Spaces:
Runtime error
Runtime error
Commit ·
40cca03
1
Parent(s): ce611b0
Create main.py
Browse files
main.py
ADDED
|
@@ -0,0 +1,181 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import time
|
| 2 |
+
|
| 3 |
+
print("\n\n ==== THE NATURAL LANGUAGE MODULE IS BEING LOADED. PLEASE WAIT ==== \n\n")
|
| 4 |
+
start_time_load = time.time()
|
| 5 |
+
|
| 6 |
+
from transformers import logging
|
| 7 |
+
logging.set_verbosity_error()
|
| 8 |
+
|
| 9 |
+
import warnings
|
| 10 |
+
warnings.filterwarnings("ignore", category=UserWarning)
|
| 11 |
+
|
| 12 |
+
import sys
|
| 13 |
+
|
| 14 |
+
import requests
|
| 15 |
+
|
| 16 |
+
from countriesIdentification import identify_locations
|
| 17 |
+
from datesIdentification import dates_binding
|
| 18 |
+
from magnitudeIdentification import magnitude_binding
|
| 19 |
+
from comparativesIdentification import comparatives_binding
|
| 20 |
+
from earthquaqeIdentification import identify_earthquake_event
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
def process_final_dict(final_dictionary):
|
| 24 |
+
"""
|
| 25 |
+
Function to convert each one of the error codes from each component into a relevant code number to be handled by the SF
|
| 26 |
+
"""
|
| 27 |
+
|
| 28 |
+
# convert all tuple error messages into dictionary error messages
|
| 29 |
+
for i, elem in enumerate(final_dictionary):
|
| 30 |
+
if isinstance(elem, tuple):
|
| 31 |
+
|
| 32 |
+
if elem == (0, "MAGNITUDE", "no_magnitude"):
|
| 33 |
+
final_dictionary[i] = {"Number": 9999911}
|
| 34 |
+
|
| 35 |
+
elif elem == (0, "MAGNITUDE", "more_magnitude"):
|
| 36 |
+
final_dictionary[i] = {"Number": 9999912}
|
| 37 |
+
|
| 38 |
+
elif elem == (0, "MAGNITUDE", "format_error"):
|
| 39 |
+
final_dictionary[i] = {"Number": 9999914}
|
| 40 |
+
|
| 41 |
+
elif elem == (0, "MAGNITUDE", "unknown_error"):
|
| 42 |
+
final_dictionary[i] = {"Number": 9999913}
|
| 43 |
+
|
| 44 |
+
elif elem == (0, "EARTHQUAKE_EVENT", "no_earthquake_reference"):
|
| 45 |
+
final_dictionary[i] = {"event":9999921}
|
| 46 |
+
|
| 47 |
+
elif elem == (0, "EARTHQUAKE_EVENT", "unknown_error"):
|
| 48 |
+
final_dictionary[i] = {"event": 9999922}
|
| 49 |
+
|
| 50 |
+
elif elem == (0,'DATES', 'wrong_date_format'):
|
| 51 |
+
final_dictionary[i] = {"date": {"day": 9999931, "month": 9999931, "year": 9999931}}
|
| 52 |
+
|
| 53 |
+
elif elem == (0,'DATES', 'no_date'):
|
| 54 |
+
final_dictionary[i] = {"date": {"day": 9999932, "month": 9999932, "year": 9999932}}
|
| 55 |
+
|
| 56 |
+
elif elem == (0,'DATES', 'more_dates'):
|
| 57 |
+
final_dictionary[i] = {"date": {"day": 9999933, "month": 9999933, "year": 9999933}}
|
| 58 |
+
|
| 59 |
+
elif elem == (0,'DATES', 'unknown_error'):
|
| 60 |
+
final_dictionary[i] = {"date": {"day": 9999934, "month": 9999934, "year": 9999934}}
|
| 61 |
+
|
| 62 |
+
elif elem == (0, "LOCATION", "no_country"):
|
| 63 |
+
final_dictionary[i] = {"city":[9999941], "country":[9999941]}
|
| 64 |
+
|
| 65 |
+
elif elem == (0, "LOCATION", "more_city_or_country"):
|
| 66 |
+
final_dictionary[i] = {"city": [9999942], "country": [9999942]}
|
| 67 |
+
|
| 68 |
+
elif elem == (0, "LOCATION", "more_country"):
|
| 69 |
+
final_dictionary[i] = {"city": [9999943], "country": [9999943]}
|
| 70 |
+
|
| 71 |
+
elif elem == (0, "LOCATION", "unknown_error"):
|
| 72 |
+
final_dictionary[i] = {"city": [9999944], "country": [9999944]}
|
| 73 |
+
|
| 74 |
+
elif elem == (0, "COMPARATIVES", "more_comparatives_mentions"):
|
| 75 |
+
final_dictionary[i] = {"comparative": 9999951}
|
| 76 |
+
|
| 77 |
+
elif elem == (0, "COMPARATIVES", "no_comparatives"):
|
| 78 |
+
final_dictionary[i] = {"comparative": 9999952}
|
| 79 |
+
|
| 80 |
+
elif elem == (0, "COMPARATIVES", "more_symbol_comparatives"):
|
| 81 |
+
final_dictionary[i] = {"comparative": 9999953}
