{ "cells": [ { "cell_type": "markdown", "id": "57b58c35", "metadata": {}, "source": [ "## Quality assurance of prediction results\n", "\n", "In the previous tutorial, we looked at various ways to visualise the results of our model. \n", "\n", "These are useful for quality assurance purposes because they allow us to understand how our model works and verify that it is doing something sensible. They can also be useful to identify examples where the model is not performing as expected. \n", "\n", "In addition to these spot checks, Splink also has functions to perform more formal accuracy analysis. These functions allow you to understand the likely prevalence of false positives and false negatives in your linkage models.\n", "\n", "They rely on the existence of a sample of labelled (ground truth) matches, which may have been produced (for example) by human beings. For the accuracy analysis to be unbiased, the sample should be representative of the overall dataset." ] }, { "cell_type": "code", "execution_count": 1, "id": "fb29d421", "metadata": {}, "outputs": [], "source": [ "# Rerun our predictions to we're ready to view the charts\n", "from splink.duckdb.duckdb_linker import DuckDBLinker\n", "import pandas as pd \n", "import altair as alt\n", "alt.renderers.enable('mimetype')\n", "\n", "df = pd.read_csv(\"./data/fake_1000.csv\")\n", "linker = DuckDBLinker(df)\n", "linker.load_settings_from_json(\"./demo_settings/saved_model_from_demo.json\")\n", "df_predictions = linker.predict(threshold_match_probability=0.2)" ] }, { "cell_type": "markdown", "id": "7b0dedd9", "metadata": {}, "source": [ "## Load in labels\n", "\n", "The labels file contains a list of pairwise comparisons which represent matches and non-matches.\n", "\n", "The required format of the labels file is described [here](https://moj-analytical-services.github.io/splink/linkerqa.html#splink.linker.Linker.roc_chart_from_labels)." ] }, { "cell_type": "code", "execution_count": 2, "id": "bbfdc70c", "metadata": {}, "outputs": [], "source": [ "df_labels = pd.read_csv(\"./data/fake_1000_labels.csv\")\n", "df_labels.head(5)\n", "linker._con.register(\"labels\", df_labels)\n", "linker._initialise_df_concat_with_tf()" ] }, { "cell_type": "markdown", "id": "81e4396d", "metadata": {}, "source": [ "## Receiver operating characteristic curve\n", "\n", "A [ROC chart](https://en.wikipedia.org/wiki/Receiver_operating_characteristic) shows how the number of false positives and false negatives varies depending on the match threshold chosen. The match threshold is the match weight chosen as a cutoff for which pairwise comparisons to accept as matches.\n", "\n" ] }, { "cell_type": "code", "execution_count": 3, "id": "01dd7eec", "metadata": {}, "outputs": [ { "data": { "application/vnd.vegalite.v4+json": { "$schema": "https://vega.github.io/schema/vega-lite/v4.8.1.json", "data": { "values": [ { "FN": 0, "FN_rate": 0, "FP": 1145, "FP_rate": 1, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 0, "TN_rate": 0, "TP": 80, "TP_rate": 1, "precision": 0.06530611962080002, "recall": 1, "row_count": 1225, "truth_threshold": -16.60994994256404 }, { "FN": 0, "FN_rate": 0, "FP": 1047, "FP_rate": 0.9144104719161987, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 98, "TN_rate": 0.08558952063322067, "TP": 80, "TP_rate": 1, "precision": 0.07098491489887238, "recall": 1, "row_count": 1225, "truth_threshold": -15.488603946152788 }, { "FN": 0, "FN_rate": 0, "FP": 877, "FP_rate": 0.7659388780593872, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 268, "TN_rate": 0.234061136841774, "TP": 80, "TP_rate": 1, "precision": 0.08359456807374954, "recall": 1, "row_count": 1225, "truth_threshold": -14.677083359952446 }, { "FN": 0, "FN_rate": 0, "FP": 835, "FP_rate": 0.7292576432228088, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 310, "TN_rate": 0.27074235677719116, "TP": 80, "TP_rate": 1, "precision": 0.08743169158697128, "recall": 1, "row_count": 1225, "truth_threshold": -14.535475051974686 }, { "FN": 0, "FN_rate": 0, "FP": 749, "FP_rate": 0.6541484594345093, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 396, "TN_rate": 0.3458515405654907, "TP": 80, "TP_rate": 1, "precision": 0.09650181233882904, "recall": 1, "row_count": 1225, "truth_threshold": -14.078050765385573 }, { "FN": 0, "FN_rate": 0, "FP": 733, "FP_rate": 0.6401746869087219, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 412, "TN_rate": 0.3598253130912781, "TP": 80, "TP_rate": 1, "precision": 0.09840098768472672, "recall": 1, "row_count": 1225, "truth_threshold": -13.764688729777017 }, { "FN": 0, "FN_rate": 0, "FP": 656, "FP_rate": 0.5729257464408875, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 489, "TN_rate": 0.42707422375679016, "TP": 80, "TP_rate": 1, "precision": 0.10869564861059189, "recall": 1, "row_count": 1225, "truth_threshold": -13.555737363541194 }, { "FN": 0, "FN_rate": 0, "FP": 549, "FP_rate": 0.4794759750366211, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 596, "TN_rate": 0.5205240249633789, "TP": 80, "TP_rate": 1, "precision": 0.12718601524829865, "recall": 1, "row_count": 1225, "truth_threshold": -13.414129055563434 }, { "FN": 0, "FN_rate": 0, "FP": 509, "FP_rate": 0.44454148411750793, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 636, "TN_rate": 0.5554584860801697, "TP": 80, "TP_rate": 1, "precision": 0.13582342863082886, "recall": 1, "row_count": 1225, "truth_threshold": -12.643342733365765 }, { "FN": 1, "FN_rate": 0.012500000186264515, "FP": 419, "FP_rate": 0.36593887209892273, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 726, "TN_rate": 0.6340611577033997, "TP": 79, "TP_rate": 0.987500011920929, "precision": 0.15863454341888428, "recall": 0.987500011920929, "row_count": 1225, "truth_threshold": -12.602608469363092 }, { "FN": 1, "FN_rate": 0.012500000186264515, "FP": 384, "FP_rate": 0.3353711664676666, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 761, "TN_rate": 0.664628803730011, "TP": 79, "TP_rate": 0.987500011920929, "precision": 0.17062635719776154, "recall": 0.987500011920929, "row_count": 1225, "truth_threshold": -12.14518418277398 }, { "FN": 1, "FN_rate": 0.012500000186264515, "FP": 378, "FP_rate": 0.33013099431991577, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 767, "TN_rate": 0.6698690056800842, "TP": 79, "TP_rate": 0.987500011920929, "precision": 0.17286652326583862, "recall": 0.987500011920929, "row_count": 1225, "truth_threshold": -11.831822147165424 }, { "FN": 1, "FN_rate": 0.012500000186264515, "FP": 365, "FP_rate": 0.31877729296684265, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 780, "TN_rate": 0.6812227368354797, "TP": 79, "TP_rate": 0.987500011920929, "precision": 0.1779279261827469, "recall": 0.987500011920929, "row_count": 1225, "truth_threshold": -11.690213839187663 }, { "FN": 2, "FN_rate": 0.02500000037252903, "FP": 273, "FP_rate": 0.23842795193195343, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 872, "TN_rate": 0.7615720629692078, "TP": 78, "TP_rate": 0.9750000238418579, "precision": 0.2222222238779068, "recall": 0.9750000238418579, "row_count": 1225, "truth_threshold": -11.48126247295184 }, { "FN": 2, "FN_rate": 0.02500000037252903, "FP": 250, "FP_rate": 0.21834060549736023, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 895, "TN_rate": 0.7816593647003174, "TP": 78, "TP_rate": 0.9750000238418579, "precision": 0.23780487477779388, "recall": 0.9750000238418579, "row_count": 1225, "truth_threshold": -11.23278955259855 }, { "FN": 2, "FN_rate": 0.02500000037252903, "FP": 243, "FP_rate": 0.21222707629203796, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 902, "TN_rate": 0.7877729535102844, "TP": 