update datasets used in the tests.

.coveragerc

@@ -0,0 +1,2 @@
+[run]
+omit = gklearn/tests/*

.gitignore

@@ -15,6 +15,7 @@ datasets/*
 !datasets/AIDS/
 !datasets/monoterpenoides/
 !datasets/Fingerprint/*.txt
+!datasets/Cuneiform/*.txt
 notebooks/results/*
 notebooks/check_gm/*
 notebooks/test_parallel/*
@@ -41,3 +42,4 @@ dist/
 build/
 
 .coverage
+htmlcov

.travis.yml

@@ -22,7 +22,7 @@ install:
 
 script:
 - python setup.py bdist_wheel
-- pytest -v --cov-report term --cov=gklearn gklearn/tests/
+- pytest -v --cov-config=.coveragerc --cov-report term --cov=gklearn gklearn/tests/
 
 after_success:
 - codecov

datasets/Cuneiform/Cuneiform_graph_labels.txt

@@ -0,0 +1,267 @@
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+11
+12
+13
+14
+15
+16
+17
+18
+19
+20
+21
+22
+23
+24
+25
+26
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+11
+12
+13
+14
+15
+16
+17
+18
+19
+20
+21
+22
+23
+24
+25
+26
+27
+28
+29
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+11
+12
+13
+14
+15
+16
+17
+18
+19
+20
+21
+22
+23
+24
+25
+26
+27
+28
+29
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+11
+12
+13
+14
+15
+16
+17
+18
+19
+20
+21
+22
+23
+24
+25
+26
+27
+28
+29
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+11
+12
+13
+14
+15
+16
+17
+18
+19
+20
+21
+22
+23
+24
+25
+26
+27
+28
+29
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+11
+12
+13
+14
+15
+16
+17
+18
+19
+20
+21
+22
+23
+24
+25
+26
+27
+28
+29
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+11
+12
+13
+14
+15
+16
+17
+18
+19
+20
+21
+22
+23
+24
+25
+26
+27
+28
+29
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+11
+12
+13
+14
+15
+16
+17
+18
+19
+20
+21
+22
+23
+24
+25
+26
+27
+28
+29
+0
+1
+2
+3
+4
+5
+6
+7
+8
+9
+10
+11
+12
+13
+14
+15
+16
+17
+18
+19
+20
+21
+22
+23
+24
+25
+26
+27
+28
+29

datasets/Cuneiform/README.txt

@@ -0,0 +1,119 @@
+README for dataset Cuneiform
+
+
+=== Usage ===
+
+This folder contains the following comma separated text files 
+(replace DS by the name of the dataset):
+
+n = total number of nodes
+m = total number of edges
+N = number of graphs
+
+(1) 	DS_A.txt (m lines) 
+	sparse (block diagonal) adjacency matrix for all graphs,
+	each line corresponds to (row, col) resp. (node_id, node_id)
+
+(2) 	DS_graph_indicator.txt (n lines)
+	column vector of graph identifiers for all nodes of all graphs,
+	the value in the i-th line is the graph_id of the node with node_id i
+
+(3) 	DS_graph_labels.txt (N lines) 
+	class labels for all graphs in the dataset,
+	the value in the i-th line is the class label of the graph with graph_id i
+
+(4) 	DS_node_labels.txt (n lines)
+	column vector of node labels,
+	the value in the i-th line corresponds to the node with node_id i
+
+There are OPTIONAL files if the respective information is available:
+
+(5) 	DS_edge_labels.txt (m lines; same size as DS_A_sparse.txt)
+	labels for the edges in DS_A_sparse.txt 
+
+(6) 	DS_edge_attributes.txt (m lines; same size as DS_A.txt)
+	attributes for the edges in DS_A.txt 
+
+(7) 	DS_node_attributes.txt (n lines) 
+	matrix of node attributes,
+	the comma seperated values in the i-th line is the attribute vector of the node with node_id i
+
+(8) 	DS_graph_attributes.txt (N lines) 
+	regression values for all graphs in the dataset,
+	the value in the i-th line is the attribute of the graph with graph_id i
+
+
+=== Description ===
+
+The Cuneiform dataset contains graphs representing 29 different Hittite cuneiform signs.
+The data was obtained from nine cuneiform tablets written by scholars of Hittitology in
+the course of a study about individualistic characteristics of cuneiform hand writing.
+After automated extraction of individual wedges, the affiliation of the wedges to the 
+cuneiform signs were determined manually. The graph model is explained in detail in the
+referenced publication.
+
+
+=== References ===
+
+Nils M. Kriege, Matthias Fey, Denis Fisseler, Petra Mutzel, Frank Weichert
+Recognizing Cuneiform Signs Using Graph Based Methods. 2018. arXiv:1802.05908
+https://arxiv.org/abs/1802.05908
+
+
+=== Description of Labels ===
+
+Node labels were converted to integer values using this map:
+
+Component 0:
+	0	depthPoint
+	1	tailVertex
+	2	leftVertex
+	3	rightVertex
+
+Component 1:
+	0	vertical
+	1	Winkelhaken
+	2	horizontal
+
+
+
+Edge labels were converted to integer values using this map:
+
+Component 0:
+	0	wedge
+	1	arrangement
+
+
+
+Class labels were converted to integer values using this map:
+
+	0	tu
+	1	ta
+	2	ti
+	3	nu
+	4	na
+	5	ni
+	6	bu
+	7	ba
+	8	bi
+	9	zu
+	10	za
+	11	zi
+	12	su
+	13	sa
+	14	si
+	15	hu
+	16	ha
+	17	hi
+	18	du
+	19	da
+	20	di
+	21	ru
+	22	ra
+	23	ri
+	24	ku
+	25	ka
+	26	ki
+	27	lu
+	28	la
+	29	li

gklearn/kernels/commonWalkKernel.py

@@ -16,7 +16,6 @@ from functools import partial
 import networkx as nx
 import numpy as np
 
-sys.path.insert(0, "../")
 from gklearn.utils.utils import direct_product
 from gklearn.utils.graphdataset import get_dataset_attributes
 from gklearn.utils.parallel import parallel_gm

gklearn/kernels/marginalizedKernel.py

@@ -28,7 +28,6 @@ from gklearn.utils.kernels import deltakernel
 from gklearn.utils.utils import untotterTransformation
 from gklearn.utils.graphdataset import get_dataset_attributes
 from gklearn.utils.parallel import parallel_gm
-sys.path.insert(0, "../")
 
 
 def marginalizedkernel(*args,

gklearn/kernels/randomWalkKernel.py

@@ -6,8 +6,6 @@
     [1] S Vichy N Vishwanathan, Nicol N Schraudolph, Risi Kondor, and Karsten M Borgwardt. Graph kernels. Journal of Machine Learning Research, 11(Apr):1201–1242, 2010.
 """
 
-import sys
-sys.path.insert(0, "../")
 import time
 from functools import partial
 from tqdm import tqdm

gklearn/kernels/spKernel.py

@@ -20,7 +20,6 @@ import numpy as np
 from gklearn.utils.utils import getSPGraph
 from gklearn.utils.graphdataset import get_dataset_attributes
 from gklearn.utils.parallel import parallel_gm
-sys.path.insert(0, "../")
 
 def spkernel(*args,
              node_label='atom',

gklearn/kernels/structuralspKernel.py

@@ -25,8 +25,6 @@ from gklearn.utils.graphdataset import get_dataset_attributes
 from gklearn.utils.parallel import parallel_gm
 from gklearn.utils.trie import Trie
 
-sys.path.insert(0, "../")
-
 
 def structuralspkernel(*args,
                        node_label='atom',

gklearn/kernels/treeletKernel.py

@@ -8,7 +8,6 @@
 """
 
 import sys
-sys.path.insert(0, "../")
 import time
 from collections import Counter
 from itertools import chain

gklearn/kernels/untilHPathKernel.py

@@ -9,7 +9,6 @@
 """
 
 import sys
-sys.path.insert(0, "../")
 import time
 from collections import Counter
 from itertools import chain

gklearn/kernels/weisfeilerLehmanKernel.py

@@ -10,7 +10,6 @@
 
 import sys
 from collections import Counter
-sys.path.insert(0, "../")
 from functools import partial
 import time
 #from multiprocessing import Pool

gklearn/preimage/find_best_k.py

@@ -9,10 +9,8 @@ import numpy as np
 import random
 import csv
 
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphfiles import loadDataset
-from preimage.test_k_closest_graphs import median_on_k_closest_graphs
+from gklearn.preimage.test_k_closest_graphs import median_on_k_closest_graphs
 
 def find_best_k():
     ds = {'name': 'monoterpenoides', 

gklearn/preimage/fitDistance.py

@@ -13,15 +13,14 @@ from multiprocessing import Pool
 from functools import partial
 import time
 import random
+import sys
 
 from scipy import optimize
 from scipy.optimize import minimize
 import cvxpy as cp
 
