Weighting decoding graphs (#427)

* Update decoding_graph.py

Allowing the user to provide costume weights for DecodingGraph.weight_syndrome_graph() as a new argument "error_probs".
- counts is not a required argument anymore
- error_probs is also not required and overridden by counts if both of them are provided
- at least one of the above arguments should be specified to avoid NotImplementedError

* Update decoding_graph.py

Allowing the user to provide costume weights for DecodingGraph.weight_syndrome_graph() as a new argument "error_probs".
- counts is not a required argument anymore
- error_probs is also not required and overridden by counts if both of them are provided
- at least one of the above arguments should be specified to avoid NotImplementedError

* Update decoding_graph.py

* Update decoding_graph.py

linting

* Update decoding_graph.py

remove self-assigning variable

* Update decoding_graph.py

src/qiskit_qec/decoders/decoding_graph.py

@@ -303,22 +303,32 @@ def get_error_probs(
         else:
             return error_probs
 
-    def weight_syndrome_graph(self, counts, method: str = METHOD_SPITZ):
+    def weight_syndrome_graph(
+        self, counts: dict = None, method: str = METHOD_SPITZ, error_probs: dict = None
+    ):
         """Generate weighted syndrome graph from result counts.
 
         Args:
             counts (dict): Counts dictionary of the results used to calculate
             the weights.
             method (string): Method to used for calculation. Supported
             methods are 'spitz' (default) and 'naive'.
+            error_probs (dict): probability that the syndrome contains the node pair
+            of a given edge. Overridden by counts if both are given.
 
         Additional information:
             Uses `counts` to estimate the probability of the errors that
             create the pairs of nodes in graph. The edge weights are then
             replaced with the corresponding -log(p/(1-p).
         """
 
-        error_probs = self.get_error_probs(counts, method=method)
+        if counts:
+            error_probs = self.get_error_probs(counts, method=method)
+        elif not error_probs:
+            raise NotImplementedError(
+                "No information provided to reweight the graph."
+                + "Specify either counts or error_probs."
+            )
 
         boundary_nodes = []
         for n, node in enumerate(self.graph.nodes()):