78, "TP_rate": 0.9750000238418579, "precision": 0.2429906576871872, "recall": 0.9750000238418579, "row_count": 1225, "truth_threshold": -10.906098261537272 }, { "FN": 2, "FN_rate": 0.02500000037252903, "FP": 238, "FP_rate": 0.20786026120185852, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 907, "TN_rate": 0.7921397089958191, "TP": 78, "TP_rate": 0.9750000238418579, "precision": 0.24683544039726257, "recall": 0.9750000238418579, "row_count": 1225, "truth_threshold": -10.710476150754172 }, { "FN": 2, "FN_rate": 0.02500000037252903, "FP": 215, "FP_rate": 0.1877729296684265, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 930, "TN_rate": 0.8122270703315735, "TP": 78, "TP_rate": 0.9750000238418579, "precision": 0.266211599111557, "recall": 0.9750000238418579, "row_count": 1225, "truth_threshold": -10.56886784277641 }, { "FN": 3, "FN_rate": 0.03750000149011612, "FP": 98, "FP_rate": 0.08558952063322067, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1047, "TN_rate": 0.9144104719161987, "TP": 77, "TP_rate": 0.9624999761581421, "precision": 0.4399999976158142, "recall": 0.9624999761581421, "row_count": 1225, "truth_threshold": -9.793937423031654 }, { "FN": 3, "FN_rate": 0.03750000149011612, "FP": 95, "FP_rate": 0.08296943455934525, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1050, "TN_rate": 0.9170305728912354, "TP": 77, "TP_rate": 0.9624999761581421, "precision": 0.447674423456192, "recall": 0.9624999761581421, "row_count": 1225, "truth_threshold": -9.78475226512602 }, { "FN": 3, "FN_rate": 0.03750000149011612, "FP": 91, "FP_rate": 0.07947598397731781, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1054, "TN_rate": 0.920524001121521, "TP": 77, "TP_rate": 0.9624999761581421, "precision": 0.4583333432674408, "recall": 0.9624999761581421, "row_count": 1225, "truth_threshold": -9.75734725657607 }, { "FN": 3, "FN_rate": 0.03750000149011612, "FP": 75, "FP_rate": 0.06550218164920807, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1070, "TN_rate": 0.9344978332519531, "TP": 77, "TP_rate": 0.9624999761581421, "precision": 0.5065789222717285, "recall": 0.9624999761581421, "row_count": 1225, "truth_threshold": -9.299922969986957 }, { "FN": 3, "FN_rate": 0.03750000149011612, "FP": 74, "FP_rate": 0.06462882459163666, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1071, "TN_rate": 0.9353711605072021, "TP": 77, "TP_rate": 0.9624999761581421, "precision": 0.5099337697029114, "recall": 0.9624999761581421, "row_count": 1225, "truth_threshold": -9.158314662009195 }, { "FN": 5, "FN_rate": 0.0625, "FP": 69, "FP_rate": 0.060262009501457214, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1076, "TN_rate": 0.9397379755973816, "TP": 75, "TP_rate": 0.9375, "precision": 0.5208333134651184, "recall": 0.9375, "row_count": 1225, "truth_threshold": -9.057240845364372 }, { "FN": 5, "FN_rate": 0.0625, "FP": 68, "FP_rate": 0.059388644993305206, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1077, "TN_rate": 0.9406113624572754, "TP": 75, "TP_rate": 0.9375, "precision": 0.5244755148887634, "recall": 0.9375, "row_count": 1225, "truth_threshold": -8.831623370947918 }, { "FN": 5, "FN_rate": 0.0625, "FP": 66, "FP_rate": 0.05764191970229149, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1079, "TN_rate": 0.9423580765724182, "TP": 75, "TP_rate": 0.9375, "precision": 0.5319148898124695, "recall": 0.9375, "row_count": 1225, "truth_threshold": -8.672591426620402 }, { "FN": 5, "FN_rate": 0.0625, "FP": 61, "FP_rate": 0.053275108337402344, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1084, "TN_rate": 0.9467248916625977, "TP": 75, "TP_rate": 0.9375, "precision": 0.5514705777168274, "recall": 0.9375, "row_count": 1225, "truth_threshold": -8.636001260164818 }, { "FN": 5, "FN_rate": 0.0625, "FP": 39, "FP_rate": 0.03406113386154175, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1106, "TN_rate": 0.9659388661384583, "TP": 75, "TP_rate": 0.9375, "precision": 0.6578947305679321, "recall": 0.9375, "row_count": 1225, "truth_threshold": -8.06083704875025 }, { "FN": 5, "FN_rate": 0.0625, "FP": 35, "FP_rate": 0.030567685142159462, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1110, "TN_rate": 0.9694322943687439, "TP": 75, "TP_rate": 0.9375, "precision": 0.6818181872367859, "recall": 0.9375, "row_count": 1225, "truth_threshold": -7.8518856825144265 }, { "FN": 5, "FN_rate": 0.0625, "FP": 33, "FP_rate": 0.028820959851145744, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1112, "TN_rate": 0.9711790680885315, "TP": 75, "TP_rate": 0.9375, "precision": 0.6944444179534912, "recall": 0.9375, "row_count": 1225, "truth_threshold": -7.738872648601806 }, { "FN": 5, "FN_rate": 0.0625, "FP": 30, "FP_rate": 0.026200873777270317, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1115, "TN_rate": 0.9737991094589233, "TP": 75, "TP_rate": 0.9375, "precision": 0.7142857313156128, "recall": 0.9375, "row_count": 1225, "truth_threshold": -7.710277374536665 }, { "FN": 5, "FN_rate": 0.0625, "FP": 29, "FP_rate": 0.025327511131763458, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1116, "TN_rate": 0.9746724963188171, "TP": 75, "TP_rate": 0.9375, "precision": 0.7211538553237915, "recall": 0.9375, "row_count": 1225, "truth_threshold": -7.262038245853187 }, { "FN": 5, "FN_rate": 0.0625, "FP": 27, "FP_rate": 0.02358078584074974, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1118, "TN_rate": 0.97641921043396, "TP": 75, "TP_rate": 0.9375, "precision": 0.7352941036224365, "recall": 0.9375, "row_count": 1225, "truth_threshold": -7.225448079397602 }, { "FN": 5, "FN_rate": 0.0625, "FP": 26, "FP_rate": 0.022707423195242882, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1119, "TN_rate": 0.9772925972938538, "TP": 75, "TP_rate": 0.9375, "precision": 0.7425742745399475, "recall": 0.9375, "row_count": 1225, "truth_threshold": -6.982765954775019 }, { "FN": 5, "FN_rate": 0.0625, "FP": 25, "FP_rate": 0.021834060549736023, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1120, "TN_rate": 0.9781659245491028, "TP": 75, "TP_rate": 0.9375, "precision": 0.75, "recall": 0.9375, "row_count": 1225, "truth_threshold": -6.9394910523389965 }, { "FN": 5, "FN_rate": 0.0625, "FP": 21, "FP_rate": 0.018340611830353737, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1124, "TN_rate": 0.9816594123840332, "TP": 75, "TP_rate": 0.9375, "precision": 0.78125, "recall": 0.9375, "row_count": 1225, "truth_threshold": -6.898756788336325 }, { "FN": 5, "FN_rate": 0.0625, "FP": 18, "FP_rate": 0.01572052389383316, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1127, "TN_rate": 0.984279453754425, "TP": 75, "TP_rate": 0.9375, "precision": 0.8064516186714172, "recall": 0.9375, "row_count": 1225, "truth_threshold": -5.8273302138333785 }, { "FN": 5, "FN_rate": 0.0625, "FP": 12, "FP_rate": 0.010480348952114582, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1133, "TN_rate": 0.9895196557044983, "TP": 75, "TP_rate": 0.9375, "precision": 0.8620689511299133, "recall": 0.9375, "row_count": 1225, "truth_threshold": -5.777410791925072 }, { "FN": 5, "FN_rate": 0.0625, "FP": 9, "FP_rate": 0.00786026194691658, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1136, "TN_rate": 0.9921397566795349, "TP": 75, "TP_rate": 0.9375, "precision": 0.8928571343421936, "recall": 0.9375, "row_count": 1225, "truth_threshold": -5.2485131299666525 }, { "FN": 6, "FN_rate": 0.07500000298023224, "FP": 9, "FP_rate": 0.00786026194691658, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1136, "TN_rate": 0.9921397566795349, "TP": 74, "TP_rate": 0.925000011920929, "precision": 0.891566276550293, "recall": 0.925000011920929, "row_count": 1225, "truth_threshold": -5.006624469727403 }, { "FN": 