def serviceMode():
# Initialise the AI
ai = AI()
ai.initialise("")
# Initialise the pre-processor
preProcessor = DataPreProcessor("")
# Create a previous input to track new questions
pInput = ""
# The program will only exit when it recieves a deliberate 'exit' message
loop = True
while loop:
# Check if the file exists
path = str(os.getcwd()) + "/input.txt"
if os.path.isfile(path):
# Read our input from the file
input_file = open("input.txt", "r")
# Clean our input
preProcessor.processInput(input_file.read())
input = str(preProcessor.input.strip())
# Exit on 'exit'
if input == "exit":
loop = False
elif input != pInput:
pInput = input
# Write the the output back to a different file
output_file = open("output.txt", "w")
output = arrangeResp(input, preProcessor, ai)
output_file.write(output)
output_file.close()
def __init__(self):
# preProcessor is only utilised here to avoid duplicating string2array
self.preProcessor = DataPreProcessor("")
self.model_ = Seq2seq(
decoder_seq_length=self.decoder_seq_length,
cell_enc=tf.keras.layers.GRUCell,
cell_dec=tf.keras.layers.GRUCell,
n_layer=3,
n_units=256,
embedding_layer=tl.layers.Embedding(
vocabulary_size=self.vocabulary_size,
embedding_size=self.emb_dim),
)
load_weights = tl.files.load_npz(name="data/model.npz")
tl.files.assign_weights(load_weights, self.model_)
def arrangeResp(userInput, personality):
# Creating our AI.
ai = AI(personality.name)
# Where output is the processed output.
preProcessor = DataPreProcessor(userInput)
# Was the query too short for the AI, if so, exit.
if preProcessor.processInput() == False:
return False
# This is passed to the model.
processedInput = preProcessor.input
# Receive response from the model.
response = ai.respond(processedInput, 1)
response = preProcessor.array2String(response)
postProcessor = DataPostProcessor(response, personality)
postProcessor.postProcess()
return response
def shortenWords(self, input):
shortWords = {
'your': 'ur',
'tonight': '2nite',
'for your information': 'fyi',
'okay': 'k',
'ok': 'k',
'to': '2',
'tomorrow': '2moro',
'easy': 'ez',
'see you': 'cya',
'because': 'cuz',
'you': 'u',
}
preProcessor = DataPreProcessor(input)
input = preProcessor.string2Array(input)
for x in range(0, len(input)):
for y in shortWords:
if (str(input[x]) == y):
input[x] = shortWords[y]
input = preProcessor.array2String(input)
return input
def shortenWords(self, input):
shortWords = {
'for your information': 'fyi',
'okay': 'kk',
'ok': 'kk',
'easy': 'ez',
'see you': 'cya',
'because': 'cuz',
'very': 'v',
'to be honest': 'tbh',
'oh my god': 'omg',
'boyfriend': 'bf',
'girlfriend': 'gf',
'awesome': 'cool',
'in trouble': 'screwed',
}
preProcessor = DataPreProcessor(input)
input = preProcessor.string2Array(input)
for x in range(0, len(input)):
for y in shortWords:
if (str(input[x]) == y):
input[x] = shortWords[y]
input = preProcessor.array2String(input)
return input
class AI:
metadata, idx_q, idx_a = data.load_data(PATH='data/{}/'.format("twitter"))
src_vocab_size = len(metadata['idx2w'])
emb_dim = 1024
word2idx = metadata['w2idx']
idx2word = metadata['idx2w']
unk_id = word2idx['unk']
pad_id = word2idx['_']
start_id = src_vocab_size
end_id = src_vocab_size + 1
word2idx.update({'start_id': start_id})
word2idx.update({'end_id': end_id})
idx2word = idx2word + ['start_id', 'end_id']
src_vocab_size = src_vocab_size + 2
vocabulary_size = src_vocab_size
decoder_seq_length = 20
# Creates an instance of the AI with it's name being passed from a name-generating function
def __init__(self):
# preProcessor is only utilised here to avoid duplicating string2array
self.preProcessor = DataPreProcessor("")
self.model_ = Seq2seq(
decoder_seq_length=self.decoder_seq_length,
cell_enc=tf.keras.layers.GRUCell,
cell_dec=tf.keras.layers.GRUCell,
n_layer=3,
n_units=256,
embedding_layer=tl.layers.Embedding(
vocabulary_size=self.vocabulary_size,
embedding_size=self.emb_dim),
)
load_weights = tl.files.load_npz(name="data/model.npz")
tl.files.assign_weights(load_weights, self.model_)
# Take a name, create a mood
def initialise(self, name):
self.name = name
feelings = [
"good", "well", "great", "grand", "excellent", "ecstatic", "happy",
"sad", "annoyed", "frustrated", "angry", "tired", "okay", "alright"
]
self.feel = feelings[randint(0, len(feelings) - 1)]
# Handle the creation of a response from the given input
def respond(self, seed, number):
simpleStart = self.simpleResponse(seed)
self.model_.eval()
seed_id = [self.word2idx.get(w, self.unk_id) for w in seed.split(" ")]
sentence_id = self.model_(inputs=[[seed_id]],
seq_length=20,
start_token=self.start_id,
top_n=number)
sentence = []
for w_id in sentence_id[0]:
w = self.idx2word[w_id]
if w == 'end_id':
break
sentence = sentence + [w]
# A catch all just in case there are no responses, but we have yet to find an input to trigger this
if sentence == []:
sentence = [
"I'm", "sorry,", "I", "just", "don't", "quite", "understand",
"what", "you're", "asking..."
