class SpeechProcessor:
def __init__(self):
self.rec_obj = sr.Recognizer()
self.tracer_obj = DialogTracer(True)
def _get_audio_from_mic(self, timeout = 10):
with ignore_stderr():
with sr.Microphone() as source:
audio = self.rec_obj.listen(source, timeout=timeout)
return audio
def _google_speech_recognizer(self, audio):
result = ''
try:
result = self.rec_obj.recognize_google(audio)
except sr.UnknownValueError:
self.tracer_obj.sys_msg("Speech Recognition module could not understand audio")
except sr.RequestError as e:
self.tracer_obj.sys_msg("Could not request results from Speech Recognition service; {0}".format(e))
return result
def _google_cloud_speech_recognizer(self, audio):
result = ''
try:
result = self.rec_obj.recognize_google_cloud(audio, credentials_json=GOOGLE_CLOUD_SPEECH_CREDENTIALS)
except sr.UnknownValueError:
self.tracer_obj.sys_msg("Speech Recognition module could not understand audio")
except sr.RequestError as e:
self.tracer_obj.sys_msg("Could not request results from Speech Recognition service; {0}".format(e))
return result
def audio_to_text(self):
speech_audio = self._get_audio_from_mic()
google_cloud_result = self._google_cloud_speech_recognizer(speech_audio)
return google_cloud_result
def __init__(self):
self.dialog_tracer_obj = DialogTracer(True)
print("\n****************************************")
print("*******Speech Based Dialog System*******")
print("****************************************\n")
self._initialize_dialog_manager()
self.user_name = ''
self.food_items_tokens = ''
self.final_usda_food_items = ''
def __init__(self):
self.rec_obj = sr.Recognizer()
self.tracer_obj = DialogTracer(True)
def __init__(self):
self.category_dict = {}
self._load_category_map()
self.tracer_obj = DialogTracer(True)
class DB:
def __init__(self):
self.category_dict = {}
self._load_category_map()
self.tracer_obj = DialogTracer(True)
# Public methods
def get_available_categories_list(self):
self.tracer_obj.sys_msg("The following food categories are available..")
print("********")
for category in self.category_dict:
print(category)
print("********")
def update_db_full(self):
self.tracer_obj.sys_msg("Performing complete DB update..")
for category in self.category_dict:
self.update_db_category(category)
self.tracer_obj.sys_msg("Full DB update complete.")
def update_db_category(self, category):
category = str(category).lower()
if str(category).lower() in self.category_dict:
self.tracer_obj.sys_msg("Starting update..")
print("CATEGORY: {0}".format(category))
print("TARGET URL: {0}".format(self.category_dict[category]))
print("This may take some time, stay tuned...")
total = self._get_search_results_count(self.category_dict[category])
print("total : ", total)
food_pages = self._fetch_calorie_links(int(total[0]), self.category_dict[category])
script_path = os.path.dirname(os.path.realpath(sys.argv[0]))
execution_path = os.getcwd()
os.makedirs('{0}/data/'.format(script_path) + category, exist_ok=True)
os.chdir('{0}/data/'.format(script_path) + category)
self._download_csv(food_pages, total)
os.chdir(execution_path)
self.tracer_obj.sys_msg("DB update completed, for category: {0}".format(category))
else:
print("[ERROR]: Improper category specified, update failed!")
def update_csv_objects_to_db(self):
self.tracer_obj.sys_msg('Begin dumping csv objects to db..')
script_path = os.path.dirname(os.path.realpath(sys.argv[0]))
execution_path = os.getcwd()
os.chdir(script_path + '/data')
os.makedirs('csv_objects', exist_ok=True)
print('Dumping csv objects at: {0}'.format(script_path + '/data/csv_objects/'))
for category in self.category_dict:
csv_file_names = glob(category + '/*.csv')
print(len(csv_file_names))
ctr = 0
for csv_file in csv_file_names:
with open(script_path + '/data/csv_objects/' +
csv_file.split('/')[1].split('.')[0] + '.pkl', 'wb') as outfile:
print('Dumping csv object to file: {0}'.format(script_path + '/data/csv_objects/' +
csv_file.split('/')[1].split('.')[0] + '.pkl'))
csv_obj = CSVData(script_path + '/data/' + csv_file)
ctr += 1
print('Completed processing {0}/{1} files in category {2} ..'