7, "FN_rate": 0.08749999850988388, "FP": 7, "FP_rate": 0.00611353712156415, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1138, "TN_rate": 0.9938864707946777, "TP": 73, "TP_rate": 0.9125000238418579, "precision": 0.9125000238418579, "recall": 0.9125000238418579, "row_count": 1225, "truth_threshold": -4.893611435814783 }, { "FN": 7, "FN_rate": 0.08749999850988388, "FP": 6, "FP_rate": 0.005240174476057291, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1139, "TN_rate": 0.9947597980499268, "TP": 73, "TP_rate": 0.9125000238418579, "precision": 0.9240506291389465, "recall": 0.9125000238418579, "row_count": 1225, "truth_threshold": -4.865016161749642 }, { "FN": 7, "FN_rate": 0.08749999850988388, "FP": 4, "FP_rate": 0.0034934498835355043, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1141, "TN_rate": 0.9965065717697144, "TP": 73, "TP_rate": 0.9125000238418579, "precision": 0.948051929473877, "recall": 0.9125000238418579, "row_count": 1225, "truth_threshold": -4.646615423705956 }, { "FN": 8, "FN_rate": 0.10000000149011612, "FP": 4, "FP_rate": 0.0034934498835355043, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1141, "TN_rate": 0.9965065717697144, "TP": 72, "TP_rate": 0.8999999761581421, "precision": 0.9473684430122375, "recall": 0.8999999761581421, "row_count": 1225, "truth_threshold": -4.137504741987995 }, { "FN": 8, "FN_rate": 0.10000000149011612, "FP": 3, "FP_rate": 0.0026200872380286455, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1142, "TN_rate": 0.9973798990249634, "TP": 72, "TP_rate": 0.8999999761581421, "precision": 0.9599999785423279, "recall": 0.8999999761581421, "row_count": 1225, "truth_threshold": -3.3386640862492003 }, { "FN": 9, "FN_rate": 0.11249999701976776, "FP": 3, "FP_rate": 0.0026200872380286455, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1142, "TN_rate": 0.9973798990249634, "TP": 71, "TP_rate": 0.887499988079071, "precision": 0.9594594836235046, "recall": 0.887499988079071, "row_count": 1225, "truth_threshold": -2.932149579138049 }, { "FN": 9, "FN_rate": 0.11249999701976776, "FP": 1, "FP_rate": 0.0008733624708838761, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1144, "TN_rate": 0.9991266131401062, "TP": 71, "TP_rate": 0.887499988079071, "precision": 0.9861111044883728, "recall": 0.887499988079071, "row_count": 1225, "truth_threshold": -2.8311489013971385 }, { "FN": 10, "FN_rate": 0.125, "FP": 1, "FP_rate": 0.0008733624708838761, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1144, "TN_rate": 0.9991266131401062, "TP": 70, "TP_rate": 0.875, "precision": 0.98591548204422, "recall": 0.875, "row_count": 1225, "truth_threshold": -2.1572191593932932 }, { "FN": 10, "FN_rate": 0.125, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 70, "TP_rate": 0.875, "precision": 1, "recall": 0.875, "row_count": 1225, "truth_threshold": -0.7308664264624687 }, { "FN": 11, "FN_rate": 0.13750000298023224, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 69, "TP_rate": 0.862500011920929, "precision": 1, "recall": 0.862500011920929, "row_count": 1225, "truth_threshold": -0.36988609476305273 }, { "FN": 12, "FN_rate": 0.15000000596046448, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 68, "TP_rate": 0.8500000238418579, "precision": 1, "recall": 0.8500000238418579, "row_count": 1225, "truth_threshold": 1.1433212648855802 }, { "FN": 13, "FN_rate": 0.16249999403953552, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 67, "TP_rate": 0.8374999761581421, "precision": 1, "recall": 0.8374999761581421, "row_count": 1225, "truth_threshold": 1.8093880498049824 }, { "FN": 14, "FN_rate": 0.17499999701976776, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 66, "TP_rate": 0.824999988079071, "precision": 1, "recall": 0.824999988079071, "row_count": 1225, "truth_threshold": 2.162013082415415 }, { "FN": 15, "FN_rate": 0.1875, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 65, "TP_rate": 0.8125, "precision": 1, "recall": 0.8125, "row_count": 1225, "truth_threshold": 2.2811784283388743 }, { "FN": 16, "FN_rate": 0.20000000298023224, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 64, "TP_rate": 0.800000011920929, "precision": 1, "recall": 0.800000011920929, "row_count": 1225, "truth_threshold": 2.467789438088027 }, { "FN": 17, "FN_rate": 0.21250000596046448, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 63, "TP_rate": 0.7875000238418579, "precision": 1, "recall": 0.7875000238418579, "row_count": 1225, "truth_threshold": 3.5624320737970847 }, { "FN": 19, "FN_rate": 0.23749999701976776, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 61, "TP_rate": 0.762499988079071, "precision": 1, "recall": 0.762499988079071, "row_count": 1225, "truth_threshold": 4.2540204580463135 }, { "FN": 20, "FN_rate": 0.25, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 60, "TP_rate": 0.75, "precision": 1, "recall": 0.75, "row_count": 1225, "truth_threshold": 4.6580632234106085 }, { "FN": 24, "FN_rate": 0.30000001192092896, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 56, "TP_rate": 0.699999988079071, "precision": 1, "recall": 0.699999988079071, "row_count": 1225, "truth_threshold": 5.132707101344441 }, { "FN": 26, "FN_rate": 0.32499998807907104, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 54, "TP_rate": 0.675000011920929, "precision": 1, "recall": 0.675000011920929, "row_count": 1225, "truth_threshold": 5.473632864149248 }, { "FN": 27, "FN_rate": 0.3375000059604645, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 53, "TP_rate": 0.6625000238418579, "precision": 1, "recall": 0.6625000238418579, "row_count": 1225, "truth_threshold": 5.685026698413226 }, { "FN": 28, "FN_rate": 0.3499999940395355, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 52, "TP_rate": 0.6499999761581421, "precision": 1, "recall": 0.6499999761581421, "row_count": 1225, "truth_threshold": 6.328495348635668 }, { "FN": 30, "FN_rate": 0.375, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 50, "TP_rate": 0.625, "precision": 1, "recall": 0.625, "row_count": 1225, "truth_threshold": 6.516556921409461 }, { "FN": 32, "FN_rate": 0.4000000059604645, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 48, "TP_rate": 0.6000000238418579, "precision": 1, "recall": 0.6000000238418579, "row_count": 1225, "truth_threshold": 6.9481736236532585 }, { "FN": 35, "FN_rate": 0.4375, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 45, "TP_rate": 0.5625, "precision": 1, "recall": 0.5625, "row_count": 1225, "truth_threshold": 7.449841345046921 }, { "FN": 36, "FN_rate": 0.44999998807907104, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 44, "TP_rate": 0.550000011920929, "precision": 1, "recall": 0.550000011920929, "row_count": 1225, "truth_threshold": 7.637902917820714 }, { "FN": 38, "FN_rate": 0.4749999940395355, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 42, "TP_rate": 0.5249999761581421, "precision": 1, "recall": 0.5249999761581421, "row_count": 1225, "truth_threshold": 7.712926222461523 }, { "FN": 39, "FN_rate": 0.48750001192092896, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 41, "TP_rate": 0.512499988079071, "precision": 1, "recall": 0.512499988079071, "row_count": 1225, "truth_threshold": 8.859769247712702 }, { "FN": 42, "FN_rate": 0.5249999761581421, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 38, "TP_rate": 0.4749999940395355, "precision": 1, "recall": 0.4749999940395355, "row_count": 1225, "truth_threshold": 9.14004857454539 }, { "FN": 43, "FN_rate": 0.5375000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 37, "TP_rate": 0.4625000059604645, "precision": 1, "recall": 0.4625000059604645, "row_count": 1225, "truth_threshold": 9.676008031737986 }, { "FN": 