Revert changes to expected loss

Actually in these tests there is only a single batch, so the new parameter has no effect

Author: FlorentinD

doc/asciidoc/machine-learning/node-embeddings/graph-sage/graph-sage.adoc

@@ -278,7 +278,7 @@ RETURN
 .Results
 |===
 | modelName           | didConverge | ranEpochs | epochLosses
-| "exampleTrainModel" | true        | 1         | [186.0494680638198]
+| "exampleTrainModel" | true        | 1         | [186.04946807210226]
 |===
 --
 
@@ -504,13 +504,13 @@ YIELD nodeId, embedding
 .Results
 |===
 | nodeId | embedding
-| 0      | [0.5285002294775042, 0.46821819621782496, 0.7081378593674258]
-| 1      | [0.5285002294779538, 0.4682181962180586, 0.7081378593669356]
-| 2      | [0.5285002294774878, 0.46821819621781646, 0.7081378593674437]
-| 3      | [0.5285002294761042, 0.4682181962170975, 0.7081378593689517]
-| 4      | [0.5285002295204241, 0.4682181962401272, 0.7081378593206474]
-| 5      | [0.528500229582853, 0.468218196272567, 0.7081378592526062]
-| 6      | [0.5285002294762983, 0.4682181962171984, 0.7081378593687399]
+| 0      | [0.528500243954147, 0.46821819122905217, 0.7081378518617193]
+| 1      | [0.5285002439545966, 0.4682181912292858, 0.7081378518612291]
+| 2      | [0.5285002439541305, 0.4682181912290437, 0.7081378518617372]
+| 3      | [0.528500243952747, 0.46821819122832464, 0.7081378518632452]
+| 4      | [0.5285002439970667, 0.46821819125135444, 0.7081378518149409]
+| 5      | [0.5285002440594959, 0.46821819128379416, 0.7081378517468996]
+| 6      | [0.528500243952941, 0.46821819122842556, 0.7081378518630335]
 |===
 --