.format(ctr, len(csv_file_names), category))
pickle.dump(csv_obj, outfile, protocol=pickle.HIGHEST_PROTOCOL)
self.tracer_obj.sys_msg('CSV object update complete..')
os.chdir(execution_path)
def get_csv_objects_list(self):
self.tracer_obj.sys_msg("Loading csv objects to memory..")
script_path = os.path.dirname(os.path.realpath(sys.argv[0]))
execution_path = os.getcwd()
os.chdir(script_path + '/data/csv_objects')
objects_list = []
pickles = glob(script_path + '/data/csv_objects' + '/*.pkl')
for pkl in pickles:
with open(pkl, 'rb') as infile:
objects_list.append(pickle.load(infile))
self.tracer_obj.sys_msg("CSV objects load to list complete, total: {0}".format(len(objects_list)))
os.chdir(execution_path)
return objects_list
# Protected methods follow
def _load_category_map(self):
try:
with open('{0}/data/category_maps.txt'.format(os.path.dirname(os.path.realpath(sys.argv[0]))), 'r') as data:
map_data = data.read()
map_data = map_data.split('\n')
for line in map_data:
self.category_dict[line.split('::')[0].lower()] = line.split('::')[1]
except EnvironmentError:
print("Could not load category map, abort!")
sys.exit(2)
def _get_search_results_count(self, page_url):
conn = urlopen(page_url)
html = conn.read()
soup = BeautifulSoup(html, "lxml")
div = soup.find("div", {"class": "alert alert-info result-message"})
return re.findall(r'\d+', str(div))
def _fetch_calorie_links(self, total, page_url):
calorie_links = []
append_flag = True
root_page_nav = self._get_root_paginateURL(page_url)
for page_index in range(0, int(total / 50) + 1):
page_results = page_index * 50
page = root_page_nav + str(page_results) + "&order=asc"
page = page.replace("&", "&")
conn = urlopen(page)
html = conn.read()
soup = BeautifulSoup(html, "lxml")
links = soup.find_all('a')
for tag in links:
link = tag.get('href', None)
if link is not None:
if "/ndb/foods/show/" not in link:
continue
if append_flag:
calorie_links.append(link)
append_flag = False
else:
append_flag = True
return calorie_links
def _get_root_paginateURL(self, page_url):
conn = urlopen(page_url)
html = conn.read()
soup = BeautifulSoup(html, "lxml")
div = soup.find("a", {"class": "step"})
return "https://ndb.nal.usda.gov" + str(div).split("href=\"")[1].split('>')[0].split("offset=")[0] + "offset="
def _download_csv(self, food_pages, total_csv):
csv_ct = 1
for link in food_pages:
conn = urlopen("https://ndb.nal.usda.gov" + link)
html = conn.read()
soup = BeautifulSoup(html, "lxml")
csv_tag = soup.find("a", {"title": "Download As CSV"})
csv_link = str(csv_tag).split("href=")[1].split(" ")[0]
if csv_link.startswith('"') and csv_link.endswith('"'):
csv_link = csv_link[1:-1]
div = soup.find("div", {"id": "view-name"})
filename = slugify(str(div).split(',', 1)[1].split('\n')[0]) + ".csv"
print("********")
print("Retrieving CSV: {0} of {1}".format(csv_ct, int(total_csv[0])))
csv_ct += 1
print(filename, "https://ndb.nal.usda.gov" + csv_link)
if Path(filename).is_file():
print("File already exists at the destination, skip retrieval..")
else:
print("Retrieving file..")
