201 lines
20 KiB
HTML
201 lines
20 KiB
HTML
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
|
|
|
|
<html lang="en">
|
|
|
|
<head>
|
|
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
|
|
<title>LCOV - BayesNet Coverage Report - bayesnet/classifiers/Proposal.cc</title>
|
|
<link rel="stylesheet" type="text/css" href="../../gcov.css">
|
|
</head>
|
|
|
|
<body>
|
|
|
|
<table width="100%" border=0 cellspacing=0 cellpadding=0>
|
|
<tr><td class="title">LCOV - code coverage report</td></tr>
|
|
<tr><td class="ruler"><img src="../../glass.png" width=3 height=3 alt=""></td></tr>
|
|
|
|
<tr>
|
|
<td width="100%">
|
|
<table cellpadding=1 border=0 width="100%">
|
|
<tr>
|
|
<td width="10%" class="headerItem">Current view:</td>
|
|
<td width="10%" class="headerValue"><a href="../../index.html" target="_parent">top level</a> - <a href="index.html" target="_parent">bayesnet/classifiers</a> - Proposal.cc<span style="font-size: 80%;"> (source / <a href="Proposal.cc.func-c.html">functions</a>)</span></td>
|
|
<td width="5%"></td>
|
|
<td width="5%"></td>
|
|
<td width="5%" class="headerCovTableHead">Coverage</td>
|
|
<td width="5%" class="headerCovTableHead" title="Covered + Uncovered code">Total</td>
|
|
<td width="5%" class="headerCovTableHead" title="Exercised code only">Hit</td>
|
|
</tr>
|
|
<tr>
|
|
<td class="headerItem">Test:</td>
|
|
<td class="headerValue">BayesNet Coverage Report</td>
|
|
<td></td>
|
|
<td class="headerItem">Lines:</td>
|
|
<td class="headerCovTableEntryHi">97.7 %</td>
|
|
<td class="headerCovTableEntry">86</td>
|
|
<td class="headerCovTableEntry">84</td>
|
|
</tr>
|
|
<tr>
|
|
<td class="headerItem">Test Date:</td>
|
|
<td class="headerValue">2024-05-06 17:54:04</td>
|
|
<td></td>
|
|
<td class="headerItem">Functions:</td>
|
|
<td class="headerCovTableEntryHi">100.0 %</td>
|
|
<td class="headerCovTableEntry">8</td>
|
|
<td class="headerCovTableEntry">8</td>
|
|
</tr>
|
|
<tr>
|
|
<td class="headerItem">Legend:</td>
|
|
<td class="headerValueLeg"> Lines:
|
|
<span class="coverLegendCov">hit</span>
|
|
<span class="coverLegendNoCov">not hit</span>
|
|
</td>
|
|
<td></td>
|
|
</tr>
|
|
<tr><td><img src="../../glass.png" width=3 height=3 alt=""></td></tr>
|
|
</table>
|
|
</td>
|
|
</tr>
|
|
|
|
<tr><td class="ruler"><img src="../../glass.png" width=3 height=3 alt=""></td></tr>
|
|
</table>
|
|
|
|
<table cellpadding=0 cellspacing=0 border=0>
|
|
<tr>
|
|
<td><br></td>
|
|
</tr>
|
|
<tr>
|
|
<td>
|
|
<pre class="sourceHeading"> Line data Source code</pre>
|
|
<pre class="source">
|
|
<span id="L1"><span class="lineNum"> 1</span> : // ***************************************************************</span>
|
|
<span id="L2"><span class="lineNum"> 2</span> : // SPDX-FileCopyrightText: Copyright 2024 Ricardo Montañana Gómez</span>
|
|
<span id="L3"><span class="lineNum"> 3</span> : // SPDX-FileType: SOURCE</span>
|
|
<span id="L4"><span class="lineNum"> 4</span> : // SPDX-License-Identifier: MIT</span>
|
|
<span id="L5"><span class="lineNum"> 5</span> : // ***************************************************************</span>
|
|
<span id="L6"><span class="lineNum"> 6</span> : </span>
|
|
<span id="L7"><span class="lineNum"> 7</span> : #include <ArffFiles.h></span>
|
|
<span id="L8"><span class="lineNum"> 8</span> : #include "Proposal.h"</span>
|
|
<span id="L9"><span class="lineNum"> 9</span> : </span>
|
|
<span id="L10"><span class="lineNum"> 10</span> : namespace bayesnet {</span>
|
|
<span id="L11"><span class="lineNum"> 11</span> <span class="tlaGNC tlaBgGNC"> 424 : Proposal::Proposal(torch::Tensor& dataset_, std::vector<std::string>& features_, std::string& className_) : pDataset(dataset_), pFeatures(features_), pClassName(className_) {}</span></span>