|
| 82 |
+
|
| 83 |
+
elif elem == (0, "COMPARATIVES", "unknown_error"):
|
| 84 |
+
final_dictionary[i] = {"comparative": 9999954}
|
| 85 |
+
|
| 86 |
+
return final_dictionary
|
| 87 |
+
|
| 88 |
+
|
| 89 |
+
|
| 90 |
+
def natural_language_module(sentence):
|
| 91 |
+
"""
|
| 92 |
+
Function to execute the complete natural language module pipeline
|
| 93 |
+
"""
|
| 94 |
+
try:
|
| 95 |
+
final_dictionary = []
|
| 96 |
+
|
| 97 |
+
# identify whether the sentence is referred on earthquake events
|
| 98 |
+
earth_event = identify_earthquake_event(sentence)
|
| 99 |
+
|
| 100 |
+
if earth_event:
|
| 101 |
+
final_dictionary.append(earth_event)
|
| 102 |
+
|
| 103 |
+
# identify the target country and city in the sentence
|
| 104 |
+
location = identify_locations(sentence)
|
| 105 |
+
|
| 106 |
+
if location:
|
| 107 |
+
final_dictionary.append(location)
|
| 108 |
+
|
| 109 |
+
# identify the target comparative in the sentence
|
| 110 |
+
comparative = comparatives_binding(sentence)
|
| 111 |
+
|
| 112 |
+
if comparative:
|
| 113 |
+
final_dictionary.append(comparative)
|
| 114 |
+
|
| 115 |
+
# identify the target date in the sentence
|
| 116 |
+
date = dates_binding(sentence)
|
| 117 |
+
|
| 118 |
+
if isinstance(date, list):
|
| 119 |
+
|
| 120 |
+
date_dict = date[0]
|
| 121 |
+
date_replc = date[1]
|
| 122 |
+
|
| 123 |
+
if date_dict:
|
| 124 |
+
final_dictionary.append(date_dict[0])
|
| 125 |
+
|
| 126 |
+
# we also delete the date reference from the sentence so that there will
|
| 127 |
+
# not be any confusion with it for the magnitude identification module
|
| 128 |
+
if len(date_replc) == 1:
|
| 129 |
+
sentence = sentence.replace(date_replc[0], " ")
|
| 130 |
+
|
| 131 |
+
# in case it is a tuple we add it as it is and we do not substitute something in the sentence
|
| 132 |
+
elif isinstance(date, tuple):
|
| 133 |
+
final_dictionary.append(date)
|
| 134 |
+
|
| 135 |
+
# identify the target magnitude number in the sentence
|
| 136 |
+
magnitude = magnitude_binding(sentence)
|
| 137 |
+
|
| 138 |
+
if magnitude:
|
| 139 |
+
final_dictionary.append(magnitude)
|
| 140 |
+
|
| 141 |
+
clean_final_dictionary = process_final_dict(final_dictionary)
|
| 142 |
+
|
| 143 |
+
result = {}
|
| 144 |
+
for d in clean_final_dictionary:
|
| 145 |
+
result.update(d)
|
| 146 |
+
|
| 147 |
+
return result
|
| 148 |
+
|
| 149 |
+
except:
|
| 150 |
+
return "\n\n=== AN UNEXPECTED ERROR HAS OCCURED. PLEASE EXECUTE AGAIN THE SCRIPT OR COMMUNICATE WITH THE DEVELOPER TEAM === \n\n"
|
| 151 |
+
|
| 152 |
+
|
| 153 |
+
|
| 154 |
+
def process_json_sf(nl_json, sentence):
|
| 155 |
+
"""
|
| 156 |
+
Function to conver the captured information an a relevant json format
|
| 157 |
+
"""
|
| 158 |
+
try:
|
| 159 |
+
sf_json_format = {
|
| 160 |
+
"text": sentence,
|
| 161 |
+
"page": "1",
|
| 162 |
+
"nlp": {"event": nl_json['event'], "city": nl_json['city'][0], "country": nl_json['country'][0], "year": int(nl_json['date']['year']), "month": int(nl_json['date']['month']),
|
| 163 |
+
"day": int(nl_json['date']['day']), "magnitude": nl_json['Number'], "comparative": nl_json['comparative'], "point": False, "latitude": None,"lognitude": None}
|
| 164 |
+
}
|
| 165 |
+
|
| 166 |
+
return sf_json_format
|
| 167 |
+
|
| 168 |
+
except:
|
| 169 |
+
return "\n\n=== AN UNEXPECTED ERROR HAS OCCURED. PLEASE EXECUTE AGAIN THE SCRIPT OR COMMUNICATE WITH THE DEVELOPER TEAM === \n\n"
|
| 170 |
+
|
| 171 |
+
|
| 172 |
+
|
| 173 |
+
def main(sentence):
|
| 174 |
+
"""
|
| 175 |
+
Function to bind together all the info and be executed
|
| 176 |
+
"""
|
| 177 |
+
|
| 178 |
+
nl_data = natural_language_module(sentence)
|
| 179 |
+
nl_json = process_json_sf(nl_data, sentence)
|
| 180 |
+
|
| 181 |
+
return nl_json
|