44, "FN_rate": 0.550000011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 36, "TP_rate": 0.44999998807907104, "precision": 1, "recall": 0.44999998807907104, "row_count": 1225, "truth_threshold": 10.089978878949308 }, { "FN": 45, "FN_rate": 0.5625, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 35, "TP_rate": 0.4375, "precision": 1, "recall": 0.4375, "row_count": 1225, "truth_threshold": 11.258293846460244 }, { "FN": 46, "FN_rate": 0.574999988079071, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 34, "TP_rate": 0.42500001192092896, "precision": 1, "recall": 0.42500001192092896, "row_count": 1225, "truth_threshold": 11.317074190024028 }, { "FN": 47, "FN_rate": 0.5874999761581421, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 33, "TP_rate": 0.4124999940395355, "precision": 1, "recall": 0.4124999940395355, "row_count": 1225, "truth_threshold": 12.297008411651136 }, { "FN": 48, "FN_rate": 0.6000000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 32, "TP_rate": 0.4000000059604645, "precision": 1, "recall": 0.4000000059604645, "row_count": 1225, "truth_threshold": 12.58803772594323 }, { "FN": 49, "FN_rate": 0.612500011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 31, "TP_rate": 0.38749998807907104, "precision": 1, "recall": 0.38749998807907104, "row_count": 1225, "truth_threshold": 12.605377590133758 }, { "FN": 50, "FN_rate": 0.625, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 30, "TP_rate": 0.375, "precision": 1, "recall": 0.375, "row_count": 1225, "truth_threshold": 12.793439162907552 }, { "FN": 53, "FN_rate": 0.6625000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 27, "TP_rate": 0.3375000059604645, "precision": 1, "recall": 0.3375000059604645, "row_count": 1225, "truth_threshold": 12.93524009173633 }, { "FN": 54, "FN_rate": 0.675000011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 26, "TP_rate": 0.32499998807907104, "precision": 1, "recall": 0.32499998807907104, "row_count": 1225, "truth_threshold": 13.125097752008548 }, { "FN": 56, "FN_rate": 0.699999988079071, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 24, "TP_rate": 0.30000001192092896, "precision": 1, "recall": 0.30000001192092896, "row_count": 1225, "truth_threshold": 13.86852451537379 }, { "FN": 57, "FN_rate": 0.7124999761581421, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 23, "TP_rate": 0.2874999940395355, "precision": 1, "recall": 0.2874999940395355, "row_count": 1225, "truth_threshold": 14.056586088147585 }, { "FN": 58, "FN_rate": 0.7250000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 22, "TP_rate": 0.2750000059604645, "precision": 1, "recall": 0.2750000059604645, "row_count": 1225, "truth_threshold": 14.177875060891724 }, { "FN": 59, "FN_rate": 0.737500011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 21, "TP_rate": 0.26249998807907104, "precision": 1, "recall": 0.26249998807907104, "row_count": 1225, "truth_threshold": 14.207507875396658 }, { "FN": 60, "FN_rate": 0.75, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 20, "TP_rate": 0.25, "precision": 1, "recall": 0.25, "row_count": 1225, "truth_threshold": 14.295584819632225 }, { "FN": 62, "FN_rate": 0.7749999761581421, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 18, "TP_rate": 0.22499999403953552, "precision": 1, "recall": 0.22499999403953552, "row_count": 1225, "truth_threshold": 14.37148330224049 }, { "FN": 63, "FN_rate": 0.7875000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 17, "TP_rate": 0.21250000596046448, "precision": 1, "recall": 0.21250000596046448, "row_count": 1225, "truth_threshold": 15.446568654706729 }, { "FN": 64, "FN_rate": 0.800000011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 16, "TP_rate": 0.20000000298023224, "precision": 1, "recall": 0.20000000298023224, "row_count": 1225, "truth_threshold": 15.634630227480523 }, { "FN": 65, "FN_rate": 0.8125, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 15, "TP_rate": 0.1875, "precision": 1, "recall": 0.1875, "row_count": 1225, "truth_threshold": 17.672735341397793 }, { "FN": 68, "FN_rate": 0.8500000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 12, "TP_rate": 0.15000000596046448, "precision": 1, "recall": 0.15000000596046448, "row_count": 1225, "truth_threshold": 17.977842553783763 }, { "FN": 69, "FN_rate": 0.862500011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 11, "TP_rate": 0.13750000298023224, "precision": 1, "recall": 0.13750000298023224, "row_count": 1225, "truth_threshold": 18.488304982136434 }, { "FN": 70, "FN_rate": 0.875, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 10, "TP_rate": 0.125, "precision": 1, "recall": 0.125, "row_count": 1225, "truth_threshold": 20.626191045723527 }, { "FN": 71, "FN_rate": 0.887499988079071, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 9, "TP_rate": 0.11249999701976776, "precision": 1, "recall": 0.11249999701976776, "row_count": 1225, "truth_threshold": 20.816048705995744 }, { "FN": 72, "FN_rate": 0.8999999761581421, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 8, "TP_rate": 0.10000000149011612, "precision": 1, "recall": 0.10000000149011612, "row_count": 1225, "truth_threshold": 21.004110278769538 }, { "FN": 73, "FN_rate": 0.9125000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 7, "TP_rate": 0.08749999850988388, "precision": 1, "recall": 0.08749999850988388, "row_count": 1225, "truth_threshold": 21.121825061668357 }, { "FN": 74, "FN_rate": 0.925000011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 6, "TP_rate": 0.07500000298023224, "precision": 1, "recall": 0.07500000298023224, "row_count": 1225, "truth_threshold": 21.74753704213478 }, { "FN": 75, "FN_rate": 0.9375, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 5, "TP_rate": 0.0625, "precision": 1, "recall": 0.0625, "row_count": 1225, "truth_threshold": 22.053313397807393 }, { "FN": 76, "FN_rate": 0.949999988079071, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 4, "TP_rate": 0.05000000074505806, "precision": 1, "recall": 0.05000000074505806, "row_count": 1225, "truth_threshold": 23.158090222901993 }, { "FN": 77, "FN_rate": 0.9624999761581421, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 3, "TP_rate": 0.03750000149011612, "precision": 1, "recall": 0.03750000149011612, "row_count": 1225, "truth_threshold": 25.57656671132397 } ] }, "encoding": { "tooltip": [ { "field": "truth_threshold", "format": ".4f", "type": "quantitative" }, { "field": "FP_rate", "format": ".4f", "type": "quantitative" }, { "field": "TP_rate", "format": ".4f", "type": "quantitative" }, { "field": "TP", "format": ",.0f", "type": "quantitative" }, { "field": "TN", "format": ",.0f", "type": "quantitative" }, { "field": "FP", "format": ",.0f", "type": "quantitative" }, { "field": "FN", "format": ",.0f", "type": "quantitative" }, { "field": "precision", "format": ".4f", "type": "quantitative" }, { "field": "recall", "format": ".4f", "type": "quantitative" } ], "x": { "field": "FP_rate", "sort": [ "truth_threshold" ], "title": "False Positive Rate amongst clerically reviewed records", "type": "quantitative" }, "y": { "field": "TP_rate", "sort": [ "truth_threshold" ], "title": "True Positive Rate amongst clerically reviewed records", "type": "quantitative" } }, "height": 400, "mark": { "clip": true, "point": true, "type": "line" }, "selection": { "selector076": { "bind": "scales", "encodings": [ "x" ], "type": "interval" } }, "title": "Receiver operating characteristic curve", "width": 400 }, "image/png": "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", "text/plain": [ "\n", "\n", "If you see this message, it means the renderer has not been properly enabled\n", "for the frontend that you are using. For more information, see\n", "https://altair-viz.github.io/user_guide/troubleshooting.html\n" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "linker.roc_chart_from_labels(\"labels\")" ] }, { "cell_type": "markdown", "id": "9f749c3c", "metadata": {}, "source": [ "### Precision-recall chart\n", "\n", "An alternative representation of truth space is called a [precision recall curve](https://stats.stackexchange.com/questions/7207/roc-vs-precision-and-recall-curves).