urlretrieve("https://ndb.nal.usda.gov" + csv_link.replace("&", "&"), filename)
print("Successfully retrieved..")
print("********")
def __init__(self, complete_db_overhaul=False):
self.dialog_tracer_obj = DialogTracer(True)
self.task_manager_obj = TaskManager(complete_db_overhaul)
self.dialog_tracer_obj.sys_msg("Dialog Manager setup complete..")
class DialogManager:
def __init__(self, complete_db_overhaul=False):
self.dialog_tracer_obj = DialogTracer(True)
self.task_manager_obj = TaskManager(complete_db_overhaul)
self.dialog_tracer_obj.sys_msg("Dialog Manager setup complete..")
# Based on Food Tokens narrow the Food item and return the List
def get_all_usda_food_items_from_user(self, food_items_tokens):
self.task_manager_obj.text_to_audio(
'Let me get more information from you about these items to figure out '
'your exact intake')
result_list = []
experience = 1
counter = 1
counter_dict = {
1: 'FIRST',
2: 'SECOND',
3: 'THIRD',
4: 'FOURTH',
5: 'FIFTH'
}
while len(food_items_tokens) != 0:
# S = "======================== SELECTION OF ",counter_dict[counter]," FOOD ITEM: BEGIN ======================"
self.dialog_tracer_obj.sys_msg(
"======================== SELECTION OF " +
counter_dict[counter] +
" FOOD ITEM: BEGIN ======================")
count = self._get_current_food_item_from_user(
food_items_tokens, counter)
counter += 1
result = self.get_standard_item(food_items_tokens[:count],
experience)
experience += 1
del food_items_tokens[:count]
if result != "null":
result_list.append(result)
self.dialog_tracer_obj.sys_msg(
"======================== SELECTION OF " +
counter_dict[counter - 1] +
" FOOD ITEM: END ======================")
return result_list
# Finalizing the Food item if narrowed to less than equal to three item
def _get_current_food_item_from_user(self, food_items_tokens, counter):
counter_dict = {
1: 'first',
2: 'second',
3: 'third',
4: 'fourth',
5: 'fifth'
}
if len(food_items_tokens) == 1:
return 1
if len(food_items_tokens) == 2:
self.task_manager_obj.text_to_audio(
'Please help me finalize your {0} food item. Say, \"I had the first item\" for {1}, or, '
'\"I had the second item\" for {1} {2}'.format(
counter_dict[counter], food_items_tokens[0],
food_items_tokens[1]))
else:
self.task_manager_obj.text_to_audio(
'Please help me finalize your {0} food item. Say, \"I had the first item\" for {1}, or, '
'\"I had the second item\" for {1} {2}, or, \"I had the third item\" '
'for {1} {2} {3}'.format(counter_dict[counter],
food_items_tokens[0],
food_items_tokens[1],
food_items_tokens[2]))
result = self._get_current_food_combination_from_user()
return result
#
def _get_current_food_combination_from_user(self):
user_input = self.task_manager_obj.audio_to_text()
user_input = user_input.lower().strip()
if any(word in user_input for word in ['first', 'second', 'third']):
self.task_manager_obj.text_to_audio('Thanks for your response')
else:
self.task_manager_obj.text_to_audio(
'I could not understand, please provide your choice again.')
user_input = self.task_manager_obj.audio_to_text()
user_input = user_input.lower().strip()
if any(word in user_input
for word in ['first', 'second', 'third']):
self.task_manager_obj.text_to_audio('Thanks for your response')
return 3 if 'third' in user_input else 2 if 'second' in user_input else 1
# Get Name from User
def get_user_name_from_user(self):
self.dialog_tracer_obj.sys_msg(
"======================== EXTRACT USER NAME: BEGIN ======================"
)
self.task_manager_obj.text_to_audio('Hello! Who am I speaking to?')