-import sys
-sys.path.insert(0, "../")
-from preimage.ged import GED, get_nb_edit_operations, get_nb_edit_operations_letter, get_nb_edit_operations_nonsymbolic
-from preimage.utils import kernel_distance_matrix
+from gklearn.preimage.ged import GED, get_nb_edit_operations, get_nb_edit_operations_letter, get_nb_edit_operations_nonsymbolic
+from gklearn.preimage.utils import kernel_distance_matrix
 
 def fit_GED_to_kernel_distance(Gn, node_label, edge_label, gkernel, itr_max,
                                params_ged={'lib': 'gedlibpy', 'cost': 'CONSTANT', 

gklearn/preimage/ged.py

@@ -128,12 +128,10 @@ def GED(g1, g2, dataset='monoterpenoides', lib='gedlibpy', cost='CHEM_1', method
     elif lib == 'gedlib-bash':
         import time
         import random
-        import sys
         import os
-        sys.path.insert(0, "../")
         from gklearn.utils.graphfiles import saveDataset
         
-        tmp_dir = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/output/tmp_ged/'
+        tmp_dir = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/output/tmp_ged/'
         if not os.path.exists(tmp_dir):
             os.makedirs(tmp_dir)
         fn_collection = tmp_dir + 'collection.' + str(time.time()) + str(random.randint(0, 1e9))
@@ -144,7 +142,7 @@ def GED(g1, g2, dataset='monoterpenoides', lib='gedlibpy', cost='CHEM_1', method
         command = 'GEDLIB_HOME=\'/media/ljia/DATA/research-repo/codes/others/gedlib/gedlib2\'\n'
         command += 'LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$GEDLIB_HOME/lib\n'
         command += 'export LD_LIBRARY_PATH\n'
-        command += 'cd \'/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/bin\'\n'
+        command += 'cd \'' + os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/bin\'\n'
         command += './ged_for_python_bash monoterpenoides ' + fn_collection \
                 + ' \'' + algo_options + '\' '
         for ec in edit_cost_constant:

gklearn/preimage/iam.py

@@ -11,11 +11,9 @@ import random
 import networkx as nx
 from tqdm import tqdm
 
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphdataset import get_dataset_attributes
 from gklearn.utils.utils import graph_isIdentical, get_node_labels, get_edge_labels
-from ged import GED, ged_median
+from gklearn.preimage.ged import GED, ged_median
 
 
 def iam_upgraded(Gn_median, Gn_candidate, c_ei=3, c_er=3, c_es=1, ite_max=50, 
@@ -438,7 +436,7 @@ def iam_upgraded(Gn_median, Gn_candidate, c_ei=3, c_er=3, c_es=1, ite_max=50,
 
 def iam_bash(Gn_names, edit_cost_constant, cost='CONSTANT', initial_solutions=1,
              dataset='monoterpenoides',
-             graph_dir='/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/monoterpenoides/'):
+             graph_dir=''):
     """Compute the iam by c++ implementation (gedlib) through bash.
     """
     import os
@@ -462,18 +460,18 @@ def iam_bash(Gn_names, edit_cost_constant, cost='CONSTANT', initial_solutions=1,
             fgroup.write("\n</GraphCollection>")
             fgroup.close()
 
-    tmp_dir = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/output/tmp_ged/'
+    tmp_dir = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/output/tmp_ged/'
     fn_collection = tmp_dir + 'collection.' + str(time.time()) + str(random.randint(0, 1e9))
     createCollectionFile(Gn_names, ['dummy'] * len(Gn_names), fn_collection)
 #    fn_collection = tmp_dir + 'collection_for_debug'
-#    graph_dir = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/generated_datsets/monoterpenoides/gxl'
+#    graph_dir = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/generated_datsets/monoterpenoides/gxl'
     
 #    if dataset == 'Letter-high' or dataset == 'Fingerprint':
 #        dataset = 'letter'
     command = 'GEDLIB_HOME=\'/media/ljia/DATA/research-repo/codes/Linlin/gedlib\'\n'
     command += 'LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$GEDLIB_HOME/lib\n'
     command += 'export LD_LIBRARY_PATH\n'
-    command += 'cd \'/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/bin\'\n'
+    command += 'cd \'' + os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/bin\'\n'
     command += './iam_for_python_bash ' + dataset + ' ' + fn_collection \
             + ' \'' + graph_dir + '\' ' + ' ' + cost + ' ' + str(initial_solutions) + ' '
     if edit_cost_constant is None:
@@ -489,8 +487,8 @@ def iam_bash(Gn_names, edit_cost_constant, cost='CONSTANT', initial_solutions=1,
     sod_sm = float(output[0].strip())
     sod_gm = float(output[1].strip())
     
-    fname_sm = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/output/tmp_ged/set_median.gxl'
-    fname_gm = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/output/tmp_ged/gen_median.gxl'
+    fname_sm = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/output/tmp_ged/set_median.gxl'
+    fname_gm = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/output/tmp_ged/gen_median.gxl'
     
     return sod_sm, sod_gm, fname_sm, fname_gm
 

gklearn/preimage/knn.py

@@ -11,14 +11,12 @@ from tqdm import tqdm
 import random
 #import csv
 from shutil import copyfile
+import os
 
-
-import sys
-sys.path.insert(0, "../")
-from preimage.iam import iam_bash
+from gklearn.preimage.iam import iam_bash
 from gklearn.utils.graphfiles import loadDataset, loadGXL
-from preimage.ged import GED
-from preimage.utils import get_same_item_indices
+from gklearn.preimage.ged import GED
+from gklearn.preimage.utils import get_same_item_indices
 
 def test_knn():
     ds = {'name': 'monoterpenoides', 
@@ -30,7 +28,7 @@ def test_knn():
 #    edge_label = 'bond_type'
 #    ds_name = 'mono'
     dir_output = 'results/knn/'
-    graph_dir='/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/monoterpenoides/'
+    graph_dir = os.path.dirname(os.path.realpath(__file__)) + '../../datasets/monoterpenoides/'
     
     k_nn = 1
     percent = 0.1

gklearn/preimage/libs.py

@@ -2,5 +2,5 @@ import sys
 import pathlib
 
 # insert gedlibpy library.
-sys.path.insert(0, "../../")
+sys.path.insert(0, "../../../")
 from gedlibpy import librariesImport, gedlibpy

gklearn/preimage/preimage_random.py

@@ -14,10 +14,7 @@ from tqdm import tqdm
 import networkx as nx
 import matplotlib.pyplot as plt
 
-
-sys.path.insert(0, "../")
-
-from utils import compute_kernel, dis_gstar
+from gklearn.preimage.utils import compute_kernel, dis_gstar
 
 
 def preimage_random(Gn_init, Gn_median, alpha, idx_gi, Kmatrix, k, r_max, l, gkernel):

gklearn/preimage/test.py

@@ -52,8 +52,6 @@ def convertGraph(G):
 
 
 def testNxGrapĥ():
-    import sys
-    sys.path.insert(0, "../")
     from gklearn.utils.graphfiles import loadDataset
     ds = {'name': 'MUTAG', 'dataset': '../datasets/MUTAG/MUTAG_A.txt',
           'extra_params': {}}  # node/edge symb

gklearn/preimage/test_fitDistance.py

@@ -9,12 +9,10 @@ from matplotlib import pyplot as plt
 import numpy as np
 from tqdm import tqdm
 
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphfiles import loadDataset
-from utils import remove_edges
-from fitDistance import fit_GED_to_kernel_distance
-from utils import normalize_distance_matrix
+from gklearn.preimage.utils import remove_edges
+from gklearn.preimage.fitDistance import fit_GED_to_kernel_distance
+from gklearn.preimage.utils import normalize_distance_matrix
 
 
 def test_update_costs():
@@ -63,7 +61,7 @@ def median_paper_clcpc_python_best():
     
     y_all = ['3', '1', '4', '6', '7', '8', '9', '2']
     repeats = 50
-    collection_path = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/generated_datsets/monoterpenoides/'
+    collection_path = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/generated_datsets/monoterpenoides/'
     graph_dir = collection_path + 'gxl/'
     
     fn_edit_costs_output = 'results/median_paper/edit_costs_output.python_init40.k10.txt'
@@ -160,7 +158,7 @@ def median_paper_clcpc_python_bash_cpp():
     
     y_all = ['3', '1', '4', '6', '7', '8', '9', '2']
     repeats = 50
-    collection_path = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/generated_datsets/monoterpenoides/'
+    collection_path = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/generated_datsets/monoterpenoides/'
     graph_dir = collection_path + 'gxl/'
     
     fn_edit_costs_output = 'results/median_paper/edit_costs_output.txt'

gklearn/preimage/test_ged.py

@@ -14,13 +14,11 @@ import sys
 def test_NON_SYMBOLIC_cost():
     """Test edit cost LETTER2.
     """
-    import sys
-    sys.path.insert(0, "../")
-    from preimage.ged import GED, get_nb_edit_operations_nonsymbolic, get_nb_edit_operations_letter
-    from preimage.test_k_closest_graphs import reform_attributes
+    from gklearn.preimage.ged import GED, get_nb_edit_operations_nonsymbolic, get_nb_edit_operations_letter
+    from gklearn.preimage.test_k_closest_graphs import reform_attributes
     from gklearn.utils.graphfiles import loadDataset
 
-    dataset = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/Letter-high/Letter-high_A.txt'
+    dataset = '../../datasets/Letter-high/Letter-high_A.txt'
     Gn, y_all = loadDataset(dataset)
 
     g1 = Gn[200]
@@ -53,14 +51,12 @@ def test_NON_SYMBOLIC_cost():
 def test_LETTER2_cost():
     """Test edit cost LETTER2.
     """
-    import sys
-    sys.path.insert(0, "../")
-    from preimage.ged import GED, get_nb_edit_operations_letter
-    from preimage.test_k_closest_graphs import reform_attributes
+    from gklearn.preimage.ged import GED, get_nb_edit_operations_letter
+    from gklearn.preimage.test_k_closest_graphs import reform_attributes
     from gklearn.utils.graphfiles import loadDataset
 