]
return simpleStart + sentence
# Handle simple questions that the AI is less than optimal at answering
def simpleResponse(self, input):
sentence = []
input = self.preProcessor.string2Array(input)
tally = [0, 0, 0]
greetings = [
"hello", "hi", "greetings", "salutations", "hey", "yo", "howdy"
]
names = [["what", "who"], ["is", "are"], ["you", "your"], ["name"]]
wellbeing = [["how"], ["do", "are"], ["you"],
["doing", "feeling", "feel"]]
# Tallying key words in the user query to determine if certain questions were being asked
for x in input:
for y in range(len(greetings)):
if x == greetings[y]:
tally[0] = 1
break
for y in range(len(names)):
for z in names[y]:
if x is z:
tally[1] = tally[1] + 1
break
for y in range(len(wellbeing)):
for z in wellbeing[y]:
if x == z:
tally[2] = tally[2] + 1
break
# Handle a return greeting, and maybe ask how the user is
if tally[0] > 0:
sentence.append(greetings[randint(0, 6)])
if randint(0, 1) is 1:
sentence.append("how")
sentence.append("are")
sentence.append("you")
value = randint(0, 2)
if value is 0:
sentence.append("doing")
elif value is 1:
sentence.append("feeling")
# Handle questions about it's name with a simple answer
if tally[1] > 2 and len(input) < 5:
if randint(0, 1) is 1:
sentence.append("I")
sentence.append("am")
else:
sentence.append("my")
sentence.append("name")
sentence.append("is")
sentence.append(self.name)
# Handle a 'how are you' type question with a pre-determined emotional state
if tally[2] > 2 and len(input) < 5:
sentence.append("I")
sentence.append("am")
if randint(0, 1) is 1:
sentence.append("feeling")
sentence.append(self.feel)
return sentence
def terminalMode():
# Simple bit of beautifying for our command-line output
c_background = "\033[44m"
c_blue = "\033[94m"
c_green = "\033[92m"
c_red = "\033[91m"
c_close = "\033[0m"
# Set up our timer
timer = BackgroundScheduler({"apscheduler.timezone": "Europe/Dublin"})
accepted_timer = False
# Small bit of starting text to introduce the user
startingText = [
"Ask a question, and see how our AI responds.",
"Please ensure queries have more than three words.",
"You will have 60 seconds upon entering your first query to make further queries.",
"The program will wait for your last query before exiting."
]
print(c_green + "\n".join(startingText) + c_close)
# The timer feature is optional over the command line because it will be implemented by the UI team
print(c_background + "Do you wish to enable the on-minute timer? (y/n)" +
c_close)
userInput = input()
if userInput == "y" or userInput == "Y":
timer.add_job(endProgram, "interval", minutes=1)
accepted_timer = True
print(c_background + "Enter the input, or enter 'exit' to end:" + c_close)
personality = None
# Retrieve a name
name = naming()
# Personality is randomly assigned between teenager, young adult and adult with a random name each time
perNum = random.randint(1, 3)
if perNum == 1:
personality = Teenager(name)
elif perNum == 2:
personality = YoungAdult(name)
else:
personality = Adult(name)
# Initialise the ai
ai = AI()
ai.initialise(name)
# Initialise the pre-processor
preProcessor = DataPreProcessor("")
# Initialise the post-processor with the personality
postProcessor = DataPostProcessor(personality)
# Our input/feedback loop, starting our timer on its initial run
initial = True
global end
while end is False:
# Start the timer after the first user input if it was accepted
if initial and accepted_timer:
timer.start()
elif not initial:
print(c_background + "Enter another input:" + c_close)
initial = False
# Take the user input
userInput = input()
# Exit on 'exit'
if (userInput == "exit"):
break
# Ensure input string is long enough if the terminal is being used
if preProcessor.processInput(userInput) is not False:
# Determine our input - whether an error or a valid query
output = arrangeResp(preProcessor.input, preProcessor, ai)
# Post-processing of data
output = postProcessor.postProcess(output)
# Provide result to the user
print(c_blue + output + c_close)
else:
# Handle queries that are too short
print(c_red +
"The input string must have a minimum of three words!" +
c_close)
# Terminate the program
print(c_green + "Program exited." + c_close)
def test_ConvertChar(self):