\n", "\n", "This can be plotted as follows:" ] }, { "cell_type": "code", "execution_count": 4, "id": "18d25327", "metadata": {}, "outputs": [ { "data": { "application/vnd.vegalite.v4+json": { "$schema": "https://vega.github.io/schema/vega-lite/v4.8.1.json", "data": { "values": [ { "FN": 0, "FN_rate": 0, "FP": 1145, "FP_rate": 1, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 0, "TN_rate": 0, "TP": 80, "TP_rate": 1, "precision": 0.06530611962080002, "recall": 1, "row_count": 1225, "truth_threshold": -16.60994994256404 }, { "FN": 0, "FN_rate": 0, "FP": 1047, "FP_rate": 0.9144104719161987, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 98, "TN_rate": 0.08558952063322067, "TP": 80, "TP_rate": 1, "precision": 0.07098491489887238, "recall": 1, "row_count": 1225, "truth_threshold": -15.488603946152788 }, { "FN": 0, "FN_rate": 0, "FP": 877, "FP_rate": 0.7659388780593872, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 268, "TN_rate": 0.234061136841774, "TP": 80, "TP_rate": 1, "precision": 0.08359456807374954, "recall": 1, "row_count": 1225, "truth_threshold": -14.677083359952446 }, { "FN": 0, "FN_rate": 0, "FP": 835, "FP_rate": 0.7292576432228088, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 310, "TN_rate": 0.27074235677719116, "TP": 80, "TP_rate": 1, "precision": 0.08743169158697128, "recall": 1, "row_count": 1225, "truth_threshold": -14.535475051974686 }, { "FN": 0, "FN_rate": 0, "FP": 749, "FP_rate": 0.6541484594345093, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 396, "TN_rate": 0.3458515405654907, "TP": 80, "TP_rate": 1, "precision": 0.09650181233882904, "recall": 1, "row_count": 1225, "truth_threshold": -14.078050765385573 }, { "FN": 0, "FN_rate": 0, "FP": 733, "FP_rate": 0.6401746869087219, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 412, "TN_rate": 0.3598253130912781, "TP": 80, "TP_rate": 1, "precision": 0.09840098768472672, "recall": 1, "row_count": 1225, "truth_threshold": -13.764688729777017 }, { "FN": 0, "FN_rate": 0, "FP": 656, "FP_rate": 0.5729257464408875, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 489, "TN_rate": 0.42707422375679016, "TP": 80, "TP_rate": 1, "precision": 0.10869564861059189, "recall": 1, "row_count": 1225, "truth_threshold": -13.555737363541194 }, { "FN": 0, "FN_rate": 0, "FP": 549, "FP_rate": 0.4794759750366211, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 596, "TN_rate": 0.5205240249633789, "TP": 80, "TP_rate": 1, "precision": 0.12718601524829865, "recall": 1, "row_count": 1225, "truth_threshold": -13.414129055563434 }, { "FN": 0, "FN_rate": 0, "FP": 509, "FP_rate": 0.44454148411750793, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 636, "TN_rate": 0.5554584860801697, "TP": 80, "TP_rate": 1, "precision": 0.13582342863082886, "recall": 1, "row_count": 1225, "truth_threshold": -12.643342733365765 }, { "FN": 1, "FN_rate": 0.012500000186264515, "FP": 419, "FP_rate": 0.36593887209892273, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 726, "TN_rate": 0.6340611577033997, "TP": 79, "TP_rate": 0.987500011920929, "precision": 0.15863454341888428, "recall": 0.987500011920929, "row_count": 1225, "truth_threshold": -12.602608469363092 }, { "FN": 1, "FN_rate": 0.012500000186264515, "FP": 384, "FP_rate": 0.3353711664676666, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 761, "TN_rate": 0.664628803730011, "TP": 79, "TP_rate": 0.987500011920929, "precision": 0.17062635719776154, "recall": 0.987500011920929, "row_count": 1225, "truth_threshold": -12.14518418277398 }, { "FN": 1, "FN_rate": 0.012500000186264515, "FP": 378, "FP_rate": 0.33013099431991577, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 767, "TN_rate": 0.6698690056800842, "TP": 79, "TP_rate": 0.987500011920929, "precision": 0.17286652326583862, "recall": 0.987500011920929, "row_count": 1225, "truth_threshold": -11.831822147165424 }, { "FN": 1, "FN_rate": 0.012500000186264515, "FP": 365, "FP_rate": 0.31877729296684265, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 780, "TN_rate": 0.6812227368354797, "TP": 79, "TP_rate": 0.987500011920929, "precision": 0.1779279261827469, "recall": 0.987500011920929, "row_count": 1225, "truth_threshold": -11.690213839187663 }, { "FN": 2, "FN_rate": 0.02500000037252903, "FP": 273, "FP_rate": 0.23842795193195343, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 872, "TN_rate": 0.7615720629692078, "TP": 78, "TP_rate": 0.9750000238418579, "precision": 0.2222222238779068, "recall": 0.9750000238418579, "row_count": 1225, "truth_threshold": -11.48126247295184 }, { "FN": 2, "FN_rate": 0.02500000037252903, "FP": 250, "FP_rate": 0.21834060549736023, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 895, "TN_rate": 0.7816593647003174, "TP": 78, "TP_rate": 0.9750000238418579, "precision": 0.23780487477779388, "recall": 0.9750000238418579, "row_count": 1225, "truth_threshold": -11.23278955259855 }, { "FN": 2, "FN_rate": 0.02500000037252903, "FP": 243, "FP_rate": 0.21222707629203796, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 902, "TN_rate": 0.7877729535102844, "TP": 78, "TP_rate": 0.9750000238418579, "precision": 0.2429906576871872, "recall": 0.9750000238418579, "row_count": 1225, "truth_threshold": -10.906098261537272 }, { "FN": 2, "FN_rate": 0.02500000037252903, "FP": 238, "FP_rate": 0.20786026120185852, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 907, "TN_rate": 0.7921397089958191, "TP": 78, "TP_rate": 0.9750000238418579, "precision": 0.24683544039726257, "recall": 0.9750000238418579, "row_count": 1225, "truth_threshold": -10.710476150754172 }, { "FN": 2, "FN_rate": 0.02500000037252903, "FP": 215, "FP_rate": 0.1877729296684265, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 930, "TN_rate": 0.8122270703315735, "TP": 78, "TP_rate": 0.9750000238418579, "precision": 0.266211599111557, "recall": 0.9750000238418579, "row_count": 1225, "truth_threshold": -10.56886784277641 }, { "FN": 3, "FN_rate": 0.03750000149011612, "FP": 98, "FP_rate": 0.08558952063322067, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1047, "TN_rate": 0.9144104719161987, "TP": 77, "TP_rate": 0.9624999761581421, "precision": 0.4399999976158142, "recall": 0.9624999761581421, "row_count": 1225, "truth_threshold": -9.793937423031654 }, { "FN": 3, "FN_rate": 0.03750000149011612, "FP": 95, "FP_rate": 0.08296943455934525, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1050, "TN_rate": 0.9170305728912354, "TP": 77, "TP_rate": 0.9624999761581421, "precision": 0.447674423456192, "recall": 0.9624999761581421, "row_count": 1225, "truth_threshold": -9.78475226512602 }, { "FN": 3, "FN_rate": 0.03750000149011612, "FP": 91, "FP_rate": 0.07947598397731781, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1054, "TN_rate": 0.920524001121521, "TP": 77, "TP_rate": 0.9624999761581421, "precision": 0.4583333432674408, "recall": 0.9624999761581421, "row_count": 1225, "truth_threshold": -9.75734725657607 }, { "FN": 3, "FN_rate": 0.03750000149011612, "FP": 75, "FP_rate": 0.06550218164920807, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1070, "TN_rate": 0.9344978332519531, "TP": 77, "TP_rate": 0.9624999761581421, "precision": 0.5065789222717285, "recall": 