-    ds = {'dataset': '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/collections/Letter.xml',
-          'graph_dir': '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/datasets/Letter/HIGH/'}  # node/edge symb
+    ds = {'dataset': 'cpp_ext/data/collections/Letter.xml',
+          'graph_dir': os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/data/datasets/Letter/HIGH/'}  # node/edge symb
     Gn, y_all = loadDataset(ds['dataset'], extra_params=ds['graph_dir'])
 
     g1 = Gn[200]
@@ -96,14 +92,12 @@ def test_get_nb_edit_operations_letter():
     should be the same as the cost computed by number of operations and edit
     cost constants.
     """
-    import sys
-    sys.path.insert(0, "../")
-    from preimage.ged import GED, get_nb_edit_operations_letter
-    from preimage.test_k_closest_graphs import reform_attributes
+    from gklearn.preimage.ged import GED, get_nb_edit_operations_letter
+    from gklearn.preimage.test_k_closest_graphs import reform_attributes
     from gklearn.utils.graphfiles import loadDataset
 
-    ds = {'dataset': '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/collections/Letter.xml',
-          'graph_dir': '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/datasets/Letter/HIGH/'}  # node/edge symb
+    ds = {'dataset': 'cpp_ext/data/collections/Letter.xml',
+          'graph_dir': os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/data/datasets/Letter/HIGH/'}  # node/edge symb
     Gn, y_all = loadDataset(ds['dataset'], extra_params=ds['graph_dir'])
 
     g1 = Gn[200]
@@ -136,13 +130,12 @@ def test_get_nb_edit_operations():
     numbers of edit operations. The distance/cost computed by GED should be the
     same as the cost computed by number of operations and edit cost constants.
     """
-    import sys
-    sys.path.insert(0, "../")
-    from preimage.ged import GED, get_nb_edit_operations
+    from gklearn.preimage.ged import GED, get_nb_edit_operations
     from gklearn.utils.graphfiles import loadDataset
+    import os
 
-    ds = {'dataset': '../datasets/monoterpenoides/dataset_10+.ds',
-          'graph_dir': '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/monoterpenoides/'}  # node/edge symb
+    ds = {'dataset': '../../datasets/monoterpenoides/dataset_10+.ds',
+          'graph_dir': os.path.dirname(os.path.realpath(__file__)) + '../../datasets/monoterpenoides/'}  # node/edge symb
     Gn, y_all = loadDataset(ds['dataset'])
 
     g1 = Gn[20]
@@ -173,11 +166,10 @@ def test_get_nb_edit_operations():
 def test_ged_python_bash_cpp():
     """Test ged computation with python invoking the c++ code by bash command (with updated library).
     """
-    sys.path.insert(0, "../")
     from gklearn.utils.graphfiles import loadDataset
-    from preimage.ged import GED
+    from gklearn.preimage.ged import GED
 
-    data_dir_prefix = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/'
+    data_dir_prefix = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/'
 #    collection_file = data_dir_prefix + 'generated_datsets/monoterpenoides/gxl/monoterpenoides.xml'
     collection_file = data_dir_prefix + 'generated_datsets/monoterpenoides/monoterpenoides_3_20.xml'
     graph_dir = data_dir_prefix +'generated_datsets/monoterpenoides/gxl/'
@@ -233,7 +225,7 @@ def test_ged_best_settings_updated():
     """Test ged computation with best settings the same as in the C++ code (with updated library).
     """
 
-    data_dir_prefix = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/'
+    data_dir_prefix = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/'
     collection_file = data_dir_prefix + 'generated_datsets/monoterpenoides/gxl/monoterpenoides.xml'
 #    collection_file = data_dir_prefix + 'generated_datsets/monoterpenoides/monoterpenoides_3_20.xml'
 
@@ -292,7 +284,7 @@ def test_ged_best_settings():
     """Test ged computation with best settings the same as in the C++ code.
     """
 
-    data_dir_prefix = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/'
+    data_dir_prefix = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/'
     collection_file = data_dir_prefix + 'generated_datsets/monoterpenoides/gxl/monoterpenoides.xml'
     graph_dir = data_dir_prefix +'generated_datsets/monoterpenoides/gxl/'
 
@@ -350,7 +342,7 @@ def test_ged_default():
     """Test ged computation with default settings.
     """
 
-    data_dir_prefix = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/'
+    data_dir_prefix = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/'
     collection_file = data_dir_prefix + 'generated_datsets/monoterpenoides/gxl/monoterpenoides.xml'
     graph_dir = data_dir_prefix +'generated_datsets/monoterpenoides/gxl/'
 
@@ -404,11 +396,10 @@ def test_ged_default():
 def test_ged_min():
     """Test ged computation with the "min" stabilizer.
     """
-    sys.path.insert(0, "../")
     from gklearn.utils.graphfiles import loadDataset
-    from preimage.ged import GED
+    from gklearn.preimage.ged import GED
 
-    data_dir_prefix = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/'
+    data_dir_prefix = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/'
     collection_file = data_dir_prefix + 'generated_datsets/monoterpenoides/gxl/monoterpenoides.xml'
     graph_dir = data_dir_prefix +'generated_datsets/monoterpenoides/gxl/'
 
@@ -487,8 +478,6 @@ def convertGraph(G):
 
 
 def testNxGrapĥ():
-    import sys
-    sys.path.insert(0, "../")
     from gklearn.utils.graphfiles import loadDataset
     ds = {'name': 'MUTAG', 'dataset': '../datasets/MUTAG/MUTAG_A.txt',
           'extra_params': {}}  # node/edge symb

gklearn/preimage/test_iam.py

@@ -13,14 +13,11 @@ import time
 import random
 #from tqdm import tqdm
 
-#import os
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphfiles import loadDataset
 #from gklearn.utils.logger2file import *
-from iam import iam_upgraded
-from utils import remove_edges, compute_kernel, get_same_item_indices, dis_gstar
-#from ged import ged_median
+from gklearn.preimage.iam import iam_upgraded
+from gklearn.preimage.utils import remove_edges, compute_kernel, get_same_item_indices, dis_gstar
+#from gklearn.preimage.ged import ged_median
 
 
 def test_iam_monoterpenoides_with_init40():
@@ -52,7 +49,7 @@ def test_iam_monoterpenoides_with_init40():
                   'stabilizer': ged_stabilizer}
 
     
-    collection_path = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/generated_datsets/monoterpenoides/'
+    collection_path = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/generated_datsets/monoterpenoides/'
     graph_dir = collection_path + 'gxl/'
     y_all = ['3', '1', '4', '6', '7', '8', '9', '2']
     repeats = 50

gklearn/preimage/test_k_closest_graphs.py

@@ -17,15 +17,12 @@ import multiprocessing
 from multiprocessing import Pool
 from functools import partial
 
-#import os
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphfiles import loadDataset, loadGXL
 #from gklearn.utils.logger2file import *
-from iam import iam_upgraded, iam_bash
-from utils import compute_kernel, dis_gstar, kernel_distance_matrix
-from fitDistance import fit_GED_to_kernel_distance
-#from ged import ged_median
+from gklearn.preimage.iam import iam_upgraded, iam_bash
+from gklearn.preimage.utils import compute_kernel, dis_gstar, kernel_distance_matrix
+from gklearn.preimage.fitDistance import fit_GED_to_kernel_distance
+#from gklearn.preimage.ged import ged_median
 
 
 def fit_edit_cost_constants(fit_method, edit_cost_name, 
@@ -204,6 +201,8 @@ def median_on_k_closest_graphs(Gn, node_label, edge_label, gkernel, k, fit_metho
     if Kmatrix is not None:
         Kmatrix_median = np.copy(Kmatrix[group_min,:])
         Kmatrix_median = Kmatrix_median[:,group_min]
+    else:
+        Kmatrix_median = None
         
 
     # 1. fit edit cost constants. 
@@ -379,7 +378,7 @@ def test_k_closest_graphs_with_cv():
     
     y_all = ['3', '1', '4', '6', '7', '8', '9', '2']
     repeats = 50
-    collection_path = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/generated_datsets/monoterpenoides/'
+    collection_path = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/generated_datsets/monoterpenoides/'
     graph_dir = collection_path + 'gxl/'
     
     sod_sm_list = []

gklearn/preimage/test_others.py

@@ -11,12 +11,10 @@ import matplotlib.pyplot as plt
 import time
 from tqdm import tqdm
 
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphfiles import loadDataset
-from median import draw_Letter_graph
-from ged import GED, ged_median
-from utils import get_same_item_indices, compute_kernel, gram2distances, \
+from gklearn.preimage.median import draw_Letter_graph
+from gklearn.preimage.ged import GED, ged_median
+from gklearn.preimage.utils import get_same_item_indices, compute_kernel, gram2distances, \
     dis_gstar, remove_edges
 
 

gklearn/preimage/test_preimage_iam.py

@@ -13,14 +13,11 @@ import time
 import random
 #from tqdm import tqdm
 
-#import os
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphfiles import loadDataset
-from utils import remove_edges, compute_kernel, get_same_item_indices
-from ged import ged_median
+from gklearn.preimage.utils import remove_edges, compute_kernel, get_same_item_indices
+from gklearn.preimage.ged import ged_median
 
