delete redundant parameters for astar (#71)

README.md

@@ -55,7 +55,7 @@ Now the pygmtools is available with the ``numpy`` backend.
 The following packages are required, and shall be automatically installed by ``pip``:
 
 ```
-Python >= 3.7
+Python >= 3.8
 requests >= 2.25.1
 scipy >= 1.4.1
 Pillow >= 7.2.0
@@ -64,6 +64,7 @@ easydict >= 1.7
 appdirs >= 1.4.4
 tqdm >= 4.64.1
 wget >= 3.2
+networkx >= 2.8.8
 ```
 
 ## Available Graph Matching Solvers

pygmtools/classic_solvers.py

@@ -1062,8 +1062,7 @@ def ipfp(K, n1=None, n2=None, n1max=None, n2max=None, x0=None,
         return result
 
 
-def astar(feat1, feat2, A1, A2, n1=None, n2=None, channel=None, dropout=0, beam_width=0, 
-          trust_fact=1, no_pred_size=0, backend=None):
+def astar(feat1, feat2, A1, A2, n1=None, n2=None, channel=None, beam_width=0, backend=None):
     r"""
     A\* (A-star) solver for graph matching (Lawler's QAP).
     The **A\*** solver was originally proposed to solve the graph edit distance (GED) problem. It finds the optimal
@@ -1081,10 +1080,7 @@ def astar(feat1, feat2, A1, A2, n1=None, n2=None, channel=None, dropout=0, beam_
     :param channel: (default: None)  Channel size of the input layer. If given, it must match the feature dimension (d) of feat1, feat2. 
         If not given, it will be defined by the feature dimension (d) of feat1, feat2. 
         Ignored if the network object isgiven (ignored if network!=None)
-    :param dropout: (default: 0) Dropout probability
     :param beam_width: (default: 0) Size of beam-search witdh (0 = no beam).
-    :param trust_fact: (default: 1) The trust factor on GNN prediction (0 = no GNN).
-    :param no_pred_size: (default: 0) If the smaller graph has no more than x nodes, stop using heuristics.
     :param backend: (default: ``pygmtools.BACKEND`` variable) the backend for computation.
     :return: :math:`(b\times n_1 \times n_2)` the doubly-stochastic matching matrix
 
@@ -1188,7 +1184,7 @@ def astar(feat1, feat2, A1, A2, n1=None, n2=None, channel=None, dropout=0, beam_
     if n1 is not None: _check_data_type(n1, 'n1', backend)
     if n2 is not None: _check_data_type(n2, 'n2', backend)
 
-    args = (feat1, feat2, A1, A2, n1, n2, channel, dropout, beam_width, trust_fact, no_pred_size)
+    args = (feat1, feat2, A1, A2, n1, n2, channel, beam_width)
     try:
         mod = importlib.import_module(f'pygmtools.{backend}_backend')
         fn = mod.astar

pygmtools/neural_solvers.py

@@ -1274,7 +1274,7 @@ def ngm(K, n1=None, n2=None, n1max=None, n2max=None, x0=None,
 
 
 def genn_astar(feat1, feat2, A1, A2, n1=None, n2=None, channel=None, filters_1=64, filters_2=32, filters_3=16,
-           tensor_neurons=16, dropout=0, beam_width=0, trust_fact=1, no_pred_size=0,
+           tensor_neurons=16, beam_width=0, trust_fact=1, no_pred_size=0,
            network=None, return_network=False, pretrain='AIDS700nef', backend=None):
     r"""
     The **GENN-A\*** (Graph Edit Neural Network A\*) solver for graph matching (and graph edit distance)
@@ -1306,7 +1306,6 @@ def genn_astar(feat1, feat2, A1, A2, n1=None, n2=None, channel=None, filters_1=6
     :param filters_2: (default: 32) Filters (neurons) in 2nd convolution.
     :param filters_3: (default: 16) Filters (neurons) in 2nd convolution.
     :param tensor_neurons: (default: 16) Neurons in tensor network layer.
-    :param dropout: (default: 0) Dropout probability
     :param beam_width: (default: 0) Size of beam-search witdh (0 = no beam).
     :param trust_fact: (default: 1) The trust factor on GNN prediction (0 = no GNN).
     :param no_pred_size: (default: 0) If the smaller graph has no more than x nodes, stop using heuristics.
@@ -1472,7 +1471,7 @@ def genn_astar(feat1, feat2, A1, A2, n1=None, n2=None, channel=None, filters_1=6
     if n2 is not None: _check_data_type(n2, 'n2', backend)
 
     args = (feat1, feat2, A1, A2, n1, n2, channel, filters_1, filters_2, filters_3, 
-            tensor_neurons, dropout, beam_width, trust_fact, no_pred_size, network, pretrain)
+            tensor_neurons, beam_width, trust_fact, no_pred_size, network, pretrain)
     try:
         mod = importlib.import_module(f'pygmtools.{backend}_backend')
         fn = mod.genn_astar

