1/*M///////////////////////////////////////////////////////////////////////////////////////
2//
3//  IMPORTANT: READ BEFORE DOWNLOADING, COPYING, INSTALLING OR USING.
4//
5//  By downloading, copying, installing or using the software you agree to this license.
6//  If you do not agree to this license, do not download, install,
7//  copy or use the software.
8//
9//
10//                        Intel License Agreement
11//                For Open Source Computer Vision Library
12//
13// Copyright (C) 2000, Intel Corporation, all rights reserved.
14// Third party copyrights are property of their respective owners.
15//
16// Redistribution and use in source and binary forms, with or without modification,
17// are permitted provided that the following conditions are met:
18//
19//   * Redistribution's of source code must retain the above copyright notice,
20//     this list of conditions and the following disclaimer.
21//
22//   * Redistribution's in binary form must reproduce the above copyright notice,
23//     this list of conditions and the following disclaimer in the documentation
24//     and/or other materials provided with the distribution.
25//
26//   * The name of Intel Corporation may not be used to endorse or promote products
27//     derived from this software without specific prior written permission.
28//
29// This software is provided by the copyright holders and contributors "as is" and
30// any express or implied warranties, including, but not limited to, the implied
31// warranties of merchantability and fitness for a particular purpose are disclaimed.
32// In no event shall the Intel Corporation or contributors be liable for any direct,
33// indirect, incidental, special, exemplary, or consequential damages
34// (including, but not limited to, procurement of substitute goods or services;
35// loss of use, data, or profits; or business interruption) however caused
36// and on any theory of liability, whether in contract, strict liability,
37// or tort (including negligence or otherwise) arising in any way out of
38// the use of this software, even if advised of the possibility of such damage.
39//
40//M*/
41
42#include "test_precomp.hpp"
43
44#include <iostream>
45#include <fstream>
46
47using namespace cv;
48using namespace std;
49
50CV_SLMLTest::CV_SLMLTest( const char* _modelName ) : CV_MLBaseTest( _modelName )
51{
52    validationFN = "slvalidation.xml";
53}
54
55int CV_SLMLTest::run_test_case( int testCaseIdx )
56{
57    int code = cvtest::TS::OK;
58    code = prepare_test_case( testCaseIdx );
59
60    if( code == cvtest::TS::OK )
61    {
62        data->setTrainTestSplit(data->getNTrainSamples(), true);
63        code = train( testCaseIdx );
64        if( code == cvtest::TS::OK )
65        {
66            get_test_error( testCaseIdx, &test_resps1 );
67            fname1 = tempfile(".yml.gz");
68            save( fname1.c_str() );
69            load( fname1.c_str() );
70            get_test_error( testCaseIdx, &test_resps2 );
71            fname2 = tempfile(".yml.gz");
72            save( fname2.c_str() );
73        }
74        else
75            ts->printf( cvtest::TS::LOG, "model can not be trained" );
76    }
77    return code;
78}
79
80int CV_SLMLTest::validate_test_results( int testCaseIdx )
81{
82    int code = cvtest::TS::OK;
83
84    // 1. compare files
85    FILE *fs1 = fopen(fname1.c_str(), "rb"), *fs2 = fopen(fname2.c_str(), "rb");
86    size_t sz1 = 0, sz2 = 0;
87    if( !fs1 || !fs2 )
88        code = cvtest::TS::FAIL_MISSING_TEST_DATA;
89    if( code >= 0 )
90    {
91        fseek(fs1, 0, SEEK_END); fseek(fs2, 0, SEEK_END);
92        sz1 = ftell(fs1);
93        sz2 = ftell(fs2);
94        fseek(fs1, 0, SEEK_SET); fseek(fs2, 0, SEEK_SET);
95    }
96
97    if( sz1 != sz2 )
98        code = cvtest::TS::FAIL_INVALID_OUTPUT;
99