-from preimage_iam import preimage_iam 
+from gklearn.preimage.preimage_iam import preimage_iam 
 
 
 ###############################################################################

gklearn/preimage/test_preimage_mix.py

@@ -13,13 +13,10 @@ import time
 import random
 #from tqdm import tqdm
 
-#import os
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphfiles import loadDataset
-from ged import ged_median
-from utils import compute_kernel, get_same_item_indices, remove_edges
-from preimage_iam import preimage_iam_random_mix
+from gklearn.preimage.ged import ged_median
+from gklearn.preimage.utils import compute_kernel, get_same_item_indices, remove_edges
+from gklearn.preimage.preimage_iam import preimage_iam_random_mix
 
 ###############################################################################
 # tests on different values on grid of median-sets and k.

gklearn/preimage/test_preimage_random.py

@@ -13,14 +13,10 @@ import time
 import random
 #from tqdm import tqdm
 
-#import os
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphfiles import loadDataset
-
-from preimage_random import preimage_random
-from ged import ged_median
-from utils import compute_kernel, get_same_item_indices, remove_edges
+from gklearn.preimage.preimage_random import preimage_random
+from gklearn.preimage.ged import ged_median
+from gklearn.preimage.utils import compute_kernel, get_same_item_indices, remove_edges
 
 
 ###############################################################################

gklearn/preimage/utils.py

@@ -11,8 +11,6 @@ Useful functions.
 import multiprocessing
 import numpy as np
 
-import sys
-sys.path.insert(0, "../")
 from gklearn.kernels.marginalizedKernel import marginalizedkernel
 from gklearn.kernels.untilHPathKernel import untilhpathkernel
 from gklearn.kernels.spKernel import spkernel
@@ -41,7 +39,7 @@ def dis_gstar(idx_g, idx_gi, alpha, Kmatrix, term3=0, withterm3=True):
     return np.sqrt(term1 - term2 + term3)
 
 
-def compute_kernel(Gn, graph_kernel, node_label, edge_label, verbose):
+def compute_kernel(Gn, graph_kernel, node_label, edge_label, verbose, parallel='imap_unordered'):
     if graph_kernel == 'marginalizedkernel':
         Kmatrix, _ = marginalizedkernel(Gn, node_label=node_label, edge_label=edge_label,
                                   p_quit=0.03, n_iteration=10, remove_totters=False,
@@ -49,6 +47,7 @@ def compute_kernel(Gn, graph_kernel, node_label, edge_label, verbose):
     elif graph_kernel == 'untilhpathkernel':
         Kmatrix, _ = untilhpathkernel(Gn, node_label=node_label, edge_label=edge_label,
                                   depth=7, k_func='MinMax', compute_method='trie',
+                                  parallel=parallel,
                                   n_jobs=multiprocessing.cpu_count(), verbose=verbose)
     elif graph_kernel == 'spkernel':
         mixkernel = functools.partial(kernelproduct, deltakernel, gaussiankernel)
@@ -66,18 +65,18 @@ def compute_kernel(Gn, graph_kernel, node_label, edge_label, verbose):
         Kmatrix, _ = structuralspkernel(Gn, node_label=node_label, 
                               edge_label=edge_label, node_kernels=sub_kernels,
                               edge_kernels=sub_kernels,
-                              parallel=None, n_jobs=multiprocessing.cpu_count(), 
+                              parallel=parallel, n_jobs=multiprocessing.cpu_count(), 
                               verbose=verbose)
     elif graph_kernel == 'treeletkernel':
         pkernel = functools.partial(polynomialkernel, d=2, c=1e5)
 #        pkernel = functools.partial(gaussiankernel, gamma=1e-6)
         mixkernel = functools.partial(kernelproduct, deltakernel, gaussiankernel)
         Kmatrix, _ = treeletkernel(Gn, node_label=node_label, edge_label=edge_label,
-                                   sub_kernel=pkernel,
+                                   sub_kernel=pkernel, parallel=parallel,
                                    n_jobs=multiprocessing.cpu_count(), verbose=verbose)
     elif graph_kernel == 'weisfeilerlehmankernel':
         Kmatrix, _ = weisfeilerlehmankernel(Gn, node_label=node_label, edge_label=edge_label,
-                                   height=4, base_kernel='subtree',
+                                   height=4, base_kernel='subtree', parallel=None,
                                    n_jobs=multiprocessing.cpu_count(), verbose=verbose)
         
     # normalization

gklearn/preimage/visualization.py

@@ -11,11 +11,8 @@ import matplotlib.pyplot as plt
 from mpl_toolkits.axes_grid1.inset_locator import zoomed_inset_axes, mark_inset
 from tqdm import tqdm
 
-
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphfiles import loadDataset, loadGXL
-from utils import kernel_distance_matrix, compute_kernel, dis_gstar, get_same_item_indices
+from gklearn.preimage.utils import kernel_distance_matrix, compute_kernel, dis_gstar, get_same_item_indices
 
 
 def visualize_graph_dataset(dis_measure, visual_method, draw_figure, 
@@ -115,11 +112,11 @@ def visualize_distances_in_kernel():
 #    Gn = Gn[0:50]
     fname_medians = 'expert.treelet'
     # add set median.
-    fname_sm = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/preimage/results/test_k_closest_graphs/set_median.' + fname_medians + '.gxl'
+    fname_sm = 'results/test_k_closest_graphs/set_median.' + fname_medians + '.gxl'
     set_median = loadGXL(fname_sm)
     Gn.append(set_median)
     # add generalized median (estimated pre-image.)
-    fname_gm = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/preimage/results/test_k_closest_graphs/gen_median.' + fname_medians + '.gxl'
+    fname_gm = 'results/test_k_closest_graphs/gen_median.' + fname_medians + '.gxl'
     gen_median = loadGXL(fname_gm)
     Gn.append(gen_median)
     
@@ -166,19 +163,19 @@ def visualize_distances_in_kernel():
         
     
 def visualize_distances_in_ged():
-    from fitDistance import compute_geds
-    from ged import GED
+    from gklearn.preimage.fitDistance import compute_geds
+    from gklearn.preimage.ged import GED
     ds = {'name': 'monoterpenoides', 
           'dataset': '../datasets/monoterpenoides/dataset_10+.ds'}  # node/edge symb
     Gn, y_all = loadDataset(ds['dataset'])
 #    Gn = Gn[0:50]
     # add set median.
     fname_medians = 'expert.treelet'
-    fname_sm = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/preimage/results/test_k_closest_graphs/set_median.' + fname_medians + '.gxl'
+    fname_sm = 'preimage/results/test_k_closest_graphs/set_median.' + fname_medians + '.gxl'
     set_median = loadGXL(fname_sm)
     Gn.append(set_median)
     # add generalized median (estimated pre-image.)
-    fname_gm = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/preimage/results/test_k_closest_graphs/gen_median.' + fname_medians + '.gxl'
+    fname_gm = 'preimage/results/test_k_closest_graphs/gen_median.' + fname_medians + '.gxl'
     gen_median = loadGXL(fname_gm)
     Gn.append(gen_median)
     
@@ -227,9 +224,10 @@ def visualize_distances_in_ged():
     
     
 def visualize_distances_in_kernel_monoterpenoides():
-    
+    import os
+
     ds = {'dataset': '../datasets/monoterpenoides/dataset_10+.ds',
-          'graph_dir': '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/monoterpenoides/'}  # node/edge symb
+          'graph_dir': os.path.dirname(os.path.realpath(__file__))  + '../../datasets/monoterpenoides/'}  # node/edge symb
     Gn_original, y_all = loadDataset(ds['dataset'])
 #    Gn = Gn[0:50]
     
@@ -301,11 +299,12 @@ def visualize_distances_in_kernel_monoterpenoides():
         
     
 def visualize_distances_in_ged_monoterpenoides():
-    from fitDistance import compute_geds
-    from ged import GED
+    from gklearn.preimage.fitDistance import compute_geds
+    from gklearn.preimage.ged import GED
+    import os
     
     ds = {'dataset': '../datasets/monoterpenoides/dataset_10+.ds',
-          'graph_dir': '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/monoterpenoides/'}  # node/edge symb
+          'graph_dir': os.path.dirname(os.path.realpath(__file__)) + '../../datasets/monoterpenoides/'}  # node/edge symb
     Gn_original, y_all = loadDataset(ds['dataset'])
 #    Gn = Gn[0:50]
     
@@ -379,8 +378,8 @@ def visualize_distances_in_ged_monoterpenoides():
     
 def visualize_distances_in_kernel_letter_h():
     
-    ds = {'dataset': '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/collections/Letter.xml',
-          'graph_dir': '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/datasets/Letter/HIGH/'}  # node/edge symb
+    ds = {'dataset': 'cpp_ext/data/collections/Letter.xml',
+          'graph_dir': os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/data/datasets/Letter/HIGH/'}  # node/edge symb
     Gn_original, y_all = loadDataset(ds['dataset'], extra_params=ds['graph_dir'])
 #    Gn = Gn[0:50]
     
@@ -455,8 +454,8 @@ def visualize_distances_in_ged_letter_h():
     from fitDistance import compute_geds
     from preimage.test_k_closest_graphs import reform_attributes
     
