def edgelist_in_dimacs_out(ifname):
'''
args: graph is the input nx graph
returns: output filename
'''
g = nx.read_edgelist(ifname, data=False)
g.name = graph_name(ifname)
ofname = '../datasets/{}.dimacs'.format(g.name)
if path.exists(ofname):
return None
n = g.number_of_nodes()
m = g.number_of_edges()
edges = g.edges()
edges = [(int(e[0]), int(e[1])) for e in edges]
df = pd.DataFrame(edges)
df.sort_values(by=[0], inplace=True)
with open(ofname, 'w') as f:
f.write('c {}\n'.format(g.name))
f.write('p edge\t{}\t{}\n'.format(n + 1, m))
output_edges = lambda x: f.write("e\t{}\t{}\n".format(
x[0] + 1, x[1] + 1))
df.apply(output_edges, axis=1)
if path.exists(ofname):
Info("Wrote: %s" % ofname)
# ToDo: a few extra checks could be added
return ofname
def proc_prod_rules_orig(fname):
gn = graph_name(fname)
df = pd.read_csv(fname, header=None, sep="\t")
df['rhs'] = df[2].apply(listify_rhs)
print df['rhs'].apply(lambda x: [k for k in x if 'N' in k]).head()
df['rhs_n'] = df['rhs'].apply(lambda x: len([k for k in x if 'N' in k]))
df['lhs_n'] = df[1].apply(lambda x: len(x.split(",")))
print df.head()
def proc_prod_rules_single(fname):
gn = graph_name(fname)
PRS_dir = "../ProdRules/"
files = glob(PRS_dir + "*{}*prs*".format(gn))
pp.pprint(files)
# print(os.getcwd())
# print
rhs_nonterm_nbrs = lambda RHS: [x for x in RHS if "N" in RHS]
mdf = pd.DataFrame()
for f in files:
df = pd.read_csv(f, header=None, sep="\t")
df['varel'] = (os.path.basename(f).split(".")[2])
df['rhs'] = df[2].apply(listify_rhs)
df['lhs_n'] = df[1].apply(lambda x: len(x.split(",")))
df['rhs_n'] = df['rhs'].apply(
lambda rhs: len([x for x in df['rhs'].values[0] if 'N' in x]))
df['rhs_n'] = df['rhs'].apply(
lambda x: len([k for k in x if 'N' in k]))
# df['rhs_t'] = df['rhs'].apply(lambda rhs: len([x for x in df['rhs'].values[0] if 'N' in x]) )
# # df['rhs_t']
# print df.apply(lambda x: (len(x[1].split(",")), len(x['rhs'])), axis=1)
# '"lhs:", len([x.split(',') for x in df[1]]), "rhs:", len(df['rhs'].values)
mdf = pd.concat([df, mdf])
# print(mdf.head())
# print len(df.loc[0]['rhs'])
# print (df['lhs'].shape)
print df.head()
gb = mdf.groupby('varel').groups
# print (gb['lexm'])
# print (df.describe())
# df.boxplot(ax=xa[1])
# print
# print (mdf.tail())
fig, xa = plt.subplots(1, len(gb.keys()))
# mdf.groupby('varel').hist(ax=xa[0])
# Visualize pairplot of df
# sns.pairplot(mdf, hue='varel');
for j, ve in enumerate(gb.keys()):
print(ve)
# mdf[mdf['varel']==ve][['lhs_n','rhs_n']].hist(ax=xa[0])
# xa[0].histogram(mdf[mdf['varel']==ve].lhs_n)
numBins = 4
xa[j].hist(mdf[mdf['varel'] == ve][['lhs_n', 'rhs_n']],
numBins,
alpha=0.8)
if j == 0: xa[j].legend(('lhs_n', 'rhs_n'))
xa[j].set_title(ve)
# mdf[mdf['varel']==ve].hist(ax=xa[0],x=ve,y=)
# mdf.loc(gb[ve].values).head() #.hist(ax=xa[j],label=ve)
# print (mdf.loc(gb[ve]))
plt.savefig('tmpfig', bbox_inches='tight')
def hstar_fixed_graph_gen(args):
import networkx as nx
orig_fname = args['grow'][0]
gn = graph_name(orig_fname)
if os.path.exists("../datasets/{}.p".format(gn)):
origG = nx.read_gpickle("../datasets/{}.p".format(gn))
else:
print("we load edgelist into an nx.obj")
prs_files = glob("../ProdRules/{}*prs".format(gn))
for f in prs_files:
prod_rules = get_prod_rules(f)
g = pcfg.Grammar('S')
for (id, lhs, rhs, prob) in prod_rules:
# print (id, lhs, rhs, prob)
g.add_rule(pcfg.Rule(id, lhs, rhs, prob))
# exit() # Takes this out
# ToDo: We nee to get these rules in the right format
num_nodes = origG.number_of_nodes()
print "Starting max size"
g.set_max_size(num_nodes)
print "Done with max size"
Hstars = []
num_samples = 20
print '*' * 40
for i in range(0, num_samples):
rule_list = g.sample(num_nodes)
hstar = PHRG.grow(rule_list, g)[0]
Hstars.append(hstar)
import pickle
pickle.dump({
'origG': origG,
'hstars': Hstars
}, open('../Results/{}_hstars.p'.format(gn), "wb"))
if os.path.exists('../Results/{}_hstars.p'.format(gn)):
print("Pickle written")
def edgelist_to_dimacs(fname):
g = nx.read_edgelist(fname, comments="%", data=False, nodetype=int)
g.name = graph_name(fname)
dimacsFiles = convert_nx_gObjs_to_dimacs_gObjs([g])
return dimacsFiles #convert_nx_gObjs_to_dimacs_gObjs([g])
from utils import Info, graph_name
import sys, os
import pprint as pp
from glob import glob
from isomorph_overlap_hl import stack_prod_rules_bygroup_into_list
from prs import proc_prod_rules_orig
results = []
def prs_count_per(prs_lst):
for f in prs_lst:
pp.pprint([os.path.basename(f), len(open(f).readlines())])
if __name__ == '__main__':
if len(sys.argv) < 2:
Info("add an out.* dataset with its full path")
exit()
f = sys.argv[1]
gn = graph_name(f)
f = "../ProdRules/" + gn + "*.prs"
files = glob(f)
prs_cnt_per = prs_count_per(files)
# prs_stack = stack_prod_rules_bygroup_into_list(files)
sys.exit(0)