def mainController(cursor):
# FileExporter is preparing a file for write
exporter = predictionio.FileExporter(file_name="my_events.json")
jsonSetItem(cursor, exporter)
jsonSetUser(cursor, exporter)
jsonUserViewItem(cursor, exporter)
jsonUserBuyItem(cursor, exporter)
# close the FileExporter when finish writing all events
exporter.close()
def test_export(self):
app_info = subprocess.check_output(['pio', 'app', 'show', app_name])
app_id = re.search('App ID: ([0-9]+)', app_info.decode('utf-8')).group(1)
print(app_id)
exporter = predictionio.FileExporter(file_name=filename)
first_event_properties = {
"prop1" : 1,
"prop2" : "value2",
"prop3" : [1, 2, 3],
"prop4" : True,
"prop5" : ["a", "b", "c"],
"prop6" : 4.56 ,
}
first_event_time = datetime(
2004, 12, 13, 21, 39, 45, 618000, pytz.timezone('US/Mountain'))
exporter.create_event(
event="my_event",
entity_type="user",
entity_id="uid",
properties=first_event_properties,
event_time=first_event_time,
)
# Second event
second_event_properties = {
"someProperty" : "value1",
"anotherProperty" : "value2",
}
exporter.create_event(
event="my_event",
entity_type="user",
entity_id="uid",
target_entity_type="item",
target_entity_id="iid",
properties=second_event_properties,
event_time=datetime(2014, 12, 13, 21, 38, 45, 618000, pytz.utc))
exporter.close()
subprocess.call(['pio', 'import', '--appid', app_id, '--input ', filename])
sys.exit(1)
logger.info(f'Accessing an access key')
appid_cmd = f'{args.pio_cmd} app list'
app_id = get_app_id(args)
if len(app_id) == 0:
logger.error('app_id is empty.')
sys.exit(1)
csv_file = f'{args.input_dir}/{args.csv_name}.csv'
if not os.path.exists(csv_file):
logger.error(f"{csv_file} is not found")
sys.exit(1)
logger.info(f'Loading {args.csv_name}.csv')
exporter = predictionio.FileExporter(
file_name=f'/tmp/{args.csv_name}.ndjson')
df = pd.read_csv(csv_file, encoding="utf-8")
for i, v in df.iterrows():
event_response = exporter.create_event(
event=f"{args.csv_name}",
entity_type='data',
entity_id=str(i),
properties={x: str(v[x])
for x in df.columns})
exporter.close()
logger.info(f'Storing /tmp/{args.csv_name}.ndjson')
subprocess.run(
f'{args.pio_cmd} import --appid {app_id} --input /tmp/{args.csv_name}.ndjson'
.split(' '))
return (x, y)
def add_event(exporter, id, x, y):
# Build event properties
props = {}
for i, xi in enumerate(x):
props['x' + str(i)] = xi
props['y'] = y
exporter.create_event(event="$set",
entity_type="point",
entity_id=id,
properties=props)
if __name__ == '__main__':
exporter = predictionio.FileExporter(
file_name=EXPORTED_EVENTS_JSON_FILE_NAME)
train_df = pd.read_csv(KAGGLE_HOUSE_PRICES_CSV_FILE_PATH)
for (_, row) in train_df.iterrows():
id = get_row_id(row)
(x, y) = row_to_example(row)
add_event(exporter, id, x, y)
print('Exported event data to {}'.format(EXPORTED_EVENTS_JSON_FILE_NAME))
exporter.close()
"""
Import sample data for recommendation engine
"""
import predictionio
from datetime import datetime
import pytz
import ujson
import mysql.connector
if __name__ == '__main__':
exporter = predictionio.FileExporter(file_name="set_events.json")
cnx = mysql.connector.connect(user='******',
password='******',
host='127.0.0.1',
database='ren')
cursor = cnx.cursor()
query = ("SELECT hashid, info, url from product")
cursor.execute(query)
count = 0
for (hashid, info, url) in cursor:
info_string = ujson.loads(info)
eid = info_string['Code']
del info_string['Code']
for key in info_string.keys():
info_string[key] = [info_string[key]]
event_properties = info_string
def new_exporter(file_name):
return predictionio.FileExporter(file_name=file_name)
"""
Import sample data for recommendation engine
"""
import random
import predictionio
from datetime import datetime
import pytz
import ujson
import mysql.connector
if __name__ == '__main__':
exporter = predictionio.FileExporter(file_name="hp_events.json")
hp_list = [
"9788498387568", "9788498383621", "9788700631625", "9780605039070",
"9788884516374", "9780195799163"
]
usr_hp = [1, 2, 3, 4, 5]
count = 0
for x in xrange(1, 100):
exporter.create_event(
event="view",
entity_type="user",
entity_id="l5;" + str(random.choice(usr_hp)),
target_entity_type="item",
target_entity_id="l5;" + random.choice(hp_list),
event_time=datetime.utcnow().replace(tzinfo=pytz.utc))
count = count + 1
for x in xrange(1, 20):
exporter.create_event(
def _lazy_get_exporter_func(self, **kwargs):
"""kwargs must contain filename"""
if self.exporter is None:
self.filename = kwargs['filename']
self.exporter = predictionio.FileExporter(file_name=self.filename)
return self.exporter.create_event
target_entity_type="item",
target_entity_id=phrase,
event_time=current_date)
#print "Event: " + SEARCH_EVENT + " entity_id: " + user_id + " target_entity_id: " + phrase + \
# " current_date: " + current_date.isoformat()
current_date += event_time_increment
count += 1
f.close()
print "%s lines are imported." % count
if __name__ == '__main__':
parser = argparse.ArgumentParser(
description="Import sample data for recommendation engine")
parser.add_argument('--access_key', default='invald_access_key')
parser.add_argument('--url', default="http://localhost:7070")
parser.add_argument('--file', default="./data/sample-handmade-data.txt")
parser.add_argument('--output', default="./data/ps-out.json")
args = parser.parse_args()
print args
# client = predictionio.EventClient(
# access_key=args.access_key,
# url=args.url,
# threads=5,
# qsize=500
# )
client = predictionio.FileExporter(args.output)
import_events(client, args.file)
import predictionio
from datetime import datetime
import pytz
exporter = predictionio.FileExporter(file_name="test.json")
first_event_properties = {
"prop1": 1,
"prop2": "value2",
"prop3": [1, 2, 3],
"prop4": True,
"prop5": ["a", "b", "c"],
"prop6": 4.56,
}
first_event_time = datetime(2004, 12, 13, 21, 39, 45, 618000,
pytz.timezone('US/Mountain'))
exporter.create_event(
event="my_event",
entity_type="user",
entity_id="uid",
properties=first_event_properties,
event_time=first_event_time,
)
# Second event
second_event_properties = {
"someProperty": "value1",
"anotherProperty": "value2",
}
exporter.create_event(event="my_event",
entity_type="user",