user_input = self.task_manager_obj.audio_to_text()
google_cloud_result = self.task_manager_obj.extract_user_name_from_text(
user_input)
if len(google_cloud_result) > 0:
name = google_cloud_result[0].split()[0]
self.task_manager_obj.text_to_audio(
'{0}, Did I get your name right? '
'Say yes to confirm, No to try again'.format(name))
result = self.task_manager_obj.audio_to_text()
result = result.lower().strip()
if len(result) == 0:
self.task_manager_obj.text_to_audio(
'I could not understand your response. '
'Say yes if I correctly identified your name. '
'Say No otherwise')
result = self.task_manager_obj.audio_to_text()
result = result.lower().strip()
if any(word in result for word in [
'yes', 'yeah', 'yup', 'yep', 'yo', 'ok', 'okay', 'right',
'correct', 'sure'
]):
self.task_manager_obj.text_to_audio(
'Great! Nice to meet you, {0}'.format(name))
self.dialog_tracer_obj.sys_msg(
'Final user_name: {0}'.format(name))
self.dialog_tracer_obj.sys_msg(
"======================== EXTRACT USER NAME: END ======================"
)
return name
self.task_manager_obj.text_to_audio(
'Sorry! I did not understand. Please tell me your name in a sentence.'
)
user_input = self.task_manager_obj.audio_to_text()
google_cloud_result = self.task_manager_obj.extract_user_name_from_text(
user_input)
if len(google_cloud_result) > 0:
name = google_cloud_result[0].split()[0]
self.task_manager_obj.text_to_audio(
'{0}, Did I get it right this time? Say yes to confirm.'.
format(name))
result = self.task_manager_obj.audio_to_text()
result = result.lower().strip()
if len(result) == 0:
self.task_manager_obj.text_to_audio(
'I could not understand your response. '
'Say yes if I correctly identified your name.'
'Say No otherwise')
result = self.task_manager_obj.audio_to_text()
result = result.lower().strip()
if any(word in result for word in [
'yes', 'yeah', 'yup', 'yep', 'yo', 'ok', 'okay', 'right',
'correct', 'sure'
]):
self.task_manager_obj.text_to_audio(
'Great! Nice to meet you, {0}'.format(name))
self.dialog_tracer_obj.sys_msg(
'Final user_name: {0}'.format(name))
self.dialog_tracer_obj.sys_msg(
"======================== EXTRACT USER NAME: END ======================"
)
return name
self.task_manager_obj.text_to_audio(
'Sorry! I could not understand your name.'
'For now, let me address you as, User')
self.dialog_tracer_obj.sys_msg('Final user_name: {0}'.format("USER"))
self.dialog_tracer_obj.sys_msg(
"========================= EXTRACT USER NAME: END ======================"
)
return 'user'
# Ask user for the food he had and return the Food Item tokens from it
def get_food_items_tokens_from_user(self, user_name):
self.dialog_tracer_obj.sys_msg(
"========================= EXTRACT FOOD TOKENS : BEGIN ======================"
)
food_tokens = self._get_food_consumption_from_user(user_name, False)
satisfied = self._confirm_food_items_from_user(user_name, food_tokens)
if not satisfied:
# food_tokens, food_token_text = self._get_food_consumption_from_user(user_name, True)
food_tokens = self._get_food_consumption_from_user(user_name, True)
satisfied = self._confirm_food_items_from_user(
user_name, food_tokens)
if not satisfied:
self.task_manager_obj.text_to_audio(
'I am sorry {0}, But the food items mentioned by you did not match '
'my knowledge'.format(user_name))
sys.exit()
self.dialog_tracer_obj.sys_msg(
"========================= EXTRACT FOOD TOKENS : END ======================"
)
return food_tokens
#
def _get_food_consumption_from_user(self, user_name, second):
if second:
self.task_manager_obj.text_to_audio(
' {0}, Mention the food items you had today with more detail'.
format(user_name))
else:
self.task_manager_obj.text_to_audio(
'Let us work on your caloric intake. What did you eat today, {0}?'