100    if( code >= 0 )
101    {
102        const int BUFSZ = 1024;
103        uchar buf1[BUFSZ], buf2[BUFSZ];
104        for( size_t pos = 0; pos < sz1;  )
105        {
106            size_t r1 = fread(buf1, 1, BUFSZ, fs1);
107            size_t r2 = fread(buf2, 1, BUFSZ, fs2);
108            if( r1 != r2 || memcmp(buf1, buf2, r1) != 0 )
109            {
110                ts->printf( cvtest::TS::LOG,
111                           "in test case %d first (%s) and second (%s) saved files differ in %d-th kb\n",
112                           testCaseIdx, fname1.c_str(), fname2.c_str(),
113                           (int)pos );
114                code = cvtest::TS::FAIL_INVALID_OUTPUT;
115                break;
116            }
117            pos += r1;
118        }
119    }
120
121    if(fs1)
122        fclose(fs1);
123    if(fs2)
124        fclose(fs2);
125
126    // delete temporary files
127    if( code >= 0 )
128    {
129        remove( fname1.c_str() );
130        remove( fname2.c_str() );
131    }
132
133    if( code >= 0 )
134    {
135        // 2. compare responses
136        CV_Assert( test_resps1.size() == test_resps2.size() );
137        vector<float>::const_iterator it1 = test_resps1.begin(), it2 = test_resps2.begin();
138        for( ; it1 != test_resps1.end(); ++it1, ++it2 )
139        {
140            if( fabs(*it1 - *it2) > FLT_EPSILON )
141            {
142                ts->printf( cvtest::TS::LOG, "in test case %d responses predicted before saving and after loading is different", testCaseIdx );
143                code = cvtest::TS::FAIL_INVALID_OUTPUT;
144                break;
145            }
146        }
147    }
148    return code;
149}
150
151TEST(ML_NaiveBayes, save_load) { CV_SLMLTest test( CV_NBAYES ); test.safe_run(); }
152TEST(ML_KNearest, save_load) { CV_SLMLTest test( CV_KNEAREST ); test.safe_run(); }
153TEST(ML_SVM, save_load) { CV_SLMLTest test( CV_SVM ); test.safe_run(); }
154TEST(ML_ANN, save_load) { CV_SLMLTest test( CV_ANN ); test.safe_run(); }
155TEST(ML_DTree, save_load) { CV_SLMLTest test( CV_DTREE ); test.safe_run(); }
156TEST(ML_Boost, save_load) { CV_SLMLTest test( CV_BOOST ); test.safe_run(); }
157TEST(ML_RTrees, save_load) { CV_SLMLTest test( CV_RTREES ); test.safe_run(); }
158TEST(DISABLED_ML_ERTrees, save_load) { CV_SLMLTest test( CV_ERTREES ); test.safe_run(); }
159
160class CV_LegacyTest : public cvtest::BaseTest
161{
162public:
163    CV_LegacyTest(const std::string &_modelName, const std::string &_suffixes = std::string())
164        : cvtest::BaseTest(), modelName(_modelName), suffixes(_suffixes)
165    {
166    }
167    virtual ~CV_LegacyTest() {}
168protected:
169    void run(int)
170    {
171        unsigned int idx = 0;
172        for (;;)
173        {
174            if (idx >= suffixes.size())
175                break;
176            int found = (int)suffixes.find(';', idx);
177            string piece = suffixes.substr(idx, found - idx);
178            if (piece.empty())
179                break;
180            oneTest(piece);
181            idx += (unsigned int)piece.size() + 1;
182        }
183    }
184    void oneTest(const string & suffix)
185    {
186        using namespace cv::ml;
187
188        int code = cvtest::TS::OK;
189        string filename = ts->get_data_path() + "legacy/" + modelName + suffix;
190        bool isTree = modelName == CV_BOOST || modelName == CV_DTREE || modelName == CV_RTREES;
191        Ptr<StatModel> model;
192        if (modelName == CV_BOOST)
193            model = Algorithm::load<Boost>(filename);
194        else if (modelName == CV_ANN)
195            model = Algorithm::load<ANN_MLP>(filename);
196        else if (modelName == CV_DTREE)
197            model = Algorithm::load<DTrees>(filename);
198        else if (modelName == CV_NBAYES)
199            model = Algorithm::load<NormalBayesClassifier>(filename);