|
|
<span id="L12"><span class="lineNum"> 12</span> <span class="tlaGNC"> 200 : Proposal::~Proposal()</span></span>
|
|
<span id="L13"><span class="lineNum"> 13</span> : {</span>
|
|
<span id="L14"><span class="lineNum"> 14</span> <span class="tlaGNC"> 1896 : for (auto& [key, value] : discretizers) {</span></span>
|
|
<span id="L15"><span class="lineNum"> 15</span> <span class="tlaGNC"> 1696 : delete value;</span></span>
|
|
<span id="L16"><span class="lineNum"> 16</span> : }</span>
|
|
<span id="L17"><span class="lineNum"> 17</span> <span class="tlaGNC"> 200 : }</span></span>
|
|
<span id="L18"><span class="lineNum"> 18</span> <span class="tlaGNC"> 228 : void Proposal::checkInput(const torch::Tensor& X, const torch::Tensor& y)</span></span>
|
|
<span id="L19"><span class="lineNum"> 19</span> : {</span>
|
|
<span id="L20"><span class="lineNum"> 20</span> <span class="tlaGNC"> 228 : if (!torch::is_floating_point(X)) {</span></span>
|
|
<span id="L21"><span class="lineNum"> 21</span> <span class="tlaUNC tlaBgUNC"> 0 : throw std::invalid_argument("X must be a floating point tensor");</span></span>
|
|
<span id="L22"><span class="lineNum"> 22</span> : }</span>
|
|
<span id="L23"><span class="lineNum"> 23</span> <span class="tlaGNC tlaBgGNC"> 228 : if (torch::is_floating_point(y)) {</span></span>
|
|
<span id="L24"><span class="lineNum"> 24</span> <span class="tlaUNC tlaBgUNC"> 0 : throw std::invalid_argument("y must be an integer tensor");</span></span>
|
|
<span id="L25"><span class="lineNum"> 25</span> : }</span>
|
|
<span id="L26"><span class="lineNum"> 26</span> <span class="tlaGNC tlaBgGNC"> 228 : }</span></span>
|
|
<span id="L27"><span class="lineNum"> 27</span> <span class="tlaGNC"> 212 : map<std::string, std::vector<int>> Proposal::localDiscretizationProposal(const map<std::string, std::vector<int>>& oldStates, Network& model)</span></span>
|
|
<span id="L28"><span class="lineNum"> 28</span> : {</span>
|
|
<span id="L29"><span class="lineNum"> 29</span> : // order of local discretization is important. no good 0, 1, 2...</span>
|
|
<span id="L30"><span class="lineNum"> 30</span> : // although we rediscretize features after the local discretization of every feature</span>
|
|
<span id="L31"><span class="lineNum"> 31</span> <span class="tlaGNC"> 212 : auto order = model.topological_sort();</span></span>
|
|
<span id="L32"><span class="lineNum"> 32</span> <span class="tlaGNC"> 212 : auto& nodes = model.getNodes();</span></span>
|
|
<span id="L33"><span class="lineNum"> 33</span> <span class="tlaGNC"> 212 : map<std::string, std::vector<int>> states = oldStates;</span></span>
|
|
<span id="L34"><span class="lineNum"> 34</span> <span class="tlaGNC"> 212 : std::vector<int> indicesToReDiscretize;</span></span>
|
|
<span id="L35"><span class="lineNum"> 35</span> <span class="tlaGNC"> 212 : bool upgrade = false; // Flag to check if we need to upgrade the model</span></span>
|
|
<span id="L36"><span class="lineNum"> 36</span> <span class="tlaGNC"> 1776 : for (auto feature : order) {</span></span>
|
|
<span id="L37"><span class="lineNum"> 37</span> <span class="tlaGNC"> 1564 : auto nodeParents = nodes[feature]->getParents();</span></span>
|
|
<span id="L38"><span class="lineNum"> 38</span> <span class="tlaGNC"> 1564 : if (nodeParents.size() < 2) continue; // Only has class as parent</span></span>
|
|
<span id="L39"><span class="lineNum"> 39</span> <span class="tlaGNC"> 1324 : upgrade = true;</span></span>
|
|