-    ds = {'dataset': '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/collections/Letter.xml',
-          'graph_dir': '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/datasets/Letter/HIGH/'}  # node/edge symb
+    ds = {'dataset': 'cpp_ext/data/collections/Letter.xml',
+          'graph_dir': os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/data/datasets/Letter/HIGH/'}  # node/edge symb
     Gn_original, y_all = loadDataset(ds['dataset'], extra_params=ds['graph_dir'])
 #    Gn = Gn[0:50]
     

gklearn/preimage/xp_fit_method.py

@@ -11,35 +11,37 @@ import csv
 from shutil import copyfile
 import networkx as nx
 import matplotlib.pyplot as plt
+import os
 
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphfiles import loadDataset, loadGXL, saveGXL
-from preimage.test_k_closest_graphs import median_on_k_closest_graphs, reform_attributes
-from preimage.utils import get_same_item_indices, kernel_distance_matrix, compute_kernel
-from preimage.find_best_k import getRelations
+from gklearn.preimage.test_k_closest_graphs import median_on_k_closest_graphs, reform_attributes
+from gklearn.preimage.utils import get_same_item_indices, kernel_distance_matrix, compute_kernel
+from gklearn.preimage.find_best_k import getRelations
 
 
 def get_dataset(ds_name):
     if ds_name == 'Letter-high': # node non-symb
-        dataset = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/collections/Letter.xml'
-        graph_dir = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/datasets/Letter/HIGH/' 
+        dataset = 'cpp_ext/data/collections/Letter.xml'
+        graph_dir = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/data/datasets/Letter/HIGH/' 
         Gn, y_all = loadDataset(dataset, extra_params=graph_dir)
         for G in Gn:
             reform_attributes(G)
     elif ds_name == 'Fingerprint':
-        dataset = '/media/ljia/DATA/research-repo/codes/Linlin/gedlib/data/collections/Fingerprint.xml'
-        graph_dir = '/media/ljia/DATA/research-repo/codes/Linlin/gedlib/data/datasets/Fingerprint/data/'
-        Gn, y_all = loadDataset(dataset, extra_params=graph_dir)
-        for G in Gn:
-            reform_attributes(G)
+#        dataset = 'cpp_ext/data/collections/Fingerprint.xml'
+#        graph_dir = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/generated_datsets/Fingerprint/node_attrs/'
+#        Gn, y_all = loadDataset(dataset, extra_params=graph_dir)
+#        for G in Gn:
+#            reform_attributes(G)
+        dataset = '../../datasets/Fingerprint/Fingerprint_A.txt'
+        graph_dir = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/generated_datsets/Fingerprint/node_attrs/'
+        Gn, y_all = loadDataset(dataset)
     elif ds_name == 'SYNTHETIC':
         pass
     elif ds_name == 'SYNTHETICnew':
-        dataset = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/SYNTHETICnew/SYNTHETICnew_A.txt'
-        graph_dir = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/generated_datsets/SYNTHETICnew'
-#        dataset = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/Letter-high/Letter-high_A.txt'
-#        graph_dir = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/datasets/Letter/HIGH/'
+        dataset = '../../datasets/SYNTHETICnew/SYNTHETICnew_A.txt'
+        graph_dir = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/generated_datsets/SYNTHETICnew'
+#        dataset = '../../datasets/Letter-high/Letter-high_A.txt'
+#        graph_dir = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/data/datasets/Letter/HIGH/'
         Gn, y_all = loadDataset(dataset)
     elif ds_name == 'Synthie':
         pass
@@ -184,6 +186,8 @@ def xp_fit_method_for_non_symbolic(parameters, save_results=True, initial_soluti
             if Kmatrix is not None:
                 Kmatrix_sub = Kmatrix[values,:]
                 Kmatrix_sub = Kmatrix_sub[:,values]
+            else:
+                Kmatrix_sub = None
             
             for repeat in range(repeats):
                 print('\nrepeat =', repeat)
@@ -273,11 +277,11 @@ def xp_fit_method_for_non_symbolic(parameters, save_results=True, initial_soluti
                     nb_dis_k_gi2gm[2] += 1
                     
                 # save median graphs.
-                fname_sm = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/output/tmp_ged/set_median.gxl'
+                fname_sm = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/output/tmp_ged/set_median.gxl'
                 fn_pre_sm_new = dir_output + 'medians/set_median.' + fit_method \
                     + '.k' + str(int(k)) + '.y' + str(y) + '.repeat' + str(repeat)
                 copyfile(fname_sm, fn_pre_sm_new + '.gxl')
-                fname_gm = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/output/tmp_ged/gen_median.gxl'
+                fname_gm = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/output/tmp_ged/gen_median.gxl'
                 fn_pre_gm_new = dir_output + 'medians/gen_median.' + fit_method \
                     + '.k' + str(int(k)) + '.y' + str(y) + '.repeat' + str(repeat)
                 copyfile(fname_gm, fn_pre_gm_new + '.gxl')
@@ -427,63 +431,101 @@ if __name__ == "__main__":
 #                                       initial_solutions=40,
 #                                       Gn_data = [Gn, y_all, graph_dir],
 #                                       k_dis_data = [dis_mat, dis_max, dis_min, dis_mean])
-    
-    
-#    #### xp 3: Fingerprint, sspkernel, using LETTER2.
+        
+        
+#    #### xp 3: SYNTHETICnew, sspkernel, using NON_SYMBOLIC.
+#    gmfile = np.load('results/xp_fit_method/Kmatrix.SYNTHETICnew.structuralspkernel.gm.npz')
+#    Kmatrix = gmfile['Kmatrix']
+#    run_time = gmfile['run_time']
+#    # normalization
+#    Kmatrix_diag = Kmatrix.diagonal().copy()
+#    for i in range(len(Kmatrix)):
+#        for j in range(i, len(Kmatrix)):
+#            Kmatrix[i][j] /= np.sqrt(Kmatrix_diag[i] * Kmatrix_diag[j])
+#            Kmatrix[j][i] = Kmatrix[i][j]
+##    np.savez('results/xp_fit_method/Kmatrix.SYNTHETICnew.spkernel.gm',
+##             Kmatrix=Kmatrix, run_time=run_time)
 #    # load dataset.
 #    print('getting dataset and computing kernel distance matrix first...')
-#    ds_name = 'Fingerprint'
+#    ds_name = 'SYNTHETICnew'
 #    gkernel = 'structuralspkernel'
 #    Gn, y_all, graph_dir = get_dataset(ds_name)
 #    # remove graphs without nodes and edges.
-#    Gn = [(idx, G) for idx, G in enumerate(Gn) if (nx.number_of_edges(G) != 0
+#    Gn = [(idx, G) for idx, G in enumerate(Gn) if (nx.number_of_nodes(G) != 0
 #          and nx.number_of_edges(G) != 0)]
 #    idx = [G[0] for G in Gn]
 #    Gn = [G[1] for G in Gn]
 #    y_all = [y_all[i] for i in idx]
-##    Gn = Gn[0:50]
-##    y_all = y_all[0:50]
+##    Gn = Gn[0:10]
+##    y_all = y_all[0:10]
+#    for G in Gn:
+#        G.graph['filename'] = 'graph' + str(G.graph['name']) + '.gxl'
 #    # compute pair distances.
-##    dis_mat, dis_max, dis_min, dis_mean = kernel_distance_matrix(Gn, None, None, 
-##        Kmatrix=None, gkernel=gkernel, verbose=True)
-#    dis_mat, dis_max, dis_min, dis_mean = 0, 0, 0, 0
+#    dis_mat, dis_max, dis_min, dis_mean = kernel_distance_matrix(Gn, None, None, 
+#        Kmatrix=Kmatrix, gkernel=gkernel, verbose=True)
+##    dis_mat, dis_max, dis_min, dis_mean = 0, 0, 0, 0
 #    # fitting and computing.
-#    fit_methods = ['k-graphs', 'expert', 'random', 'random', 'random']
+#    fit_methods = ['k-graphs', 'random', 'random', 'random']
 #    for fit_method in fit_methods:
 #        print('\n-------------------------------------')
 #        print('fit method:', fit_method)
 #        parameters = {'ds_name': ds_name,
 #                      'gkernel': gkernel,
-#                      'edit_cost_name': 'LETTER2',
+#                      'edit_cost_name': 'NON_SYMBOLIC',
 #                      'ged_method': 'mIPFP',
 #                      'attr_distance': 'euclidean',
 #                      'fit_method': fit_method}
 #        xp_fit_method_for_non_symbolic(parameters, save_results=True, 
-#                                       initial_solutions=40,
+#                                       initial_solutions=1,
 #                                       Gn_data = [Gn, y_all, graph_dir],
-#                                       k_dis_data = [dis_mat, dis_max, dis_min, dis_mean])
+#                                       k_dis_data = [dis_mat, dis_max, dis_min, dis_mean],
+#                                       Kmatrix=Kmatrix)
         
