array= []\n\nfor row in df_asnpairings.iterrows():\n    if float(row[1]['bad_count'])>30:\n        aasn_array.append(int(row[0][0]))\n        basn_array.append(int(row[0][1]))\n        parameter_array.append(float(row[1]['bad_count'])/float(row[1]['count']))\n    \n\n#asnlist=sorted(list(set(df['a_saddr_asn'].unique()).union(set(df['b_saddr_asn'].unique()))))\n\nno_entries=len(aasn_array)\nasnlist=sorted(list(set(aasn_array).union(set(basn_array))))\n\naasnlist=sorted(list(set(aasn_array)))\nbasnlist=sorted(list(set(basn_array)))\nno_aasn=len(aasnlist)\nno_basn=len(basnlist)\n\n\n#no_asn=len(asnlist)\n#print(no_aasn)\n#print(no_basn)\n\nasn_par=np.zeros([no_aasn,no_basn])\nfor i in range(0,no_entries):\n    row=aasnlist.index(aasn_array[i])\n    column=basnlist.index(basn_array[i])\n    asn_par[row,column]=parameter_array[i]\n    \n\n\naasorglist=list()\nbasorglist=list()\n\nfor asn in iter(aasnlist):\n    aasorglist.append(df_pairings[df_pairings['a_saddr_asn']==asn]['a_saddr_asorg'][0:1].values[0])\n\nfor asn in iter(basnlist):\n    basorglist.append(df_pairings[df_pairings['b_saddr_asn']==asn]['b_saddr_asorg'][0:1].values[0])",
      "execution_count": 95,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:13: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n  del sys.path[0]\n/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:11: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n  # This is added back by InteractiveShellApp.init_path()\n"
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "plt.rcParams['axes.labelsize'] = 10\n# Set up the matplotlib figure\nf, ax = plt.subplots(figsize=(20, 16))\n\nplt.title('ASN Pairings with High Bad Count to Good Count Ratio. A call is bad if Qualfun>0.02. ASN pairings with less than 30 bad calls are ignored.')\n\n# We want to show all ticks...\n#ax.set_xticks(np.arange(len(farmers)))\n#ax.set_yticks(np.arange(len(vegetables)))\n# ... and label them with the respective list entries\n#ax.set_xticklabels(basnlist)\n#ax.set_yticklabels(aasnlist)\n\n# Generate a custom diverging colormap\ncmap = sns.diverging_palette(10, 220, as_cmap=True)\n\n\n# Draw the heatmap with the mask and correct aspect ratio\nsns.heatmap(asn_par, cmap=cmap, center=0,\n            square=True, linewidths=.5, cbar_kws={\"shrink\": .5},yticklabels=cloudpbxlist,xticklabels=basorglist);\nplt.tight_layout()\nplt.savefig('./img/asorg_badcallsratio_byQualfun_1millionstar_aCloudPBX_all.png', dpi=600)",
      "execution_count": 96,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<matplotlib.figure.Figure at 0x7fa7b7d6e400>",
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "",
      "execution_count": null,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "# Create matrix\n\ndf_pairings_raw=df_pairings_comb\n\ndf_pairings_raw['bad_count']=df_pairings_raw.apply(getbadcount,axis=1)\ndf_pairings_raw['bad_ratio']=df_pairings_raw.apply(getbadratio,axis=1)\n\ndf_pairings=df_pairings_raw[df_pairings_raw['b_saddr_asorg']=='CloudPBX']\n\ndf_asnpairings=df_pairings.groupby(['a_saddr_asn','b_saddr_asn'])[['count','bad_count']].sum()\n\naasn_array = []\nbasn_array = []\nparameter_array= []\n\nfor row in df_asnpairings.iterrows():\n    if float(row[1]['bad_count'])>30:\n        aasn_array.append(int(row[0][0]))\n        basn_array.append(int(row[0][1]))\n        parameter_array.append(float(row[1]['bad_count'])/float(row[1]['count']))\n    \n\n#asnlist=sorted(list(set(df['a_saddr_asn'].unique()).union(set(df['b_saddr_asn'].unique()))))\n\nno_entries=len(aasn_array)\nasnlist=sorted(list(set(aasn_array).union(set(basn_array))))\n\naasnlist=sorted(list(set(aasn_array)))\nbasnlist=sorted(list(set(basn_array)))\nno_aasn=len(aasnlist)\nno_basn=len(basnlist)\n\n\n#no_asn=len(asnlist)\n#print(no_aasn)\n#print(no_basn)\n\nasn_par=np.zeros([no_aasn,no_basn])\nfor