def main(args):
data = load_data(args)
g = data.graph
if isinstance(g, dgl.DGLGraph):
csr = g.adjacency_matrix_scipy(transpose=True)
else:
csr = nx.to_scipy_sparse_matrix(g, weight=None, format='csr')
graph_io.save_graph(args.out, csr)
def main(args):
if args.dataset == 'segtree':
g = build_segtree(batch_size=32, seq_len=512)
print('#Nodes: %d #Edges: %d' % (g.number_of_nodes(), g.number_of_edges()))
csr = g.adjacency_matrix_scipy(fmt='csr')
else:
data = load_data(args)
g = data.graph
csr = nx.to_scipy_sparse_matrix(g, weight=None, format='csr')
graph_io.save_graph(args.out, csr)
def main(args):
if args.dataset == 'segtree':
g = build_segtree(batch_size=32, seq_len=512)
print('#Nodes: %d #Edges: %d' %
(g.number_of_nodes(), g.number_of_edges()))
csr = g.adjacency_matrix_scipy(fmt='csr')
n, m = 32 * 512, 32 * 512
else:
data = load_data(args)
g = data.graph
if isinstance(g, dgl.DGLGraph):
csr = g.adjacency_matrix_scipy(transpose=True)
else:
csr = nx.to_scipy_sparse_matrix(g, weight=None, format='csr')
n, m = csr.indptr.shape[0] - 1, csr.indptr.shape[0] - 1
graph_io.save_graph(args.out, csr, n, m)
def compare_modules(m_g0, g_m0, m_g1, g_m1, g0, g1):
"""
currently saves two images of the biggest module for graph0 and graph1.
"""
print 'calling compare_modules'
top10_0 = get_top_n_modules(m_g0, 10)
top10_1 = get_top_n_modules(m_g1, 10)
top_module_genes0 = list(m_g0[top10_0[0]])
top_module_genes1 = list(m_g1[top10_1[0]])
H0 = g0.subgraph(top_module_genes0)
H1 = g1.subgraph(top_module_genes1)
print top_module_genes0
print top_module_genes1
print top_module_genes0 == top_module_genes1
print 'saving the two biggest modoles'
#todo get this to display the same graph, since edgelist is the same?
graph_io.save_graph(H0, '../results/Merlin/h0.png')
graph_io.save_graph(H1, '../results/Merlin/h1.png')
def gen_er(args):
g = nx.fast_gnp_random_graph(args.er_n, args.er_p)
csr = nx.to_scipy_sparse_matrix(g, weight=None, format='csr')
graph_io.save_graph(args.out, csr, args.er_n, args.er_n)
def gen_ba(args):
g = nx.barabasi_albert_graph(args.ba_n, args.ba_m)
csr = nx.to_scipy_sparse_matrix(g, weight=None, format='csr')
graph_io.save_graph(args.out, csr, args.er_n, args.er_n)