<span id="L40"><span class="lineNum"> 40</span> <span class="tlaGNC"> 1324 : int index = find(pFeatures.begin(), pFeatures.end(), feature) - pFeatures.begin();</span></span>
|
|
<span id="L41"><span class="lineNum"> 41</span> <span class="tlaGNC"> 1324 : indicesToReDiscretize.push_back(index); // We need to re-discretize this feature</span></span>
|
|
<span id="L42"><span class="lineNum"> 42</span> <span class="tlaGNC"> 1324 : std::vector<std::string> parents;</span></span>
|
|
<span id="L43"><span class="lineNum"> 43</span> <span class="tlaGNC"> 4020 : transform(nodeParents.begin(), nodeParents.end(), back_inserter(parents), [](const auto& p) { return p->getName(); });</span></span>
|
|
<span id="L44"><span class="lineNum"> 44</span> : // Remove class as parent as it will be added later</span>
|
|
<span id="L45"><span class="lineNum"> 45</span> <span class="tlaGNC"> 1324 : parents.erase(remove(parents.begin(), parents.end(), pClassName), parents.end());</span></span>
|
|
<span id="L46"><span class="lineNum"> 46</span> : // Get the indices of the parents</span>
|
|
<span id="L47"><span class="lineNum"> 47</span> <span class="tlaGNC"> 1324 : std::vector<int> indices;</span></span>
|
|
<span id="L48"><span class="lineNum"> 48</span> <span class="tlaGNC"> 1324 : indices.push_back(-1); // Add class index</span></span>
|
|
<span id="L49"><span class="lineNum"> 49</span> <span class="tlaGNC"> 2696 : transform(parents.begin(), parents.end(), back_inserter(indices), [&](const auto& p) {return find(pFeatures.begin(), pFeatures.end(), p) - pFeatures.begin(); });</span></span>
|
|
<span id="L50"><span class="lineNum"> 50</span> : // Now we fit the discretizer of the feature, conditioned on its parents and the class i.e. discretizer.fit(X[index], X[indices] + y)</span>
|
|
<span id="L51"><span class="lineNum"> 51</span> <span class="tlaGNC"> 1324 : std::vector<std::string> yJoinParents(Xf.size(1));</span></span>
|
|
<span id="L52"><span class="lineNum"> 52</span> <span class="tlaGNC"> 4020 : for (auto idx : indices) {</span></span>
|
|
<span id="L53"><span class="lineNum"> 53</span> <span class="tlaGNC"> 958640 : for (int i = 0; i < Xf.size(1); ++i) {</span></span>
|
|
<span id="L54"><span class="lineNum"> 54</span> <span class="tlaGNC"> 2867832 : yJoinParents[i] += to_string(pDataset.index({ idx, i }).item<int>());</span></span>
|
|
<span id="L55"><span class="lineNum"> 55</span> : }</span>
|
|
<span id="L56"><span class="lineNum"> 56</span> : }</span>
|
|
<span id="L57"><span class="lineNum"> 57</span> <span class="tlaGNC"> 1324 : auto arff = ArffFiles();</span></span>
|
|
<span id="L58"><span class="lineNum"> 58</span> <span class="tlaGNC"> 1324 : auto yxv = arff.factorize(yJoinParents);</span></span>
|
|
<span id="L59"><span class="lineNum"> 59</span> <span class="tlaGNC"> 2648 : auto xvf_ptr = Xf.index({ index }).data_ptr<float>();</span></span>
|
|
<span id="L60"><span class="lineNum"> 60</span> <span class="tlaGNC"> 1324 : auto xvf = std::vector<mdlp::precision_t>(xvf_ptr, xvf_ptr + Xf.size(1));</span></span>
|
|
<span id="L61"><span class="lineNum"> 61</span> <span class="tlaGNC"> 1324 : discretizers[feature]->fit(xvf, yxv);</span></span>
|
|
<span id="L62"><span class="lineNum"> 62</span> <span class="tlaGNC"> 1804 : }</span></span>
|
|
<span id="L63"><span class="lineNum"> 63</span> <span class="tlaGNC"> 212 : if (upgrade) {</span></span>
|
|
<span id="L64"><span class="lineNum"> 64</span> : // Discretize again X (only the affected indices) with the new fitted discretizers</span>
|
|
<span id="L65"><span class="lineNum"> 65</span> <span class="tlaGNC"> 1536 : for (auto index : indicesToReDiscretize) {</span></span>