200        else if (modelName == CV_SVM)
201            model = Algorithm::load<SVM>(filename);
202        else if (modelName == CV_RTREES)
203            model = Algorithm::load<RTrees>(filename);
204        if (!model)
205        {
206            code = cvtest::TS::FAIL_INVALID_TEST_DATA;
207        }
208        else
209        {
210            Mat input = Mat(isTree ? 10 : 1, model->getVarCount(), CV_32F);
211            ts->get_rng().fill(input, RNG::UNIFORM, 0, 40);
212
213            if (isTree)
214                randomFillCategories(filename, input);
215
216            Mat output;
217            model->predict(input, output, StatModel::RAW_OUTPUT | (isTree ? DTrees::PREDICT_SUM : 0));
218            // just check if no internal assertions or errors thrown
219        }
220        ts->set_failed_test_info(code);
221    }
222    void randomFillCategories(const string & filename, Mat & input)
223    {
224        Mat catMap;
225        Mat catCount;
226        std::vector<uchar> varTypes;
227
228        FileStorage fs(filename, FileStorage::READ);
229        FileNode root = fs.getFirstTopLevelNode();
230        root["cat_map"] >> catMap;
231        root["cat_count"] >> catCount;
232        root["var_type"] >> varTypes;
233
234        int offset = 0;
235        int countOffset = 0;
236        uint var = 0, varCount = (uint)varTypes.size();
237        for (; var < varCount; ++var)
238        {
239            if (varTypes[var] == ml::VAR_CATEGORICAL)
240            {
241                int size = catCount.at<int>(0, countOffset);
242                for (int row = 0; row < input.rows; ++row)
243                {
244                    int randomChosenIndex = offset + ((uint)ts->get_rng()) % size;
245                    int value = catMap.at<int>(0, randomChosenIndex);
246                    input.at<float>(row, var) = (float)value;
247                }
248                offset += size;
249                ++countOffset;
250            }
251        }
252    }
253    string modelName;
254    string suffixes;
255};
256
257TEST(ML_ANN, legacy_load) { CV_LegacyTest test(CV_ANN, "_waveform.xml"); test.safe_run(); }
258TEST(ML_Boost, legacy_load) { CV_LegacyTest test(CV_BOOST, "_adult.xml;_1.xml;_2.xml;_3.xml"); test.safe_run(); }
259TEST(ML_DTree, legacy_load) { CV_LegacyTest test(CV_DTREE, "_abalone.xml;_mushroom.xml"); test.safe_run(); }
260TEST(ML_NBayes, legacy_load) { CV_LegacyTest test(CV_NBAYES, "_waveform.xml"); test.safe_run(); }
261TEST(ML_SVM, legacy_load) { CV_LegacyTest test(CV_SVM, "_poletelecomm.xml;_waveform.xml"); test.safe_run(); }
262TEST(ML_RTrees, legacy_load) { CV_LegacyTest test(CV_RTREES, "_waveform.xml"); test.safe_run(); }
263
264/*TEST(ML_SVM, throw_exception_when_save_untrained_model)
265{
266    Ptr<cv::ml::SVM> svm;
267    string filename = tempfile("svm.xml");
268    ASSERT_THROW(svm.save(filename.c_str()), Exception);
269    remove(filename.c_str());
270}*/
271
272TEST(DISABLED_ML_SVM, linear_save_load)
273{
274    Ptr<cv::ml::SVM> svm1, svm2, svm3;
275
276    svm1 = Algorithm::load<SVM>("SVM45_X_38-1.xml");
277    svm2 = Algorithm::load<SVM>("SVM45_X_38-2.xml");
278    string tname = tempfile("a.xml");
279    svm2->save(tname);
280    svm3 = Algorithm::load<SVM>(tname);
281
282    ASSERT_EQ(svm1->getVarCount(), svm2->getVarCount());
283    ASSERT_EQ(svm1->getVarCount(), svm3->getVarCount());
284
285    int m = 10000, n = svm1->getVarCount();
286    Mat samples(m, n, CV_32F), r1, r2, r3;
287    randu(samples, 0., 1.);
288
289    svm1->predict(samples, r1);
290    svm2->predict(samples, r2);
291    svm3->predict(samples, r3);
292
293    double eps = 1e-4;
294    EXPECT_LE(norm(r1, r2, NORM_INF), eps);
295    EXPECT_LE(norm(r1, r3, NORM_INF), eps);
296
297    remove(tname.c_str());
298}
299
300/* End of file. */
301