.format(user_name))
google_cloud_result = self.task_manager_obj.audio_to_text().lower()
food_tokens = self.task_manager_obj.get_tokens_from_audio(
google_cloud_result)
if len(food_tokens) == 0:
self.task_manager_obj.text_to_audio(
'Sorry {0}, can you repeat the food item names again.'.format(
user_name))
google_cloud_result = self.task_manager_obj.audio_to_text().lower()
food_tokens = self.task_manager_obj.get_tokens_from_audio(
google_cloud_result)
if len(food_tokens) == 0:
self.task_manager_obj.text_to_audio(
'I am sorry {0}, But the food items mentioned by you did not match '
'my knowledge'.format(user_name))
sys.exit()
return food_tokens
# Confirm the tokens extracted are the food item he had
def _confirm_food_items_from_user(self, user, food_tokens):
output = self.task_manager_obj.refractor_tokens_to_spoken_string(
food_tokens)
self.task_manager_obj.text_to_audio(
'Based on My Knowledge, these are the food items consumed '
'by you: {0}'.format(output))
self.dialog_tracer_obj.sys_msg(
'FINAL FOOD ITEM TOKENS:::: [{0}]'.format(output))
self.task_manager_obj.text_to_audio(
'{0},Please say yes, if all the food items are covered.'.format(
user))
result = self.task_manager_obj.audio_to_text()
result = result.lower().strip()
if any(word in result for word in [
'yes', 'yeah', 'yup', 'yep', 'yo', 'ok', 'okay', 'right',
'correct', 'sure'
]):
return True
return False
# Generate List of Contender for the Match and also get additional Token for
# the item asking user for more information by giving Hints
def get_usda_obj(self, food_item_token):
token_set = set()
for i in range(0, len(food_item_token)):
token_set.add(food_item_token[i])
s = ""
for item in food_item_token:
s = s + item
s += " "
result = self.task_manager_obj.get_fully_matched_food_results(
token_set)
if len(result) == 0:
self.task_manager_obj.text_to_audio(
"Sorry, I could not find any information for your selection, {0}. "
"Let us move to the next item".format(s))
return token_set, {}, result
# print(result)
dict_freq = self.task_manager_obj.freq_generator(result, token_set)
sorted_x = sorted(dict_freq.items(), key=operator.itemgetter(1))
high_freq_item = []
for item in sorted_x:
high_freq_item.append(item[0])
# print (sorted_x)
# print (high_freq_item)
if len(high_freq_item) > 3:
high_freq_item = high_freq_item[-3:]
# print (high_freq_item)
descriptors = ""
for word in high_freq_item:
descriptors = descriptors + word
descriptors += " , "
self.dialog_tracer_obj.sys_msg(descriptors)
if len(descriptors) == 0:
return token_set
self.task_manager_obj.text_to_audio(
'For your selection {0}, would you like to provide some description? '
'For instance, you could tell me which restaurant you ate it from, or '
'provide preparation information like raw, or cooked, boiled, or fried,'
'roasted or grilled etcetra. Few suggestions for your current '
'selection are: '
'{1}.. Say, \'No description\' if you want to skip'.format(
s, descriptors))
google_cloud_result = self.task_manager_obj.audio_to_text().lower()
if any(word in google_cloud_result for word in
['no description', 'nope', 'no'
'no, thanks', 'no, thank you']):
self.task_manager_obj.text_to_audio(
'Thanks. Let\'s proceed further.')
return token_set, dict_freq, result
token = self.task_manager_obj.get_tokens_from_audio(
google_cloud_result)
if len(google_cloud_result) == 0 or len(token) == 0:
self.task_manager_obj.text_to_audio(
'Sorry, I could not find matches for your description. '
'Please try explaining in a sentence. For example, \'I '
'ate my dish at Burger King\', or, \' My chicken was fried.\'.. Say, \''
'No description\' if you want to skip')
google_cloud_result = self.task_manager_obj.audio_to_text().lower()
if any(word in google_cloud_result for word in
['no description', 'nope', 'no'
'no, thanks', 'no, thank you']):
self.task_manager_obj.text_to_audio(
'Thanks. Let\'s proceed further.')
