Force atlas fix to accept renumbered graphs

datasets/README.md

@@ -9,13 +9,16 @@ This directory contains small public datasets in `mtx` and `csv` format used by
 | karate        | 34    | 156   | No       | No       |
 | dolphin       | 62    | 318   | No       | No       |
 | netscience    | 1,589 | 5,484 | No       | Yes      |
+| dining_prefs  | 26    | 52    | Yes      | Yes      |
 
 **karate** : The graph "karate" contains the network of friendships between the 34 members of a karate club at a US university, as described by Wayne Zachary in 1977.
 
 **dolphin** : The graph dolphins contains an undirected social network of frequent associations between 62 dolphins in a community living off Doubtful Sound, New Zealand, as compiled by Lusseau et al. (2003).
 
 **netscience** : The graph netscience contains a coauthorship network of scientists working on network theory and experiment, as compiled by M. Newman in May 2006.
 
+**dining_prefs** : The graph dining_prefs contains dining partner preferences from a classic social network dataset originated by J. L. Moreno (1960). The Sociometry Reader. The Free Press, Glencoe, Illinois, pg.35
+
 
 
 ### Modified datasets 

datasets/dining_prefs.csv

@@ -0,0 +1,52 @@
+Ada Cora 1
+Ada Louise 2
+Cora Ada 1
+Cora Jean 2
+Louise Marion 1
+Louise Lena 2
+Jean Helen 1
+Jean Robin 2
+Helen Jean 1
+Helen Eva 2
+Martha Marion 2
+Martha Anna 1
+Alice Martha 2
+Alice Eva 1
+Robin Helen 2
+Robin Eva 1
+Marion Martha 1
+Marion Frances 2
+Maxine Eva 2
+Maxine Adele 1
+Lena Louise 2
+Lena Marion 1
+Hazel Hilda 1
+Hazel Anna 2
+Hilda Hazel 2
+Hilda Betty 1
+Frances Marion 2
+Frances Eva 1
+Eva Marion 2
+Eva Maxine 1
+Ruth Hilda 2
+Ruth Jane 1
+Edna Adele 2
+Edna Mary 1
+Adele Marion 2
+Adele Frances 1
+Jane Adele 1
+Jane Mary 2
+Anna Maxine 1
+Anna Lena 2
+Mary Edna 1
+Mary Jane 2
+Betty Hilda 2
+Betty Edna 1
+Ella Helen 2
+Ella Ellen 1
+Ellen Edna 2
+Ellen Anna 1
+Laura Eva 1
+Laura Edna 2
+Irene Hilda 1
+Irene Ellen 2

datasets/dining_prefs.mtx

@@ -0,0 +1,54 @@
+%%MatrixMarket matrix coordinate pattern symmetric
+26 26 52
+1 2 1
+1 3 2
+2 1 1
+2 4 2
+3 9 1
+3 11 2
+4 5 1
+4 8 2
+5 4 1
+5 15 2
+6 9 2
+6 20 1
+7 6 2
+7 15 1
+8 5 2
+8 15 1
+9 6 1
+9 14 2
+10 15 2
+10 18 1
+11 3 2
+11 9 1
+12 13 1
+12 20 2
+13 12 2
+13 22 1
+14 9 2
+14 15 1
+15 9 2
+15 10 1
+16 13 2
+16 19 1
+17 18 2
+17 21 1
+18 9 2
+18 14 1
+19 18 1
+19 21 2
+20 10 1
+20 11 2
+21 17 1
+21 19 2
+22 13 2
+22 17 1
+23 5 2
+23 24 1
+24 17 2
+24 20 1
+25 15 1
+25 17 2
+26 13 1
+26 24 2