0.9624999761581421, "row_count": 1225, "truth_threshold": -9.299922969986957 }, { "FN": 3, "FN_rate": 0.03750000149011612, "FP": 74, "FP_rate": 0.06462882459163666, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1071, "TN_rate": 0.9353711605072021, "TP": 77, "TP_rate": 0.9624999761581421, "precision": 0.5099337697029114, "recall": 0.9624999761581421, "row_count": 1225, "truth_threshold": -9.158314662009195 }, { "FN": 5, "FN_rate": 0.0625, "FP": 69, "FP_rate": 0.060262009501457214, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1076, "TN_rate": 0.9397379755973816, "TP": 75, "TP_rate": 0.9375, "precision": 0.5208333134651184, "recall": 0.9375, "row_count": 1225, "truth_threshold": -9.057240845364372 }, { "FN": 5, "FN_rate": 0.0625, "FP": 68, "FP_rate": 0.059388644993305206, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1077, "TN_rate": 0.9406113624572754, "TP": 75, "TP_rate": 0.9375, "precision": 0.5244755148887634, "recall": 0.9375, "row_count": 1225, "truth_threshold": -8.831623370947918 }, { "FN": 5, "FN_rate": 0.0625, "FP": 66, "FP_rate": 0.05764191970229149, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1079, "TN_rate": 0.9423580765724182, "TP": 75, "TP_rate": 0.9375, "precision": 0.5319148898124695, "recall": 0.9375, "row_count": 1225, "truth_threshold": -8.672591426620402 }, { "FN": 5, "FN_rate": 0.0625, "FP": 61, "FP_rate": 0.053275108337402344, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1084, "TN_rate": 0.9467248916625977, "TP": 75, "TP_rate": 0.9375, "precision": 0.5514705777168274, "recall": 0.9375, "row_count": 1225, "truth_threshold": -8.636001260164818 }, { "FN": 5, "FN_rate": 0.0625, "FP": 39, "FP_rate": 0.03406113386154175, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1106, "TN_rate": 0.9659388661384583, "TP": 75, "TP_rate": 0.9375, "precision": 0.6578947305679321, "recall": 0.9375, "row_count": 1225, "truth_threshold": -8.06083704875025 }, { "FN": 5, "FN_rate": 0.0625, "FP": 35, "FP_rate": 0.030567685142159462, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1110, "TN_rate": 0.9694322943687439, "TP": 75, "TP_rate": 0.9375, "precision": 0.6818181872367859, "recall": 0.9375, "row_count": 1225, "truth_threshold": -7.8518856825144265 }, { "FN": 5, "FN_rate": 0.0625, "FP": 33, "FP_rate": 0.028820959851145744, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1112, "TN_rate": 0.9711790680885315, "TP": 75, "TP_rate": 0.9375, "precision": 0.6944444179534912, "recall": 0.9375, "row_count": 1225, "truth_threshold": -7.738872648601806 }, { "FN": 5, "FN_rate": 0.0625, "FP": 30, "FP_rate": 0.026200873777270317, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1115, "TN_rate": 0.9737991094589233, "TP": 75, "TP_rate": 0.9375, "precision": 0.7142857313156128, "recall": 0.9375, "row_count": 1225, "truth_threshold": -7.710277374536665 }, { "FN": 5, "FN_rate": 0.0625, "FP": 29, "FP_rate": 0.025327511131763458, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1116, "TN_rate": 0.9746724963188171, "TP": 75, "TP_rate": 0.9375, "precision": 0.7211538553237915, "recall": 0.9375, "row_count": 1225, "truth_threshold": -7.262038245853187 }, { "FN": 5, "FN_rate": 0.0625, "FP": 27, "FP_rate": 0.02358078584074974, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1118, "TN_rate": 0.97641921043396, "TP": 75, "TP_rate": 0.9375, "precision": 0.7352941036224365, "recall": 0.9375, "row_count": 1225, "truth_threshold": -7.225448079397602 }, { "FN": 5, "FN_rate": 0.0625, "FP": 26, "FP_rate": 0.022707423195242882, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1119, "TN_rate": 0.9772925972938538, "TP": 75, "TP_rate": 0.9375, "precision": 0.7425742745399475, "recall": 0.9375, "row_count": 1225, "truth_threshold": -6.982765954775019 }, { "FN": 5, "FN_rate": 0.0625, "FP": 25, "FP_rate": 0.021834060549736023, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1120, "TN_rate": 0.9781659245491028, "TP": 75, "TP_rate": 0.9375, "precision": 0.75, "recall": 0.9375, "row_count": 1225, "truth_threshold": -6.9394910523389965 }, { "FN": 5, "FN_rate": 0.0625, "FP": 21, "FP_rate": 0.018340611830353737, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1124, "TN_rate": 0.9816594123840332, "TP": 75, "TP_rate": 0.9375, "precision": 0.78125, "recall": 0.9375, "row_count": 1225, "truth_threshold": -6.898756788336325 }, { "FN": 5, "FN_rate": 0.0625, "FP": 18, "FP_rate": 0.01572052389383316, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1127, "TN_rate": 0.984279453754425, "TP": 75, "TP_rate": 0.9375, "precision": 0.8064516186714172, "recall": 0.9375, "row_count": 1225, "truth_threshold": -5.8273302138333785 }, { "FN": 5, "FN_rate": 0.0625, "FP": 12, "FP_rate": 0.010480348952114582, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1133, "TN_rate": 0.9895196557044983, "TP": 75, "TP_rate": 0.9375, "precision": 0.8620689511299133, "recall": 0.9375, "row_count": 1225, "truth_threshold": -5.777410791925072 }, { "FN": 5, "FN_rate": 0.0625, "FP": 9, "FP_rate": 0.00786026194691658, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1136, "TN_rate": 0.9921397566795349, "TP": 75, "TP_rate": 0.9375, "precision": 0.8928571343421936, "recall": 0.9375, "row_count": 1225, "truth_threshold": -5.2485131299666525 }, { "FN": 6, "FN_rate": 0.07500000298023224, "FP": 9, "FP_rate": 0.00786026194691658, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1136, "TN_rate": 0.9921397566795349, "TP": 74, "TP_rate": 0.925000011920929, "precision": 0.891566276550293, "recall": 0.925000011920929, "row_count": 1225, "truth_threshold": -5.006624469727403 }, { "FN": 7, "FN_rate": 0.08749999850988388, "FP": 7, "FP_rate": 0.00611353712156415, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1138, "TN_rate": 0.9938864707946777, "TP": 73, "TP_rate": 0.9125000238418579, "precision": 0.9125000238418579, "recall": 0.9125000238418579, "row_count": 1225, "truth_threshold": -4.893611435814783 }, { "FN": 7, "FN_rate": 0.08749999850988388, "FP": 6, "FP_rate": 0.005240174476057291, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1139, "TN_rate": 0.9947597980499268, "TP": 73, "TP_rate": 0.9125000238418579, "precision": 0.9240506291389465, "recall": 0.9125000238418579, "row_count": 1225, "truth_threshold": -4.865016161749642 }, { "FN": 7, "FN_rate": 0.08749999850988388, "FP": 4, "FP_rate": 0.0034934498835355043, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1141, "TN_rate": 0.9965065717697144, "TP": 73, "TP_rate": 0.9125000238418579, "precision": 0.948051929473877, "recall": 0.9125000238418579, "row_count": 1225, "truth_threshold": -4.646615423705956 }, { "FN": 8, "FN_rate": 0.10000000149011612, "FP": 4, "FP_rate": 0.0034934498835355043, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1141, "TN_rate": 0.9965065717697144, "TP": 72, "TP_rate": 0.8999999761581421, "precision": 0.9473684430122375, "recall": 0.8999999761581421, "row_count": 1225, "truth_threshold": -4.137504741987995 }, { "FN": 8, "FN_rate": 0.10000000149011612, "FP": 3, "FP_rate": 0.0026200872380286455, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1142, "TN_rate": 0.9973798990249634, "TP": 72, "TP_rate": 0.8999999761581421, "precision": 0.9599999785423279, "recall": 0.8999999761581421, "row_count": 1225, "truth_threshold": -3.3386640862492003 }, { "FN": 9, "FN_rate": 0.11249999701976776, "FP": 3, "FP_rate": 0.0026200872380286455, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1142, "TN_rate": 0.9973798990249634, "TP": 71, "TP_rate": 0.887499988079071, "precision": 0.9594594836235046, "recall": 0.887499988079071, "row_count": 1225, "truth_threshold": -2.932149579138049 }, { "FN": 9, "FN_rate": 0.11249999701976776, "FP": 1, "FP_rate": 