return token_set, dict_freq, result
self.task_manager_obj.text_to_audio('Thanks. Let\'s proceed further.')
token_set = token_set.union(set(token))
temp_result = self.task_manager_obj.get_fully_matched_food_results(
token_set)
if len(temp_result) == 0:
token_set = token_set.difference(set(token))
for i in range(0, len(token)):
token_set.add(token[i])
temp_result = self.task_manager_obj.get_fully_matched_food_results(
token_set)
if len(temp_result) == 0:
token_set.remove(token[i])
else:
for element in temp_result:
result.append(element)
else:
result = temp_result
dict_freq = self.task_manager_obj.freq_generator(result, token_set)
self.dialog_tracer_obj.sys_msg(token_set)
return token_set, dict_freq, list(set(result))
# Throw user set of 3 most probable desriptors , and get feedback
def ask_user_choice_from_items(self, item_list, contender_list, dict_freq,
experience):
item = str(item_list[0]) + "\n" + str(item_list[1]) + "\n" + str(
item_list[2])
item_1 = item_list[0][0]
item_2 = item_list[1][0]
item_3 = item_list[2][0]
self.dialog_tracer_obj.sys_msg(item)
#self.dialog_tracer_obj.sys_msg(item_list[0])
#self.dialog_tracer_obj.sys_msg(item_list[1])
#self.dialog_tracer_obj.sys_msg(item_list[2])
contender_len = len(contender_list)
display = "Contender List Strength : " + str(
contender_len) + "\nSize of Dictionary : " + str(len(dict_freq))
# self.dialog_tracer_obj.sys_msg("Contender List Strength : {0}".format(contender_len))
# self.dialog_tracer_obj.sys_msg("Size of Dictionary : {0}".format(len(dict_freq)))
self.dialog_tracer_obj.sys_msg(display)
if (experience == 1):
self.task_manager_obj.text_to_audio(
'Currently we have {3} choices which match your food item description. '
'. Say, \"My food item relates to first choice.\" for {0}. Or, \"My food '
'item relates to second choice.\" for {1}. Or, \"My food item relates '
'to third choice.\" for {2} . If you have no descriptor matches please '
'Say, \" None of the these matches\" . If Not sure please Say \" I am not '
'sure \"'.format(item_1, item_2, item_3, contender_len))
else:
self.task_manager_obj.text_to_audio('{3} Contender remain.'
'. Say, first, for {0}. Or,'
'Second for {1}. Or, '
'third for {2} '.format(
item_1, item_2, item_3,
contender_len))
audio_response = self.task_manager_obj.audio_to_text()
audio_response = audio_response.lower().strip()
self.task_manager_obj.text_to_audio('Thanks for your response')
if not ('first' in audio_response or 'second' in audio_response
or 'third' in audio_response or 'none' in audio_response or
('sure' in audio_response and 'not' in audio_response)):
self.task_manager_obj.text_to_audio(
' I could not understand your response, kindly speak again')
audio_response = self.task_manager_obj.audio_to_text()
audio_response = audio_response.lower().strip()
self.task_manager_obj.text_to_audio('Thanks for your response')
if 'first' in audio_response:
return item_1
if 'second' in audio_response:
return item_2
if 'third' in audio_response:
return item_3
if 'none' in audio_response:
return 'none'
if 'sure' in audio_response and 'not' in audio_response:
return 'not sure'
return 'not sure'
# If the contender list is reduced to less than equal to three, throw three contender item to user
# and ask to pick one
def get_final_item(self, contender_list):
if len(contender_list) == 2:
self.task_manager_obj.text_to_audio(
'Based on your input, two items best match your food item Say, \"My food item relates to '
'first item.\" for {0}. Or Say, \"My food item relates to second item.\" for '
'{1}'.format(contender_list[0], contender_list[1]))
audio_response = self.task_manager_obj.audio_to_text()
audio_response = audio_response.lower().strip()
self.task_manager_obj.text_to_audio('Thanks for your response')