         
-#    #### xp 4: SYNTHETICnew, sspkernel, using NON_SYMBOLIC.
+#    ### xp 4: SYNTHETICnew, spkernel, using NON_SYMBOLIC.
+#    gmfile = np.load('results/xp_fit_method/Kmatrix.SYNTHETICnew.spkernel.gm.npz')
+#    Kmatrix = gmfile['Kmatrix']
+#    # normalization
+#    Kmatrix_diag = Kmatrix.diagonal().copy()
+#    for i in range(len(Kmatrix)):
+#        for j in range(i, len(Kmatrix)):
+#            Kmatrix[i][j] /= np.sqrt(Kmatrix_diag[i] * Kmatrix_diag[j])
+#            Kmatrix[j][i] = Kmatrix[i][j]
+#    run_time = 21821.35
+#    np.savez('results/xp_fit_method/Kmatrix.SYNTHETICnew.spkernel.gm',
+#             Kmatrix=Kmatrix, run_time=run_time)
+#    
 #    # load dataset.
 #    print('getting dataset and computing kernel distance matrix first...')
 #    ds_name = 'SYNTHETICnew'
-#    gkernel = 'structuralspkernel'
+#    gkernel = 'spkernel'
 #    Gn, y_all, graph_dir = get_dataset(ds_name)
-#    # remove graphs without nodes and edges.
-#    Gn = [(idx, G) for idx, G in enumerate(Gn) if (nx.number_of_edges(G) != 0
-#          and nx.number_of_edges(G) != 0)]
-#    idx = [G[0] for G in Gn]
-#    Gn = [G[1] for G in Gn]
-#    y_all = [y_all[i] for i in idx]
-#    Gn = Gn[0:10]
-#    y_all = y_all[0:10]
+##    # remove graphs without nodes and edges.
+##    Gn = [(idx, G) for idx, G in enumerate(Gn) if (nx.number_of_node(G) != 0
+##          and nx.number_of_edges(G) != 0)]
+##    idx = [G[0] for G in Gn]
+##    Gn = [G[1] for G in Gn]
+##    y_all = [y_all[i] for i in idx]
+##    Gn = Gn[0:5]
+##    y_all = y_all[0:5]
 #    for G in Gn:
 #        G.graph['filename'] = 'graph' + str(G.graph['name']) + '.gxl'
-#    # compute pair distances.
+#    
+#    # compute/read Gram matrix and pair distances.
+##    Kmatrix = compute_kernel(Gn, gkernel, None, None, True)
+##    np.savez('results/xp_fit_method/Kmatrix.' + ds_name + '.' + gkernel + '.gm', 
+##         Kmatrix=Kmatrix)
+#    gmfile = np.load('results/xp_fit_method/Kmatrix.' + ds_name + '.' + gkernel + '.gm.npz')
+#    Kmatrix = gmfile['Kmatrix']
+#    run_time = gmfile['run_time']
+##    Kmatrix = Kmatrix[[0,1,2,3,4],:]
+##    Kmatrix = Kmatrix[:,[0,1,2,3,4]]
+#    print('\nTime to compute Gram matrix for the whole dataset: ', run_time)
 #    dis_mat, dis_max, dis_min, dis_mean = kernel_distance_matrix(Gn, None, None, 
-#        Kmatrix=None, gkernel=gkernel, verbose=True)
+#        Kmatrix=Kmatrix, gkernel=gkernel, verbose=True)
+##    Kmatrix = np.zeros((len(Gn), len(Gn)))
 ##    dis_mat, dis_max, dis_min, dis_mean = 0, 0, 0, 0
+#    
 #    # fitting and computing.
 #    fit_methods = ['k-graphs', 'random', 'random', 'random']
 #    for fit_method in fit_methods:
@@ -496,68 +538,69 @@ if __name__ == "__main__":
 #                      'attr_distance': 'euclidean',
 #                      'fit_method': fit_method}
 #        xp_fit_method_for_non_symbolic(parameters, save_results=True, 
-#                                       initial_solutions=40,
-#                                       Gn_data = [Gn, y_all, graph_dir],
-#                                       k_dis_data = [dis_mat, dis_max, dis_min, dis_mean])
-        
-        
-    ### xp 5: SYNTHETICnew, spkernel, using NON_SYMBOLIC.
-    gmfile = np.load('results/xp_fit_method/Kmatrix.SYNTHETICnew.spkernel.gm.npz')
-    Kmatrix = gmfile['Kmatrix']
-    # normalization
-    Kmatrix_diag = Kmatrix.diagonal().copy()
-    for i in range(len(Kmatrix)):
-        for j in range(i, len(Kmatrix)):
-            Kmatrix[i][j] /= np.sqrt(Kmatrix_diag[i] * Kmatrix_diag[j])
-            Kmatrix[j][i] = Kmatrix[i][j]
-    run_time = 21821.35
-    np.savez('results/xp_fit_method/Kmatrix.SYNTHETICnew.spkernel.gm',
-             Kmatrix=Kmatrix, run_time=run_time)
+#                                       initial_solutions=1,
+#                                       Gn_data=[Gn, y_all, graph_dir],
+#                                       k_dis_data=[dis_mat, dis_max, dis_min, dis_mean],
+#                                       Kmatrix=Kmatrix)
+    
     
+    #### xp 5: Fingerprint, sspkernel, using LETTER2.
     # load dataset.
     print('getting dataset and computing kernel distance matrix first...')
-    ds_name = 'SYNTHETICnew'
-    gkernel = 'spkernel'
+    ds_name = 'Fingerprint'
+    gkernel = 'structuralspkernel'
     Gn, y_all, graph_dir = get_dataset(ds_name)
-#    # remove graphs without nodes and edges.
-#    Gn = [(idx, G) for idx, G in enumerate(Gn) if (nx.number_of_edges(G) != 0
+    # remove graphs without nodes and edges.
+    Gn = [(idx, G) for idx, G in enumerate(Gn) if (nx.number_of_nodes(G) != 0)]
 #          and nx.number_of_edges(G) != 0)]
-#    idx = [G[0] for G in Gn]
-#    Gn = [G[1] for G in Gn]
-#    y_all = [y_all[i] for i in idx]
-#    Gn = Gn[0:5]
-#    y_all = y_all[0:5]
+    idx = [G[0] for G in Gn]
+    Gn = [G[1] for G in Gn]
+    y_all = [y_all[i] for i in idx]
+    y_idx = get_same_item_indices(y_all)
+    # remove unused labels.
+    for G in Gn:
+        G.graph['edge_attrs'] = []
+        for edge in G.edges:
+            del G.edges[edge]['attributes']
+            del G.edges[edge]['orient']
+            del G.edges[edge]['angle']
+    Gn = Gn[805:815]
+    y_all = y_all[805:815]
     for G in Gn:
         G.graph['filename'] = 'graph' + str(G.graph['name']) + '.gxl'
-    
+            
     # compute/read Gram matrix and pair distances.
-#    Kmatrix = compute_kernel(Gn, gkernel, None, None, True)
-#    np.savez('results/xp_fit_method/Kmatrix.' + ds_name + '.' + gkernel + '.gm', 
-#         Kmatrix=Kmatrix)
-    gmfile = np.load('results/xp_fit_method/Kmatrix.' + ds_name + '.' + gkernel + '.gm.npz')
-    Kmatrix = gmfile['Kmatrix']
-    run_time = gmfile['run_time']
+    Kmatrix = compute_kernel(Gn, gkernel, None, None, True, parallel='imap_unordered')
+    np.savez('results/xp_fit_method/Kmatrix.' + ds_name + '.' + gkernel + '.gm', 
+         Kmatrix=Kmatrix)
+#    gmfile = np.load('results/xp_fit_method/Kmatrix.' + ds_name + '.' + gkernel + '.gm.npz')
+#    Kmatrix = gmfile['Kmatrix']
+#    run_time = gmfile['run_time']
 #    Kmatrix = Kmatrix[[0,1,2,3,4],:]
 #    Kmatrix = Kmatrix[:,[0,1,2,3,4]]
-    print('\nTime to compute Gram matrix for the whole dataset: ', run_time)
+#    print('\nTime to compute Gram matrix for the whole dataset: ', run_time)
     dis_mat, dis_max, dis_min, dis_mean = kernel_distance_matrix(Gn, None, None, 
         Kmatrix=Kmatrix, gkernel=gkernel, verbose=True)
 #    Kmatrix = np.zeros((len(Gn), len(Gn)))
 #    dis_mat, dis_max, dis_min, dis_mean = 0, 0, 0, 0
-    
+            
+    # compute pair distances.
+#    dis_mat, dis_max, dis_min, dis_mean = kernel_distance_matrix(Gn, None, None, 
+#        Kmatrix=None, gkernel=gkernel, verbose=True)
+#    dis_mat, dis_max, dis_min, dis_mean = 0, 0, 0, 0
     # fitting and computing.
-    fit_methods = ['k-graphs', 'random', 'random', 'random']
+    fit_methods = ['k-graphs', 'expert', 'random', 'random', 'random']
     for fit_method in fit_methods:
         print('\n-------------------------------------')
         print('fit method:', fit_method)
         parameters = {'ds_name': ds_name,
                       'gkernel': gkernel,
-                      'edit_cost_name': 'NON_SYMBOLIC',
+                      'edit_cost_name': 'LETTER2',
                       'ged_method': 'mIPFP',
                       'attr_distance': 'euclidean',
                       'fit_method': fit_method}
         xp_fit_method_for_non_symbolic(parameters, save_results=True, 
-                                       initial_solutions=1,
-                                       Gn_data=[Gn, y_all, graph_dir],
-                                       k_dis_data=[dis_mat, dis_max, dis_min, dis_mean],
+                                       initial_solutions=40,
+                                       Gn_data = [Gn, y_all, graph_dir],
+                                       k_dis_data = [dis_mat, dis_max, dis_min, dis_mean],
                                        Kmatrix=Kmatrix)

gklearn/preimage/xp_letter_h.py

@@ -12,17 +12,15 @@ from shutil import copyfile
 import networkx as nx
 import matplotlib.pyplot as plt
 