0.0008733624708838761, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1144, "TN_rate": 0.9991266131401062, "TP": 71, "TP_rate": 0.887499988079071, "precision": 0.9861111044883728, "recall": 0.887499988079071, "row_count": 1225, "truth_threshold": -2.8311489013971385 }, { "FN": 10, "FN_rate": 0.125, "FP": 1, "FP_rate": 0.0008733624708838761, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1144, "TN_rate": 0.9991266131401062, "TP": 70, "TP_rate": 0.875, "precision": 0.98591548204422, "recall": 0.875, "row_count": 1225, "truth_threshold": -2.1572191593932932 }, { "FN": 10, "FN_rate": 0.125, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 70, "TP_rate": 0.875, "precision": 1, "recall": 0.875, "row_count": 1225, "truth_threshold": -0.7308664264624687 }, { "FN": 11, "FN_rate": 0.13750000298023224, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 69, "TP_rate": 0.862500011920929, "precision": 1, "recall": 0.862500011920929, "row_count": 1225, "truth_threshold": -0.36988609476305273 }, { "FN": 12, "FN_rate": 0.15000000596046448, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 68, "TP_rate": 0.8500000238418579, "precision": 1, "recall": 0.8500000238418579, "row_count": 1225, "truth_threshold": 1.1433212648855802 }, { "FN": 13, "FN_rate": 0.16249999403953552, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 67, "TP_rate": 0.8374999761581421, "precision": 1, "recall": 0.8374999761581421, "row_count": 1225, "truth_threshold": 1.8093880498049824 }, { "FN": 14, "FN_rate": 0.17499999701976776, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 66, "TP_rate": 0.824999988079071, "precision": 1, "recall": 0.824999988079071, "row_count": 1225, "truth_threshold": 2.162013082415415 }, { "FN": 15, "FN_rate": 0.1875, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 65, "TP_rate": 0.8125, "precision": 1, "recall": 0.8125, "row_count": 1225, "truth_threshold": 2.2811784283388743 }, { "FN": 16, "FN_rate": 0.20000000298023224, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 64, "TP_rate": 0.800000011920929, "precision": 1, "recall": 0.800000011920929, "row_count": 1225, "truth_threshold": 2.467789438088027 }, { "FN": 17, "FN_rate": 0.21250000596046448, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 63, "TP_rate": 0.7875000238418579, "precision": 1, "recall": 0.7875000238418579, "row_count": 1225, "truth_threshold": 3.5624320737970847 }, { "FN": 19, "FN_rate": 0.23749999701976776, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 61, "TP_rate": 0.762499988079071, "precision": 1, "recall": 0.762499988079071, "row_count": 1225, "truth_threshold": 4.2540204580463135 }, { "FN": 20, "FN_rate": 0.25, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 60, "TP_rate": 0.75, "precision": 1, "recall": 0.75, "row_count": 1225, "truth_threshold": 4.6580632234106085 }, { "FN": 24, "FN_rate": 0.30000001192092896, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 56, "TP_rate": 0.699999988079071, "precision": 1, "recall": 0.699999988079071, "row_count": 1225, "truth_threshold": 5.132707101344441 }, { "FN": 26, "FN_rate": 0.32499998807907104, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 54, "TP_rate": 0.675000011920929, "precision": 1, "recall": 0.675000011920929, "row_count": 1225, "truth_threshold": 5.473632864149248 }, { "FN": 27, "FN_rate": 0.3375000059604645, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 53, "TP_rate": 0.6625000238418579, "precision": 1, "recall": 0.6625000238418579, "row_count": 1225, "truth_threshold": 5.685026698413226 }, { "FN": 28, "FN_rate": 0.3499999940395355, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 52, "TP_rate": 0.6499999761581421, "precision": 1, "recall": 0.6499999761581421, "row_count": 1225, "truth_threshold": 6.328495348635668 }, { "FN": 30, "FN_rate": 0.375, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 50, "TP_rate": 0.625, "precision": 1, "recall": 0.625, "row_count": 1225, "truth_threshold": 6.516556921409461 }, { "FN": 32, "FN_rate": 0.4000000059604645, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 48, "TP_rate": 0.6000000238418579, "precision": 1, "recall": 0.6000000238418579, "row_count": 1225, "truth_threshold": 6.9481736236532585 }, { "FN": 35, "FN_rate": 0.4375, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 45, "TP_rate": 0.5625, "precision": 1, "recall": 0.5625, "row_count": 1225, "truth_threshold": 7.449841345046921 }, { "FN": 36, "FN_rate": 0.44999998807907104, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 44, "TP_rate": 0.550000011920929, "precision": 1, "recall": 0.550000011920929, "row_count": 1225, "truth_threshold": 7.637902917820714 }, { "FN": 38, "FN_rate": 0.4749999940395355, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 42, "TP_rate": 0.5249999761581421, "precision": 1, "recall": 0.5249999761581421, "row_count": 1225, "truth_threshold": 7.712926222461523 }, { "FN": 39, "FN_rate": 0.48750001192092896, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 41, "TP_rate": 0.512499988079071, "precision": 1, "recall": 0.512499988079071, "row_count": 1225, "truth_threshold": 8.859769247712702 }, { "FN": 42, "FN_rate": 0.5249999761581421, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 38, "TP_rate": 0.4749999940395355, "precision": 1, "recall": 0.4749999940395355, "row_count": 1225, "truth_threshold": 9.14004857454539 }, { "FN": 43, "FN_rate": 0.5375000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 37, "TP_rate": 0.4625000059604645, "precision": 1, "recall": 0.4625000059604645, "row_count": 1225, "truth_threshold": 9.676008031737986 }, { "FN": 44, "FN_rate": 0.550000011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 36, "TP_rate": 0.44999998807907104, "precision": 1, "recall": 0.44999998807907104, "row_count": 1225, "truth_threshold": 10.089978878949308 }, { "FN": 45, "FN_rate": 0.5625, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 35, "TP_rate": 0.4375, "precision": 1, "recall": 0.4375, "row_count": 1225, "truth_threshold": 11.258293846460244 }, { "FN": 46, "FN_rate": 0.574999988079071, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 34, "TP_rate": 0.42500001192092896, "precision": 1, "recall": 0.42500001192092896, "row_count": 1225, "truth_threshold": 11.317074190024028 }, { "FN": 47, "FN_rate": 0.5874999761581421, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 33, "TP_rate": 0.4124999940395355, "precision": 1, "recall": 0.4124999940395355, "row_count": 1225, "truth_threshold": 12.297008411651136 }, { "FN": 48, "FN_rate": 0.6000000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 32, "TP_rate": 0.4000000059604645, "precision": 1, "recall": 0.4000000059604645, "row_count": 1225, "truth_threshold": 12.58803772594323 }, { "FN": 49, "FN_rate": 0.612500011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 31, "TP_rate": 0.38749998807907104, "precision": 1, "recall": 0.38749998807907104, "row_count": 1225, "truth_threshold": 12.605377590133758 }, { "FN": 50, "FN_rate": 0.625, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 30, "TP_rate": 0.375, "precision": 1, "recall": 0.375, "row_count": 1225, "truth_threshold": 12.793439162907552 }, { "FN": 53, "FN_rate": 0.6625000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 27, "TP_rate": 0.3375000059604645, "precision": 1, "recall": 0.3375000059604645, "row_count": 1225, "truth_threshold": 12.93524009173633 }, { "FN": 54, "FN_rate": 0.675000011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 26, "TP_rate": 0.32499998807907104, "precision": 1, "recall": 0.32499998807907104, "row_count": 1225, "truth_threshold": 13.125097752008548 }, { "FN": 56, "FN_rate": 0.699999988079071, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 