if 'first' in audio_response:
return contender_list[0]
if 'second' in audio_response:
return contender_list[1]
return contender_list[0]
else:
self.task_manager_obj.text_to_audio(
'Based on your input, three items best match your food item Say, \"My food item relates to '
'first item.\" for {0}. Or Say, \"My food item relates to second item.\" for {1}.Or Say, '
'\"My food item relates to third item.\" for '
'{2}'.format(contender_list[0], contender_list[1],
contender_list[2]))
audio_response = self.task_manager_obj.audio_to_text()
audio_response = audio_response.lower().strip()
self.task_manager_obj.text_to_audio('Thanks for your response')
if 'first' in audio_response:
return contender_list[0]
if 'second' in audio_response:
return contender_list[1]
if 'third' in audio_response:
return contender_list[2]
return contender_list[0]
# Get reponse from the user whether he selected a descriptor or not sure or none of these and
# accordingly manage the freq dictionary And Contender List
def get_response(self, sorted_list_freq, contender_list, dict_freq,
experience):
if len(sorted_list_freq) < 3 or len(dict_freq) < 3 or len(
contender_list) <= 3:
if len(dict_freq) < 3 < len(contender_list):
contender_list = contender_list[:3]
return "-----", contender_list, dict_freq
response = self.ask_user_choice_from_items(
list(reversed(sorted_list_freq[-3:])), contender_list, dict_freq,
experience)
while response == 'none' or response == 'not sure':
if response == 'none':
contender_list, dict_freq = self.task_manager_obj.dict_contender_updator_for_none(
dict_freq, sorted_list_freq[-3:], contender_list)
sorted_list_freq = sorted(dict_freq.items(),
key=operator.itemgetter(1))
else:
if len(sorted_list_freq) >= 3:
for i in range(-3, 0):
del dict_freq[sorted_list_freq[i][0]]
sorted_list_freq = sorted_list_freq[:-3]
if len(sorted_list_freq) < 3 or len(dict_freq) < 3 or len(
contender_list) <= 3:
if len(dict_freq) < 3 < len(contender_list):
contender_list = contender_list[:3]
return "-----", contender_list, dict_freq
experience = experience + 1
response = self.ask_user_choice_from_items(
list(reversed(sorted_list_freq[-3:])), contender_list,
dict_freq, experience)
return response, contender_list, dict_freq
# Iteratively tries to reduce the list of items unless the list is less than equal to 3
def narrow_list(self, contender_list, dict_freq, experience):
while len(contender_list) >= 3:
sorted_list_freq = sorted(dict_freq.items(),
key=operator.itemgetter(1))
response, contender_list, dict_freq = self.get_response(
sorted_list_freq, contender_list, dict_freq, experience)
experience += 1
if response == "-----":
break
contender_list, dict_freq = self.task_manager_obj.reduce_list(
contender_list, response, dict_freq)
if len(contender_list) == 1:
return contender_list[0]
result = self.get_final_item(contender_list)
return result
# Returning the Food item based on the Food Tokens
def get_standard_item(self, food_items, experience):
token_set, dict_freq, contender_list = self.get_usda_obj(food_items)
if len(contender_list) == 0:
return "null"
result = self.narrow_list(contender_list, dict_freq, experience)
self.dialog_tracer_obj.sys_msg(result)
return result
# Displaying the Calorific Intake for Each Item
def show_food_items_with_calories(self, final_usda_food_items):
self.dialog_tracer_obj.sys_msg(
"======================== CALORIFIC INFORMATION : BEGIN ======================"
)
if len(final_usda_food_items) == 0:
self.dialog_tracer_obj.sys_msg("No Item to Display")
for food_item in final_usda_food_items:
self.task_manager_obj.display_item(food_item)
self.dialog_tracer_obj.sys_msg(
"======================== CALORIFIC INFORMATION : END ======================"
)