testInput = "æ"
preProcessor = DataPreProcessor(testInput)
preProcessor.convertAccentedCharsToAscii()
testInput = preProcessor.input
self.assertEqual(testInput, "a")
testInput = "je ètais"
preProcessor = DataPreProcessor(testInput)
preProcessor.convertAccentedCharsToAscii()
testInput = preProcessor.input
self.assertEqual(testInput, "je etais")
testInput = "í"
preProcessor = DataPreProcessor(testInput)
preProcessor.convertAccentedCharsToAscii()
testInput = preProcessor.input
self.assertEqual(testInput, "i")
testInput = "Ø"
preProcessor = DataPreProcessor(testInput)
preProcessor.convertAccentedCharsToAscii()
testInput = preProcessor.input
self.assertEqual(testInput,"o")
testInput = "Ùaemd"
preProcessor = DataPreProcessor(testInput)
preProcessor.convertAccentedCharsToAscii()
testInput = preProcessor.input
self.assertEqual(testInput, "uaemd")
testInput = "ůűų or ŷÝŸ"
preProcessor = DataPreProcessor(testInput)
preProcessor.convertAccentedCharsToAscii()
testInput = preProcessor.input
self.assertEqual(testInput,"uuu or yyy")
def test_NumWordToDigit(self):
testInput = "fsd"
preProcessor = DataPreProcessor(testInput)
preProcessor.removeNumWords()
testInput = preProcessor.input
self.assertEqual(testInput, "fsd")
testInput = "one.two"
preProcessor = DataPreProcessor(testInput)
preProcessor.removeNumWords()
testInput = preProcessor.input
self.assertEqual(testInput, "one.two")
testInput = "one"
preProcessor = DataPreProcessor(testInput)
preProcessor.removeNumWords()
testInput = preProcessor.input
self.assertEqual(testInput, "")
testInput = "one two"
preProcessor = DataPreProcessor(testInput)
preProcessor.removeNumWords()
testInput = preProcessor.input
self.assertEqual(testInput," ")
testInput = "twelve"
preProcessor = DataPreProcessor(testInput)
preProcessor.removeNumWords()
testInput = preProcessor.input
self.assertEqual(testInput, "")
testInput = "one or three"
preProcessor = DataPreProcessor(testInput)
preProcessor.removeNumWords()
testInput = preProcessor.input
self.assertEqual(testInput," or ")
def test_ConvertChar(self):
testInput = 'æ'
preProcessor = DataPreProcessor(testInput)
preProcessor.convertAccentedCharsToAscii()
testInput = preProcessor.input
self.assertEqual(testInput, 'a')
testInput = 'je ètais'
preProcessor = DataPreProcessor(testInput)
preProcessor.convertAccentedCharsToAscii()
testInput = preProcessor.input
self.assertEqual(testInput, 'je etais')
testInput = 'í'
preProcessor = DataPreProcessor(testInput)
preProcessor.convertAccentedCharsToAscii()
testInput = preProcessor.input
self.assertEqual(testInput, 'i')
testInput = 'Ø'
preProcessor = DataPreProcessor(testInput)
preProcessor.convertAccentedCharsToAscii()
testInput = preProcessor.input
self.assertEqual(testInput,'o')
testInput = 'Ùaemd'
preProcessor = DataPreProcessor(testInput)
preProcessor.convertAccentedCharsToAscii()
testInput = preProcessor.input
self.assertEqual(testInput, 'uaemd')
testInput = 'ůűų or ŷÝŸ'
preProcessor = DataPreProcessor(testInput)
preProcessor.convertAccentedCharsToAscii()
testInput = preProcessor.input
self.assertEqual(testInput,'uuu or yyy')
def test_NumWordToDigit(self):
testInput = 'fsd'
preProcessor = DataPreProcessor(testInput)
preProcessor.convertNumberWordToDigit()
testInput = preProcessor.input
self.assertEqual(testInput, 'fsd')
# FAILS THIS TEST
testInput = 'one.two'
preProcessor = DataPreProcessor(testInput)
preProcessor.convertNumberWordToDigit()
testInput = preProcessor.input
print (f"{testInput} : testing input")
self.assertEqual(testInput, '1.2')
testInput = 'one'
preProcessor = DataPreProcessor(testInput)
preProcessor.convertNumberWordToDigit()
testInput = preProcessor.input
self.assertEqual(testInput, '1')
# FAILS THIS TEST
testInput = 'one two'
preProcessor = DataPreProcessor(testInput)
preProcessor.convertNumberWordToDigit()
testInput = preProcessor.input
self.assertEqual(testInput,'1 2')
testInput = 'twelve'
preProcessor = DataPreProcessor(testInput)
preProcessor.convertNumberWordToDigit()
testInput = preProcessor.input
self.assertEqual(testInput, '12')
testInput = 'one or three'
preProcessor = DataPreProcessor(testInput)
preProcessor.convertNumberWordToDigit()
testInput = preProcessor.input
self.assertEqual(testInput,'1 or 3')