|
|
<span id="L66"><span class="lineNum"> 66</span> <span class="tlaGNC"> 2648 : auto Xt_ptr = Xf.index({ index }).data_ptr<float>();</span></span>
|
|
<span id="L67"><span class="lineNum"> 67</span> <span class="tlaGNC"> 1324 : auto Xt = std::vector<float>(Xt_ptr, Xt_ptr + Xf.size(1));</span></span>
|
|
<span id="L68"><span class="lineNum"> 68</span> <span class="tlaGNC"> 5296 : pDataset.index_put_({ index, "..." }, torch::tensor(discretizers[pFeatures[index]]->transform(Xt)));</span></span>
|
|
<span id="L69"><span class="lineNum"> 69</span> <span class="tlaGNC"> 1324 : auto xStates = std::vector<int>(discretizers[pFeatures[index]]->getCutPoints().size() + 1);</span></span>
|
|
<span id="L70"><span class="lineNum"> 70</span> <span class="tlaGNC"> 1324 : iota(xStates.begin(), xStates.end(), 0);</span></span>
|
|
<span id="L71"><span class="lineNum"> 71</span> : //Update new states of the feature/node</span>
|
|
<span id="L72"><span class="lineNum"> 72</span> <span class="tlaGNC"> 1324 : states[pFeatures[index]] = xStates;</span></span>
|
|
<span id="L73"><span class="lineNum"> 73</span> <span class="tlaGNC"> 1324 : }</span></span>
|
|
<span id="L74"><span class="lineNum"> 74</span> <span class="tlaGNC"> 212 : const torch::Tensor weights = torch::full({ pDataset.size(1) }, 1.0 / pDataset.size(1), torch::kDouble);</span></span>
|
|
<span id="L75"><span class="lineNum"> 75</span> <span class="tlaGNC"> 212 : model.fit(pDataset, weights, pFeatures, pClassName, states);</span></span>
|
|
<span id="L76"><span class="lineNum"> 76</span> <span class="tlaGNC"> 212 : }</span></span>
|
|
<span id="L77"><span class="lineNum"> 77</span> <span class="tlaGNC"> 424 : return states;</span></span>
|
|
<span id="L78"><span class="lineNum"> 78</span> <span class="tlaGNC"> 960128 : }</span></span>
|
|
<span id="L79"><span class="lineNum"> 79</span> <span class="tlaGNC"> 232 : map<std::string, std::vector<int>> Proposal::fit_local_discretization(const torch::Tensor& y)</span></span>
|
|
<span id="L80"><span class="lineNum"> 80</span> : {</span>
|
|
<span id="L81"><span class="lineNum"> 81</span> : // Discretize the continuous input data and build pDataset (Classifier::dataset)</span>
|
|
<span id="L82"><span class="lineNum"> 82</span> <span class="tlaGNC"> 232 : int m = Xf.size(1);</span></span>
|
|
<span id="L83"><span class="lineNum"> 83</span> <span class="tlaGNC"> 232 : int n = Xf.size(0);</span></span>
|
|
<span id="L84"><span class="lineNum"> 84</span> <span class="tlaGNC"> 232 : map<std::string, std::vector<int>> states;</span></span>
|
|
<span id="L85"><span class="lineNum"> 85</span> <span class="tlaGNC"> 232 : pDataset = torch::zeros({ n + 1, m }, torch::kInt32);</span></span>
|
|
<span id="L86"><span class="lineNum"> 86</span> <span class="tlaGNC"> 232 : auto yv = std::vector<int>(y.data_ptr<int>(), y.data_ptr<int>() + y.size(0));</span></span>
|
|
<span id="L87"><span class="lineNum"> 87</span> : // discretize input data by feature(row)</span>
|
|
<span id="L88"><span class="lineNum"> 88</span> <span class="tlaGNC"> 1944 : for (auto i = 0; i < pFeatures.size(); ++i) {</span></span>
|
|
<span id="L89"><span class="lineNum"> 89</span> <span class="tlaGNC"> 1712 : auto* discretizer = new mdlp::CPPFImdlp();</span></span>
|
|
<span id="L90"><span class="lineNum"> 90</span> <span class="tlaGNC"> 3424 : auto Xt_ptr = Xf.index({ i }).data_ptr<float>();</span></span>
|
|
<span id="L91"><span class="lineNum"> 91</span> <span class="tlaGNC"> 1712 : auto Xt = std::vector<float>(Xt_ptr, Xt_ptr + Xf.size(1));</span></span>