i in range(0,no_entries):\n    row=aasnlist.index(aasn_array[i])\n    column=basnlist.index(basn_array[i])\n    asn_par[row,column]=parameter_array[i]\n    \nasn_par=asn_par.transpose()\n\naasorglist=list()\nbasorglist=list()\n\nfor asn in iter(aasnlist):\n    aasorglist.append(df_pairings[df_pairings['a_saddr_asn']==asn]['a_saddr_asorg'][0:1].values[0])\n\nfor asn in iter(basnlist):\n    basorglist.append(df_pairings[df_pairings['b_saddr_asn']==asn]['b_saddr_asorg'][0:1].values[0])",
      "execution_count": 93,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:13: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n  del sys.path[0]\n/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:11: UserWarning: Boolean Series key will be reindexed to match DataFrame index.\n  # This is added back by InteractiveShellApp.init_path()\n"
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "plt.rcParams['axes.labelsize'] = 10\n# Set up the matplotlib figure\nf, ax = plt.subplots(figsize=(20, 16))\n\nplt.title('ASN Pairings with High Bad Count to Good Count Ratio. A call is bad if Qualfun>0.02. ASN pairings with less than 30 bad calls are ignored.')\n\n# We want to show all ticks...\n#ax.set_xticks(np.arange(len(farmers)))\n#ax.set_yticks(np.arange(len(vegetables)))\n# ... and label them with the respective list entries\n#ax.set_xticklabels(basnlist)\n#ax.set_yticklabels(aasnlist)\n\n# Generate a custom diverging colormap\ncmap = sns.diverging_palette(10, 220, as_cmap=True)\n\n\n# Draw the heatmap with the mask and correct aspect ratio\nsns.heatmap(asn_par, cmap=cmap, center=0,\n            square=True, linewidths=.5, cbar_kws={\"shrink\": .5},yticklabels=cloudpbxlist,xticklabels=aasorglist);\nplt.tight_layout()\nplt.savefig('./img/asorg_badcallsratio_byQualfun_1millionstar_bCloudPBX_all.png', dpi=600)",
      "execution_count": 94,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": "<matplotlib.figure.Figure at 0x7fa7b7d8d0f0>",
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "basnlist",
      "execution_count": 89,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": "[393755, 395152, 395766]"
          },
          "execution_count": 89,
          "metadata": {}
        }
      ]
    },
    {
      "metadata": {
        "trusted": false
      },
      "cell_type": "code",
      "source": "aasnlist=[393755, 395152, 395766]\n\ncloudpbxlist=['CloudPBX Toronto','CloudPBX Vancouver','CloudPBX Montreal']",
      "execution_count": 90,
      "outputs": []
    }
  ],
  "metadata": {
    "toc": {
      "nav_menu": {},
      "number_sections": true,
      "sideBar": true,
      "skip_h1_title": false,
      "base_numbering": 1,
      "title_cell": "Table of Contents",
      "title_sidebar": "Contents",
      "toc_cell": false,
      "toc_position": {},
      "toc_section_display": true,
      "toc_window_display": false
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python [default]",
      "language": "python"
    },
    "language_info": {
      "file_extension": ".py",
      "name": "python",
      "nbconvert_exporter": "python",
      "version": "3.5.5",
      "pygments_lexer": "ipython3",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      }
    },
    "gist": {
      "id": "",
      "data": {
        "description": "asn_pairings.ipynb",
        "public": false
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 2
}

Geo Worksheet.ipynb

histograms_mos_qualfun.ipynb