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphfiles import loadDataset, loadGXL, saveGXL
-from preimage.test_k_closest_graphs import median_on_k_closest_graphs, reform_attributes
-from preimage.utils import get_same_item_indices, kernel_distance_matrix
-from preimage.find_best_k import getRelations
+from gklearn.preimage.test_k_closest_graphs import median_on_k_closest_graphs, reform_attributes
+from gklearn.preimage.utils import get_same_item_indices, kernel_distance_matrix
+from gklearn.preimage.find_best_k import getRelations
 
 
 def xp_letter_h_LETTER2_cost():
-    ds = {'dataset': '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/collections/Letter.xml',
-          'graph_dir': '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/datasets/Letter/HIGH/'}  # node/edge symb
+    ds = {'dataset': 'cpp_ext/data/collections/Letter.xml',
+          'graph_dir': os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/data/datasets/Letter/HIGH/'}  # node/edge symb
     Gn, y_all = loadDataset(ds['dataset'], extra_params=ds['graph_dir'])
     
     dis_mat, dis_max, dis_min, dis_mean = kernel_distance_matrix(Gn, None, None, Kmatrix=None, gkernel='structuralspkernel')
@@ -177,11 +175,11 @@ def xp_letter_h_LETTER2_cost():
                     nb_dis_k_gi2gm[2] += 1
                     
                 # save median graphs.
-                fname_sm = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/output/tmp_ged/set_median.gxl'
+                fname_sm = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/output/tmp_ged/set_median.gxl'
                 fn_pre_sm_new = dir_output + 'medians/set_median.' + fit_method \
                     + '.k' + str(int(k)) + '.y' + y + '.repeat' + str(repeat)
                 copyfile(fname_sm, fn_pre_sm_new + '.gxl')
-                fname_gm = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/output/tmp_ged/gen_median.gxl'
+                fname_gm = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/output/tmp_ged/gen_median.gxl'
                 fn_pre_gm_new = dir_output + 'medians/gen_median.' + fit_method \
                     + '.k' + str(int(k)) + '.y' + y + '.repeat' + str(repeat)
                 copyfile(fname_gm, fn_pre_gm_new + '.gxl')
@@ -243,8 +241,8 @@ def xp_letter_h_LETTER2_cost():
 
 
 def xp_letter_h():
-    ds = {'dataset': '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/collections/Letter.xml',
-          'graph_dir': '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/data/datasets/Letter/HIGH/'}  # node/edge symb
+    ds = {'dataset': 'cpp_ext/data/collections/Letter.xml',
+          'graph_dir': os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/data/datasets/Letter/HIGH/'}  # node/edge symb
     Gn, y_all = loadDataset(ds['dataset'], extra_params=ds['graph_dir'])
     for G in Gn:
         reform_attributes(G)
@@ -396,11 +394,11 @@ def xp_letter_h():
                     nb_dis_k_gi2gm[2] += 1
                     
                 # save median graphs.
-                fname_sm = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/output/tmp_ged/set_median.gxl'
+                fname_sm = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/output/tmp_ged/set_median.gxl'
                 fn_pre_sm_new = dir_output + 'medians/set_median.' + fit_method \
                     + '.k' + str(int(k)) + '.y' + y + '.repeat' + str(repeat)
                 copyfile(fname_sm, fn_pre_sm_new + '.gxl')
-                fname_gm = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/output/tmp_ged/gen_median.gxl'
+                fname_gm = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/output/tmp_ged/gen_median.gxl'
                 fn_pre_gm_new = dir_output + 'medians/gen_median.' + fit_method \
                     + '.k' + str(int(k)) + '.y' + y + '.repeat' + str(repeat)
                 copyfile(fname_gm, fn_pre_gm_new + '.gxl')

gklearn/preimage/xp_monoterpenoides.py

@@ -13,16 +13,16 @@ from shutil import copyfile
 import networkx as nx
 import matplotlib.pyplot as plt
 
-import sys
-sys.path.insert(0, "../")
 from gklearn.utils.graphfiles import loadDataset, loadGXL, saveGXL
-from preimage.test_k_closest_graphs import median_on_k_closest_graphs, reform_attributes
-from preimage.utils import get_same_item_indices
-from preimage.find_best_k import getRelations
+from gklearn.preimage.test_k_closest_graphs import median_on_k_closest_graphs, reform_attributes
+from gklearn.preimage.utils import get_same_item_indices
+from gklearn.preimage.find_best_k import getRelations
 
 def xp_monoterpenoides():
-    ds = {'dataset': '../datasets/monoterpenoides/dataset_10+.ds',
-          'graph_dir': '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/monoterpenoides/'}  # node/edge symb
+    import os
+
+    ds = {'dataset': '../../datasets/monoterpenoides/dataset_10+.ds',
+          'graph_dir': os.path.dirname(os.path.realpath(__file__)) + '../../datasets/monoterpenoides/'}  # node/edge symb
     Gn, y_all = loadDataset(ds['dataset'])
 #    ds = {'name': 'Letter-high', 
 #          'dataset': '../datasets/Letter-high/Letter-high_A.txt'}  # node/edge symb
@@ -169,11 +169,11 @@ def xp_monoterpenoides():
                     nb_dis_k_gi2gm[2] += 1
                     
                 # save median graphs.
-                fname_sm = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/output/tmp_ged/set_median.gxl'
+                fname_sm = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/output/tmp_ged/set_median.gxl'
                 fn_pre_sm_new = dir_output + 'medians/set_median.' + fit_method \
                     + '.k' + str(int(k)) + '.y' + str(int(y)) + '.repeat' + str(repeat)
                 copyfile(fname_sm, fn_pre_sm_new + '.gxl')
-                fname_gm = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/output/tmp_ged/gen_median.gxl'
+                fname_gm = os.path.dirname(os.path.realpath(__file__)) + '/cpp_ext/output/tmp_ged/gen_median.gxl'
                 fn_pre_gm_new = dir_output + 'medians/gen_median.' + fit_method \
                     + '.k' + str(int(k)) + '.y' + str(int(y)) + '.repeat' + str(repeat)
                 copyfile(fname_gm, fn_pre_gm_new + '.gxl')

gklearn/tests/test_graphkernels.py

@@ -15,10 +15,16 @@ def chooseDataset(ds_name):
         ds_file = 'datasets/Alkane/dataset.ds'
         ds_y = 'datasets/Alkane/dataset_boiling_point_names.txt'
         Gn, y = loadDataset(ds_file, filename_y=ds_y)
+        for G in Gn:
+            for node in G.nodes:
+                del G.nodes[node]['attributes']
     # node symbolic labels.
     elif ds_name == 'Acyclic':
         ds_file = 'datasets/acyclic/dataset_bps.ds'
         Gn, y = loadDataset(ds_file)
+        for G in Gn:
+            for node in G.nodes:
+                del G.nodes[node]['attributes']
     # node non-symbolic labels.
     elif ds_name == 'Letter-med':
         ds_file = 'datasets/Letter-med/Letter-med_A.txt'
@@ -27,14 +33,39 @@ def chooseDataset(ds_name):
     elif ds_name == 'AIDS':
         ds_file = 'datasets/AIDS/AIDS_A.txt'
         Gn, y = loadDataset(ds_file)
-        
+    
+    # edge non-symbolic labels (no node labels).
+    elif ds_name == 'Fingerprint_edge':
+        import networkx as nx
+        ds_file = 'datasets/Fingerprint/Fingerprint_A.txt'
+        Gn, y = loadDataset(ds_file)
+        Gn = [(idx, G) for idx, G in enumerate(Gn) if nx.number_of_edges(G) != 0]
+        idx = [G[0] for G in Gn]
+        Gn = [G[1] for G in Gn]
+        y = [y[i] for i in idx]
+        for G in Gn:
+            G.graph['node_attrs'] = []
+            for node in G.nodes:
+                del G.nodes[node]['attributes']
+                del G.nodes[node]['x']
+                del G.nodes[node]['y']
     # edge non-symbolic labels (and node non-symbolic labels).
     elif ds_name == 'Fingerprint':
+        import networkx as nx
         ds_file = 'datasets/Fingerprint/Fingerprint_A.txt'
         Gn, y = loadDataset(ds_file)
+        Gn = [(idx, G) for idx, G in enumerate(Gn) if nx.number_of_edges(G) != 0]
+        idx = [G[0] for G in Gn]
+        Gn = [G[1] for G in Gn]
+        y = [y[i] for i in idx]
+    # edge symbolic and non-symbolic labels (and node symbolic and non-symbolic labels).
+    elif ds_name == 'Cuneiform':
+        import networkx as nx
+        ds_file = 'datasets/Cuneiform/Cuneiform_A.txt'
+        Gn, y = loadDataset(ds_file)
     
-    Gn = Gn[0:10]
-    y = y[0:10]
+    Gn = Gn[0:3]
+    y = y[0:3]
     
     return Gn, y
 
@@ -152,7 +183,7 @@ def test_spkernel(ds_name, parallel):
 