docs/cugraph/source/references/datasets.md

@@ -2,20 +2,27 @@
 
 karate
   - W. W. Zachary, *An information flow model for conflict and fission in small groups*, Journal of Anthropological Research 33, 452-473 (1977).
+
+dining_prefs 
+  - J. L. Moreno (1960). *The Sociometry Reader*, The Free Press, Glencoe, Illinois, pg.35
+
 dolphins
   - D. Lusseau, K. Schneider, O. J. Boisseau, P. Haase, E. Slooten, and S. M. Dawson,
     *The bottlenose dolphin community of Doubtful Sound features a large proportion of long-lasting associations*, 
     Behavioral Ecology and Sociobiology 54, 396-405 (2003).
+
 netscience
   - M. E. J. Newman,
     *Finding community structure in networks using the eigenvectors of matrices*, 
     Preprint physics/0605087 (2006).
+
 email-Eu-core
   - Hao Yin, Austin R. Benson, Jure Leskovec, and David F. Gleich.
     *Local Higher-order Graph Clustering.* 
     In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2017.
   - J. Leskovec, J. Kleinberg and C. Faloutsos. 
     *Graph Evolution: Densification and Shrinking Diameters*. 
     ACM Transactions on Knowledge Discovery from Data (ACM TKDD), 1(1), 2007. http://www.cs.cmu.edu/~jure/pubs/powergrowth-tkdd.pdf 
+
 polbooks
   - V. Krebs, unpublished, http://www.orgnet.com/. 

python/cugraph/cugraph/experimental/datasets/__init__.py

@@ -24,7 +24,7 @@
 
 
 meta_path = Path(__file__).parent / "metadata"
-
+dining_prefs = Dataset(meta_path / "dining_prefs.yaml")
 karate = Dataset(meta_path / "karate.yaml")
 karate_data = Dataset(meta_path / "karate_data.yaml")
 karate_undirected = Dataset(meta_path / "karate_undirected.yaml")

python/cugraph/cugraph/experimental/datasets/metadata/dining_prefs.yaml

@@ -0,0 +1,23 @@
+name: dining_prefs
+file_type: .csv
+author: J.L. Moreno
+url:  https://data.rapids.ai/cugraph/datasets/dining_prefs.csv
+refs:
+  J. L. Moreno (1960). The Sociometry Reader. The Free Press, Glencoe, Illinois, pg.35
+delim: " "
+header: None
+col_names:
+  - src
+  - dst
+  - wgt
+col_types:
+  - string
+  - string
+  - int
+has_loop: true
+is_directed: true
+is_multigraph: false
+is_symmetric: true
+number_of_edges: 52
+number_of_nodes: 26
+number_of_lines: 52

python/cugraph/cugraph/layout/force_atlas2_wrapper.pyx

@@ -56,9 +56,11 @@ def force_atlas2(input_graph,
     if not input_graph.edgelist:
         input_graph.view_edge_list()
 
-    # FIXME: This implementation assumes that the number of vertices
-    # is the max vertex ID + 1 which is not always the case.
-    num_verts = input_graph.nodes().max() + 1
+    # this code allows handling of renumbered graphs
+    if input_graph.is_renumbered():
+        num_verts = input_graph.renumber_map.df_internal_to_external['id'].max()+1
+    else:
+        num_verts = input_graph.nodes().max() + 1
     num_edges = len(input_graph.edgelist.edgelist_df['src'])
 
     cdef GraphCOOView[int,int,float] graph_float

python/cugraph/cugraph/tests/layout/test_force_atlas2.py

@@ -19,7 +19,13 @@
 from cugraph.internals import GraphBasedDimRedCallback
 from sklearn.manifold import trustworthiness
 import scipy.io
-from cugraph.experimental.datasets import karate, polbooks, dolphins, netscience
+from cugraph.experimental.datasets import (
+    karate,
+    polbooks,
+    dolphins,
+    netscience,
+    dining_prefs,
+)
 
 # Temporarily suppress warnings till networkX fixes deprecation warnings
 # (Using or importing the ABCs from 'collections' instead of from
@@ -43,11 +49,15 @@ def cugraph_call(
     strong_gravity_mode,
     gravity,
     callback=None,
+    renumber=False,
 ):
-
     G = cugraph.Graph()
+    if cu_M["src"] is not int or cu_M["dst"] is not int:
+        renumber = True
+    else:
+        renumber = False
     G.from_cudf_edgelist(
-        cu_M, source="src", destination="dst", edge_attr="wgt", renumber=False
+        cu_M, source="src", destination="dst", edge_attr="wgt", renumber=renumber
     )
 
     t1 = time.time()
@@ -72,7 +82,13 @@ def cugraph_call(
     return pos
 
 
-DATASETS = [(karate, 0.70), (polbooks, 0.75), (dolphins, 0.66), (netscience, 0.66)]
+DATASETS = [
+    (karate, 0.70),
+    (polbooks, 0.75),
+    (dolphins, 0.66),
+    (netscience, 0.66),
+    (dining_prefs, 0.50),
+]
 
 
 MAX_ITERATIONS = [500]