pygmtools/pytorch_astar_modules.py

@@ -22,7 +22,6 @@ def default_parameter():
     params['filters_2'] = 32
     params['filters_3'] = 16
     params['tensor_neurons'] = 16
-    params['dropout'] = 0
     params['astar_beam_width'] = 0
     params['astar_trust_fact'] = 1
     params['astar_no_pred'] = 0

pygmtools/pytorch_backend.py

@@ -869,6 +869,7 @@ def __init__(self, args):
         :param number_of_labels: Number of node labels.
         """
         super(GENN, self).__init__()
+        self.training = False
         self.args = args
         if self.args['use_net']:
             self.number_labels = self.args['channel']
@@ -909,10 +910,8 @@ def convolutional_pass(self, edge_index, x, edge_weight=None):
 
         features = self.convolution_1(edge_index, x, edge_weight)
         features = F.relu(features)
-        features = F.dropout(features, p=self.args['dropout'], training=self.training)
         features = self.convolution_2(edge_index, features, edge_weight)
         features = F.relu(features)
-        features = F.dropout(features, p=self.args['dropout'], training=self.training)
         features = self.convolution_3(edge_index, features, edge_weight)
         return features
 
@@ -1098,17 +1097,17 @@ def hungarian_ged(node_cost_mat: torch.Tensor, n1, n2):
     return pred_x, ged_lower_bound
 
 
-def astar(feat1, feat2, A1, A2, n1, n2, channel, dropout, beam_width, trust_fact, no_pred_size):
+def astar(feat1, feat2, A1, A2, n1, n2, channel, beam_width):
     """ 
     Pytorch implementation of ASTAR
     """
-    return astar_kernel(feat1, feat2, A1, A2, n1, n2, channel, dropout=dropout, beam_width=beam_width, 
-                        filters_1=64, filters_2=32, filters_3=16, tensor_neurons=16, trust_fact=trust_fact, 
-                        no_pred_size=no_pred_size, pretrain=False, network=None, use_net=False)
+    return astar_kernel(feat1, feat2, A1, A2, n1, n2, channel, beam_width=beam_width, 
+                        filters_1=64, filters_2=32, filters_3=16, tensor_neurons=16, trust_fact=1.0, 
+                        no_pred_size=0, pretrain=False, network=None, use_net=False)
 
 
 def astar_kernel(feat1, feat2, A1, A2, n1, n2, channel, filters_1, filters_2, filters_3,
-          tensor_neurons, dropout, beam_width, trust_fact, no_pred_size, network, pretrain, use_net):
+          tensor_neurons, beam_width, trust_fact, no_pred_size, network, pretrain, use_net):
     """ 
     The true implementation of astar and genn_astar functions
     """
@@ -1139,7 +1138,6 @@ def astar_kernel(feat1, feat2, A1, A2, n1, n2, channel, filters_1, filters_2, fi
         args['filters_2'] = filters_2
         args['filters_3'] = filters_3
         args['tensor_neurons'] = tensor_neurons
-        args['dropout'] = dropout
         args['astar_beam_width'] = beam_width
         args['astar_trust_fact'] = trust_fact
         args['astar_no_pred'] = no_pred_size
@@ -1534,12 +1532,12 @@ def ngm(K, n1, n2, n1max, n2max, x0, gnn_channels, sk_emb, sk_max_iter, sk_tau,
 
 
 def genn_astar(feat1, feat2, A1, A2, n1, n2, channel, filters_1, filters_2, filters_3,
-          tensor_neurons, dropout, beam_width, trust_fact, no_pred_size, network, pretrain):
+          tensor_neurons, beam_width, trust_fact, no_pred_size, network, pretrain):
     """
     Pytorch implementation of GENN-ASTAR
     """
     return astar_kernel(feat1, feat2, A1, A2, n1, n2, channel, filters_1, filters_2, filters_3,
-          tensor_neurons, dropout, beam_width, trust_fact, no_pred_size, network, pretrain, use_net=True)
+          tensor_neurons, beam_width, trust_fact, no_pred_size, network, pretrain, use_net=True)
 
 
 #############################################

tests/test_classic_solvers.py

@@ -455,16 +455,12 @@ def test_astar(get_backend):
     backends = get_backends(get_backend)
     # heuristic_prediction
     args1 = (list(range(10, 16, 2)), 10, pygm.astar,{
-        "beam_width": [0, 1, 2],
-        "trust_fact": [0.9, 0.95, 1.0],
-        "no_pred_size": [0, 1],
+        "beam_width": [0, 1, 2]
     }, backends)
 
     # non-batched input
     args2 = ([10], 10, pygm.astar,{
-        "beam_width": [0, 1, 2],
-        "trust_fact": [0.9, 0.95, 1.0],
-        "no_pred_size": [0, 1],
+        "beam_width": [0, 1, 2]
     }, backends)
 
     _test_astar(*args1)