24, "TP_rate": 0.30000001192092896, "precision": 1, "recall": 0.30000001192092896, "row_count": 1225, "truth_threshold": 13.86852451537379 }, { "FN": 57, "FN_rate": 0.7124999761581421, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 23, "TP_rate": 0.2874999940395355, "precision": 1, "recall": 0.2874999940395355, "row_count": 1225, "truth_threshold": 14.056586088147585 }, { "FN": 58, "FN_rate": 0.7250000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 22, "TP_rate": 0.2750000059604645, "precision": 1, "recall": 0.2750000059604645, "row_count": 1225, "truth_threshold": 14.177875060891724 }, { "FN": 59, "FN_rate": 0.737500011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 21, "TP_rate": 0.26249998807907104, "precision": 1, "recall": 0.26249998807907104, "row_count": 1225, "truth_threshold": 14.207507875396658 }, { "FN": 60, "FN_rate": 0.75, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 20, "TP_rate": 0.25, "precision": 1, "recall": 0.25, "row_count": 1225, "truth_threshold": 14.295584819632225 }, { "FN": 62, "FN_rate": 0.7749999761581421, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 18, "TP_rate": 0.22499999403953552, "precision": 1, "recall": 0.22499999403953552, "row_count": 1225, "truth_threshold": 14.37148330224049 }, { "FN": 63, "FN_rate": 0.7875000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 17, "TP_rate": 0.21250000596046448, "precision": 1, "recall": 0.21250000596046448, "row_count": 1225, "truth_threshold": 15.446568654706729 }, { "FN": 64, "FN_rate": 0.800000011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 16, "TP_rate": 0.20000000298023224, "precision": 1, "recall": 0.20000000298023224, "row_count": 1225, "truth_threshold": 15.634630227480523 }, { "FN": 65, "FN_rate": 0.8125, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 15, "TP_rate": 0.1875, "precision": 1, "recall": 0.1875, "row_count": 1225, "truth_threshold": 17.672735341397793 }, { "FN": 68, "FN_rate": 0.8500000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 12, "TP_rate": 0.15000000596046448, "precision": 1, "recall": 0.15000000596046448, "row_count": 1225, "truth_threshold": 17.977842553783763 }, { "FN": 69, "FN_rate": 0.862500011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 11, "TP_rate": 0.13750000298023224, "precision": 1, "recall": 0.13750000298023224, "row_count": 1225, "truth_threshold": 18.488304982136434 }, { "FN": 70, "FN_rate": 0.875, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 10, "TP_rate": 0.125, "precision": 1, "recall": 0.125, "row_count": 1225, "truth_threshold": 20.626191045723527 }, { "FN": 71, "FN_rate": 0.887499988079071, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 9, "TP_rate": 0.11249999701976776, "precision": 1, "recall": 0.11249999701976776, "row_count": 1225, "truth_threshold": 20.816048705995744 }, { "FN": 72, "FN_rate": 0.8999999761581421, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 8, "TP_rate": 0.10000000149011612, "precision": 1, "recall": 0.10000000149011612, "row_count": 1225, "truth_threshold": 21.004110278769538 }, { "FN": 73, "FN_rate": 0.9125000238418579, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 7, "TP_rate": 0.08749999850988388, "precision": 1, "recall": 0.08749999850988388, "row_count": 1225, "truth_threshold": 21.121825061668357 }, { "FN": 74, "FN_rate": 0.925000011920929, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 6, "TP_rate": 0.07500000298023224, "precision": 1, "recall": 0.07500000298023224, "row_count": 1225, "truth_threshold": 21.74753704213478 }, { "FN": 75, "FN_rate": 0.9375, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 5, "TP_rate": 0.0625, "precision": 1, "recall": 0.0625, "row_count": 1225, "truth_threshold": 22.053313397807393 }, { "FN": 76, "FN_rate": 0.949999988079071, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 4, "TP_rate": 0.05000000074505806, "precision": 1, "recall": 0.05000000074505806, "row_count": 1225, "truth_threshold": 23.158090222901993 }, { "FN": 77, "FN_rate": 0.9624999761581421, "FP": 0, "FP_rate": 0, "N": 1145, "N_rate": 0.9346938729286194, "P": 80, "P_rate": 0, "TN": 1145, "TN_rate": 1, "TP": 3, "TP_rate": 0.03750000149011612, "precision": 1, "recall": 0.03750000149011612, "row_count": 1225, "truth_threshold": 25.57656671132397 } ] }, "encoding": { "tooltip": [ { "field": "truth_threshold", "format": ".4f", "type": "quantitative" }, { "field": "FP_rate", "format": ".4f", "type": "quantitative" }, { "field": "TP_rate", "format": ".4f", "type": "quantitative" }, { "field": "TP", "format": ",.0f", "type": "quantitative" }, { "field": "TN", "format": ",.0f", "type": "quantitative" }, { "field": "FP", "format": ",.0f", "type": "quantitative" }, { "field": "FN", "format": ",.0f", "type": "quantitative" }, { "field": "precision", "format": ".4f", "type": "quantitative" }, { "field": "recall", "format": ".4f", "type": "quantitative" } ], "x": { "field": "recall", "sort": [ "-recall" ], "title": "Recall", "type": "quantitative" }, "y": { "field": "precision", "sort": [ "-precision" ], "title": "Precision", "type": "quantitative" } }, "height": 400, "mark": { "clip": true, "point": true, "type": "line" }, "title": "Precision-recall curve", "width": 400 }, "image/png": "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", "text/plain": [ "\n", "\n", "If you see this message, it means the renderer has not been properly enabled\n", "for the frontend that you are using. For more information, see\n", "https://altair-viz.github.io/user_guide/troubleshooting.html\n" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "linker.precision_recall_chart_from_labels(\"labels\")" ] }, { "cell_type": "markdown", "id": "12e6ba74", "metadata": {}, "source": [ "## Truth table\n", "\n", "Fnally, Splink can also report the underlying table used to construct the ROC and precision recall curves." ] }, { "cell_type": "code", "execution_count": 5, "id": "f7c283ba", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
truth_thresholdrow_countPNTPTNFPFNP_rateN_rateTP_rateTN_rateFP_rateFN_rateprecisionrecall
0-16.6099501225.080.01145.080.00.01145.00.00.00.9346941.00.0000001.0000000.00.0653061.0
1-15.4886041225.080.01145.080.098.01047.00.00.00.9346941.00.0855900.9144100.00.0709851.0
2-14.6770831225.080.01145.080.0268.0877.00.00.00.9346941.00.2340610.7659390.00.0835951.0
3-14.5354751225.080.01145.080.0310.0835.00.00.00.9346941.00.2707420.7292580.00.0874321.0
4-14.0780511225.080.01145.080.0396.0749.00.00.00.9346941.00.3458520.6541480.00.0965021.0
\n", "
" ], "text/plain": [ " truth_threshold row_count P N TP TN FP FN P_rate \\\n", "0 -16.609950 1225.0 80.0 1145.0 80.0 0.0 1145.0 0.0 0.0 \n", "1 -15.488604 1225.0 80.0 1145.0 80.0 98.0 1047.0 0.0 0.0 \n", "2 -14.677083 1225.0 80.0 1145.0 80.0 268.0 877.0 0.0 0.0 \n", "3 -14.535475 1225.0 80.0 1145.0 80.0 310.0 835.0 0.0 0.0 \n", "4 -14.078051 1225.0 80.0 1145.0 80.0 396.0 749.0 0.0 0.0 \n", "\n", " N_rate TP_rate TN_rate FP_rate FN_rate precision recall \n", "0 0.934694 1.0 0.000000 1.000000 0.0 0.065306 1.0 \n", "1 0.934694 1.0 0.085590 0.914410 0.0 0.070985 1.0 \n", "2 0.934694 1.0 0.234061 0.765939 0.0 0.083595 1.0 \n", "3 0.934694 1.0 0.270742 0.729258 0.0 0.087432 1.0 \n", "4 0.934694 1.0 0.345852 0.654148 0.0 0.096502 1.0 " ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "linker._initialise_df_concat_with_tf()\n", "roc_table = linker.roc_table_from_labels(\"labels\")\n", "roc_table.as_pandas_dataframe(limit=5)" ] } ], "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": "3.8.3" }, "vscode": { "interpreter": { "hash": "3b53fa520a31e303a9636a08ff10a3bbc14893ee50cb37445791fa59628fc75b" } } }, "nbformat": 4, "nbformat_minor": 5 }