 
 #@pytest.mark.parametrize('ds_name', ['Alkane', 'Acyclic', 'Letter-med', 'AIDS', 'Fingerprint'])
-@pytest.mark.parametrize('ds_name', ['Alkane', 'Acyclic', 'Letter-med', 'AIDS'])
+@pytest.mark.parametrize('ds_name', ['Alkane', 'Acyclic', 'Letter-med', 'AIDS', 'Fingerprint', 'Fingerprint_edge', 'Cuneiform'])
 @pytest.mark.parametrize('parallel', ['imap_unordered', None])
 def test_structuralspkernel(ds_name, parallel):
     """Test structural shortest path kernel.
@@ -246,4 +277,5 @@ def test_weisfeilerlehmankernel(ds_name, parallel, base_kernel):
         
 
 if __name__ == "__main__":
-    test_spkernel()
+#    test_spkernel('Alkane', 'imap_unordered')
+    test_structuralspkernel('Fingerprint_edge', 'imap_unordered')

gklearn/utils/graphfiles.py

@@ -753,15 +753,12 @@ if __name__ == '__main__':
 #          'extra_params': {'am_sp_al_nl_el': [0, 0, 3, 1, 2]}}  # node/edge symb
 #    Gn, y = loadDataset(ds['dataset'], extra_params=ds['extra_params'])
 #    saveDataset(Gn, y, group='xml', filename='temp/temp')
-    dataset = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/SYNTHETICnew/SYNTHETICnew_A.txt'
-    Gn, y_all = loadDataset(dataset)
-    filename = '/media/ljia/DATA/research-repo/codes/others/gedlib/tests_linlin/generated_datsets/SYNTHETICnew/SYNTHETICnew'
-    saveDataset(Gn, y_all, gformat='gxl', group='xml', filename=filename)
     
     # test - new way to add labels and attributes.
-#    dataset = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/SYNTHETICnew/SYNTHETICnew_A.txt'
-#    dataset = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/Fingerprint/Fingerprint_A.txt'
-#    dataset = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/Letter-med/Letter-med_A.txt'
-#    dataset = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/AIDS/AIDS_A.txt'
-#    dataset = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/ENZYMES_txt/ENZYMES_A_sparse.txt'
-#    Gn, y_all = loadDataset(dataset)
+#    dataset = '../../datasets/SYNTHETICnew/SYNTHETICnew_A.txt'
+#    dataset = '../../datasets/Fingerprint/Fingerprint_A.txt'
+#    dataset = '../../datasets/Letter-med/Letter-med_A.txt'
+#    dataset = '../../datasets/AIDS/AIDS_A.txt'
+#    dataset = '../../datasets/ENZYMES_txt/ENZYMES_A_sparse.txt'
+#    Gn, y_all = loadDataset(dataset)
+    pass

gklearn/utils/model_selection_precomputed.py

@@ -11,7 +11,6 @@ from sklearn.model_selection import KFold, train_test_split, ParameterGrid
 from multiprocessing import Pool, Array
 from functools import partial
 import sys
-sys.path.insert(0, "../")
 import os
 import time
 import datetime
@@ -74,8 +73,6 @@ def model_selection_for_precomputed_kernel(datafile,
     Examples
     --------
     >>> import numpy as np
-    >>> import sys
-    >>> sys.path.insert(0, "../")
     >>> from gklearn.utils.model_selection_precomputed import model_selection_for_precomputed_kernel
     >>> from gklearn.kernels.untilHPathKernel import untilhpathkernel
     >>>

notebooks/utils/plot_all_graphs.ipynb

@@ -46466,7 +46466,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.8"
+   "version": "3.6.9"
   }
  },
  "nbformat": 4,

notebooks/utils/plot_all_graphs.py

@@ -7,10 +7,6 @@ Created on Tue Jan  7 15:25:36 2020
 """
 
 # draw all the praphs
-import sys
-import pathlib
-sys.path.insert(0, "../../")
-
 import matplotlib.pyplot as plt
 
 import networkx as nx
@@ -19,6 +15,58 @@ from gklearn.utils.graphfiles import loadDataset, loadGXL
 
 
 def main(): 
+    # MUTAG dataset.
+    dataset, y = loadDataset("../../datasets/MUTAG/MUTAG_A.txt")
+    for idx in [65]:#[6]:
+        G = dataset[idx]
+        for node in G.nodes:
+            if G.nodes[node]['atom'] == '0':
+                G.nodes[node]['atom'] = 'C'
+            elif G.nodes[node]['atom'] == '1':
+                G.nodes[node]['atom'] = 'N'
+            elif G.nodes[node]['atom'] == '2':
+                G.nodes[node]['atom'] = 'O'
+            elif G.nodes[node]['atom'] == '3':
+                G.nodes[node]['atom'] = 'F'
+            elif G.nodes[node]['atom'] == '4':
+                G.nodes[node]['atom'] = 'I'
+            elif G.nodes[node]['atom'] == '5':
+                G.nodes[node]['atom'] = 'Cl'
+            elif G.nodes[node]['atom'] == '6':
+                G.nodes[node]['atom'] = 'Br'
+        ecolors = []
+        for edge in G.edges:
+            if G.edges[edge]['bond_type'] == '0':
+                ecolors.append('orange')
+            elif G.edges[edge]['bond_type'] == '1':
+                ecolors.append('r')
+            elif G.edges[edge]['bond_type'] == '2':
+                ecolors.append('purple')
+            elif G.edges[edge]['bond_type'] == '3':
+                ecolors.append('orange')
+
+        print(idx)
+        print(nx.get_node_attributes(G, 'atom'))
+        edge_labels = nx.get_edge_attributes(G, 'bond_type')
+        print(edge_labels)
+        pos=nx.spring_layout(G)
+        nx.draw(G, 
+                pos,
+                node_size=500,
+                labels=nx.get_node_attributes(G, 'atom'), 
+                node_color='blue', 
+                font_color='w', 
+                edge_color=ecolors,
+                width=3,
+                with_labels=True)
+#        edge_labels = nx.draw_networkx_edge_labels(G, pos, 
+#                                                   edge_labels=edge_labels,
+#                                                   font_color='pink')
+        plt.savefig('mol1_graph.svg', format='svg', dpi=300)
+        plt.show()
+        plt.clf()
+    
+    
 #    # monoterpenoides dataset.
 #    dataset, y = loadDataset("../../datasets/monoterpenoides/dataset_10+.ds")
 #    for idx in [12,22,29,74]:
@@ -67,35 +115,35 @@ def main():
 #        draw_Fingerprint_graph(Gn[idx], file_prefix='')
     
     
-    # SYNTHETIC dataset.
-    dataset = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/SYNTHETICnew/SYNTHETICnew_A.txt'
-    Gn, y_all = loadDataset(dataset)
-    
-    idx_no_node = []
-    idx_no_edge = []
-    idx_no_both = []
-    for idx, G in enumerate(Gn):
-        if nx.number_of_nodes(G) == 0:
-            idx_no_node.append(idx)
-            if nx.number_of_edges(G) == 0:
-                idx_no_both.append(idx)
-        if nx.number_of_edges(G) == 0:
-            idx_no_edge.append(idx)
-#        file_prefix = '../results/graph_images/SYNTHETIC/' + G.graph['name']
-#        draw_SYNTHETIC_graph(Gn[idx], file_prefix=file_prefix, save=True)
-#        draw_SYNTHETIC_graph(Gn[idx])
-    print('nb_no_node: ', len(idx_no_node))
-    print('nb_no_edge: ', len(idx_no_edge))
-    print('nb_no_both: ', len(idx_no_both))
-    print('idx_no_node: ', idx_no_node)
-    print('idx_no_edge: ', idx_no_edge)
-    print('idx_no_both: ', idx_no_both)
-    
-    for idx in [0, 10, 100]:
-        print(idx)
-        print(Gn[idx].nodes(data=True))
-        print(Gn[idx].edges(data=True))
-        draw_SYNTHETIC_graph(Gn[idx], save=None)
+#    # SYNTHETIC dataset.
+#    dataset = '/media/ljia/DATA/research-repo/codes/Linlin/graphkit-learn/datasets/SYNTHETICnew/SYNTHETICnew_A.txt'
+#    Gn, y_all = loadDataset(dataset)
+#    
+#    idx_no_node = []
+#    idx_no_edge = []
+#    idx_no_both = []
+#    for idx, G in enumerate(Gn):
+#        if nx.number_of_nodes(G) == 0:
+#            idx_no_node.append(idx)
+#            if nx.number_of_edges(G) == 0:
+#                idx_no_both.append(idx)
+#        if nx.number_of_edges(G) == 0:
+#            idx_no_edge.append(idx)
+##        file_prefix = '../results/graph_images/SYNTHETIC/' + G.graph['name']
+##        draw_SYNTHETIC_graph(Gn[idx], file_prefix=file_prefix, save=True)
+##        draw_SYNTHETIC_graph(Gn[idx])
+#    print('nb_no_node: ', len(idx_no_node))
+#    print('nb_no_edge: ', len(idx_no_edge))
+#    print('nb_no_both: ', len(idx_no_both))
+#    print('idx_no_node: ', idx_no_node)
+#    print('idx_no_edge: ', idx_no_edge)
+#    print('idx_no_both: ', idx_no_both)
+#    
+#    for idx in [0, 10, 100]:
+#        print(idx)
+#        print(Gn[idx].nodes(data=True))
+#        print(Gn[idx].edges(data=True))
+#        draw_SYNTHETIC_graph(Gn[idx], save=None)
         
         
 def plot_a_graph(graph_filename):