def instantiate_networkdata_class(dataset_domain, dataset_path, impl_type,
max_core_nodes, cutoff_rating, store_dataset,
interact_type_val, min_interacts_beforeaftersplit_per_user):
data = None
#h = hpy()
#h.setref()
if dataset_domain == "twitter":
data = HashtagDataPreparser(dataset_path, impl_type)
elif dataset_domain== "lastfm":
data = LastfmDataPreparserCSV(dataset_path, impl_type, cutoff_rating,
max_core_nodes, store_dataset, use_artists=False)
elif dataset_domain== "lastfm_simple":
data = LastfmDataPreparserSimple(dataset_path, impl_type, cutoff_rating,
max_core_nodes, store_dataset, use_artists=False,
interact_type_val=interact_type_val,
min_interactions_per_user=min_interacts_beforeaftersplit_per_user*2)
elif dataset_domain== "lastfm_lovelisten":
data = LastfmDataPreparserLovelisten(dataset_path, impl_type, cutoff_rating,
max_core_nodes, store_dataset, use_artists=False,
interact_type_val=interact_type_val,
min_interactions_per_user=min_interacts_beforeaftersplit_per_user*2)
elif dataset_domain=="goodreads":
data = GoodreadsDataPreparser(dataset_path, impl_type, cutoff_rating,
max_core_nodes, store_dataset,
min_interactions_per_user = min_interacts_beforeaftersplit_per_user*2)
elif dataset_domain=="flixster":
data = FlixsterDataPreparser(dataset_path, impl_type, cutoff_rating,
max_core_nodes, store_dataset,
min_interactions_per_user=min_interacts_beforeaftersplit_per_user*2)
elif dataset_domain=="flickr":
data = FlickrDataPreparser(dataset_path, impl_type, cutoff_rating,
max_core_nodes, store_dataset,
min_interactions_per_user=min_interacts_beforeaftersplit_per_user*2)
try:
data.get_all_data()
BasicNetworkAnalyzer(data).show_basic_stats()
except:
raise
return data
node = self.netdata.nodes[node_id]
node.get_details(self.netdata.interaction_types[0])
return
"""
def getItemPopularityInDataset(data):
likes = {}
for k, v in data.allusers.iteritems():
for itemid, created in v.likes:
if itemid not in likes:
likes[itemid] = 0
likes[itemid] += 1
likes_hist={}
in_sum = 0
for val in likes.values():
if val<100:
in_sum += 1
if val not in likes_hist:
likes_hist[val] = 0
likes_hist[val] += 1
items_covered_ratio = in_sum /float(len(likes))
return list(likes_hist.iteritems()), items_covered_ratio, likes.values()
"""
if __name__ == "__main__":
data = HashtagDataPreparser("/home/asharma/datasets/ttest/")
data.get_all_data()
net_analyzer = BasicNetworkAnalyzer(data)
net_analyzer.show_basic_stats()
def instantiate_networkdata_class(dataset_domain, dataset_path, impl_type,
max_core_nodes, cutoff_rating, store_dataset,
interact_type_val,
min_interacts_beforeaftersplit_per_user):
data = None
#h = hpy()
#h.setref()
if dataset_domain == "twitter":
data = HashtagDataPreparser(dataset_path, impl_type)
elif dataset_domain == "lastfm":
data = LastfmDataPreparserCSV(dataset_path,
impl_type,
cutoff_rating,
max_core_nodes,
store_dataset,
use_artists=False)
elif dataset_domain == "lastfm_simple":
data = LastfmDataPreparserSimple(
dataset_path,
impl_type,
cutoff_rating,
max_core_nodes,
store_dataset,
use_artists=False,
interact_type_val=interact_type_val,
min_interactions_per_user=min_interacts_beforeaftersplit_per_user *
2)
elif dataset_domain == "lastfm_lovelisten":
data = LastfmDataPreparserLovelisten(
dataset_path,
impl_type,
cutoff_rating,
max_core_nodes,
store_dataset,
use_artists=False,
interact_type_val=interact_type_val,
min_interactions_per_user=min_interacts_beforeaftersplit_per_user *
2)
elif dataset_domain == "goodreads":
data = GoodreadsDataPreparser(
dataset_path,
impl_type,
cutoff_rating,
max_core_nodes,
store_dataset,
min_interactions_per_user=min_interacts_beforeaftersplit_per_user *
2)
elif dataset_domain == "flixster":
data = FlixsterDataPreparser(
dataset_path,
impl_type,
cutoff_rating,
max_core_nodes,
store_dataset,
min_interactions_per_user=min_interacts_beforeaftersplit_per_user *
2)
elif dataset_domain == "flickr":
data = FlickrDataPreparser(
dataset_path,
impl_type,
cutoff_rating,
max_core_nodes,
store_dataset,
min_interactions_per_user=min_interacts_beforeaftersplit_per_user *
2)
try:
data.get_all_data()
BasicNetworkAnalyzer(data).show_basic_stats()
except:
raise
return data