|
|
<span id="L92"><span class="lineNum"> 92</span> <span class="tlaGNC"> 1712 : discretizer->fit(Xt, yv);</span></span>
|
|
<span id="L93"><span class="lineNum"> 93</span> <span class="tlaGNC"> 6848 : pDataset.index_put_({ i, "..." }, torch::tensor(discretizer->transform(Xt)));</span></span>
|
|
<span id="L94"><span class="lineNum"> 94</span> <span class="tlaGNC"> 1712 : auto xStates = std::vector<int>(discretizer->getCutPoints().size() + 1);</span></span>
|
|
<span id="L95"><span class="lineNum"> 95</span> <span class="tlaGNC"> 1712 : iota(xStates.begin(), xStates.end(), 0);</span></span>
|
|
<span id="L96"><span class="lineNum"> 96</span> <span class="tlaGNC"> 1712 : states[pFeatures[i]] = xStates;</span></span>
|
|
<span id="L97"><span class="lineNum"> 97</span> <span class="tlaGNC"> 1712 : discretizers[pFeatures[i]] = discretizer;</span></span>
|
|
<span id="L98"><span class="lineNum"> 98</span> <span class="tlaGNC"> 1712 : }</span></span>
|
|
<span id="L99"><span class="lineNum"> 99</span> <span class="tlaGNC"> 232 : int n_classes = torch::max(y).item<int>() + 1;</span></span>
|
|
<span id="L100"><span class="lineNum"> 100</span> <span class="tlaGNC"> 232 : auto yStates = std::vector<int>(n_classes);</span></span>
|
|
<span id="L101"><span class="lineNum"> 101</span> <span class="tlaGNC"> 232 : iota(yStates.begin(), yStates.end(), 0);</span></span>
|
|
<span id="L102"><span class="lineNum"> 102</span> <span class="tlaGNC"> 232 : states[pClassName] = yStates;</span></span>
|
|
<span id="L103"><span class="lineNum"> 103</span> <span class="tlaGNC"> 696 : pDataset.index_put_({ n, "..." }, y);</span></span>
|
|
<span id="L104"><span class="lineNum"> 104</span> <span class="tlaGNC"> 464 : return states;</span></span>
|
|
<span id="L105"><span class="lineNum"> 105</span> <span class="tlaGNC"> 3888 : }</span></span>
|
|
<span id="L106"><span class="lineNum"> 106</span> <span class="tlaGNC"> 168 : torch::Tensor Proposal::prepareX(torch::Tensor& X)</span></span>
|
|
<span id="L107"><span class="lineNum"> 107</span> : {</span>
|
|
<span id="L108"><span class="lineNum"> 108</span> <span class="tlaGNC"> 168 : auto Xtd = torch::zeros_like(X, torch::kInt32);</span></span>
|
|
<span id="L109"><span class="lineNum"> 109</span> <span class="tlaGNC"> 1376 : for (int i = 0; i < X.size(0); ++i) {</span></span>
|
|
<span id="L110"><span class="lineNum"> 110</span> <span class="tlaGNC"> 1208 : auto Xt = std::vector<float>(X[i].data_ptr<float>(), X[i].data_ptr<float>() + X.size(1));</span></span>
|
|
<span id="L111"><span class="lineNum"> 111</span> <span class="tlaGNC"> 1208 : auto Xd = discretizers[pFeatures[i]]->transform(Xt);</span></span>
|
|
<span id="L112"><span class="lineNum"> 112</span> <span class="tlaGNC"> 3624 : Xtd.index_put_({ i }, torch::tensor(Xd, torch::kInt32));</span></span>
|
|
<span id="L113"><span class="lineNum"> 113</span> <span class="tlaGNC"> 1208 : }</span></span>
|
|
<span id="L114"><span class="lineNum"> 114</span> <span class="tlaGNC"> 336 : return Xtd;</span></span>
|
|
<span id="L115"><span class="lineNum"> 115</span> <span class="tlaGNC"> 1376 : }</span></span>
|
|
<span id="L116"><span class="lineNum"> 116</span> : }</span>
|
|
</pre>
|
|
</td>
|
|
</tr>
|
|
</table>
|
|
<br>
|
|
|
|
<table width="100%" border=0 cellspacing=0 cellpadding=0>
|
|
<tr><td class="ruler"><img src="../../glass.png" width=3 height=3 alt=""></td></tr>
|
|
<tr><td class="versionInfo">Generated by: <a href="https://github.com//linux-test-project/lcov" target="_parent">LCOV version 2.0-1</a></td></tr>
|
|
</table>
|
|
<br>
|
|
|
|
</body>
|
|
</html>
|