def get_compare_value(data, focus_value, compare_value, focus_aggre_name,
compare_aggre_name):
sentences = []
relation, degree, ratio = get_compare_relation(focus_value, compare_value,
data)
sentence = f'The value in {focus_aggre_name} is {degree} {relation} {compare_aggre_name}'
sentences.append(get_sentence_setting('compare', sentence))
if len(focus_value) + len(compare_value) == len(
data['data_array']) and len(compare_value) > 2:
sentence = f'The value in {focus_aggre_name} is {degree} {relation} others'
sentences.append(get_sentence_setting('compare', sentence))
if relation != 'same as':
if relation == 'higher than':
extreme = 'highest'
elif relation == 'lower than':
extreme = 'lowest'
if (len(focus_value) == 1):
sentence = f'{focus_aggre_name} have the {extreme} {len(focus_value)} value'
sentences.append(get_sentence_setting('compare', sentence))
else:
sentence = f'{focus_aggre_name} have the {extreme} {len(focus_value)} value'
sentences.append(get_sentence_setting('compare', sentence))
return sentences
def compare_diff(data,
focus_data,
compare_data,
focus_name,
compare_name,
share_name,
main_dimension,
version='English'):
focus_array = cal_compare_diff(focus_data)
compare_array = cal_compare_diff(compare_data)
# # print(focus_array, compare_array)
sentences = []
relation, degree, range, ratio = compare(focus_array, compare_array)
if range == "all" and len(focus_array) + len(compare_array) > 2 and len(
data[main_dimension]) == len(focus_array) + len(compare_array):
extreme = "maximum"
if ratio < 1:
extreme = "minimum"
sentence = f"The difference among {share_name} reached its {extreme} in the category of {focus_name}"
# # print(' '.join(sentence.split()))
sentences.append(get_sentence_setting('compare_diff', sentence))
if relation == 'higher' or relation == 'lower':
relation = relation + " than"
else:
degree = "almost"
relation = "the same as"
sentence = f"The difference among {share_name} in {focus_name} is {degree} {relation} than {compare_name}"
sentences.append(get_sentence_setting('compare_diff', sentence))
return sentences
def cal_ccq_count_sentence(data, focus_id, focus_main_dim, focus_second_dim,
focus_main_name, focus_second_name):
values = [data['data_array'][i]['q0'] for i in focus_id]
value_max = max(values)
value_min = min(values)
sentences = []
unit = data['unit']
sentence = f"The value of {focus_second_name} in {focus_main_name} ranges from {value_min} to {value_max} {unit}"
sentences.append(get_sentence_setting('absolute_range', sentence))
# higher than:
if sum(focus_main_dim) == len(focus_main_dim):
sentence = f"The value of {focus_second_name} ranges from {value_min} to {value_max} {unit}"
else:
sentence = f"The value of {focus_main_name} in {focus_second_name} ranges from {value_min} to {value_max} {unit}"
sentences.append(get_sentence_setting('absolute_range', sentence))
if sum(focus_main_dim) == len(focus_main_dim):
sentence = f"All values of {focus_second_name} is higher than {value_min} {unit}"
else:
sentence = f"All values of {focus_main_name} in {focus_second_name} is higher than {value_min} {unit}"
sentences.append(get_sentence_setting('absolute_higher', sentence))
if sum(focus_main_dim) == len(focus_main_dim):
sentence = f"All values of {focus_second_name} is lower than {value_max} {unit}"
else:
sentence = f"All values of {focus_main_name} in {focus_second_name} is lower than {value_max} {unit}"
sentences.append(get_sentence_setting('absolute_lower', sentence))
return sentences
def sentence_local_trend(data,
focus_id,
compare_id,
major_name='c0',
second_name='o0',
version='English',
fuzzy=True,
ordinal_name='o0',
category_name='c0'):
if len(data[second_name]) < 4:
return []
focus_id, compare_id, category_chosen, ordinal_chosen = modify_local_trend(
data, focus_id, compare_id)
sentences = []
if len(ordinal_chosen) < 3:
return []
print("see see category chosen: ", category_chosen)
ordinal_sum_quantity = get_sum_quantity_array(data, category_chosen)
max_value = max(ordinal_sum_quantity)
category_num = len(data['c0'])
sentence = ''
if ordinal_sum_quantity[ordinal_chosen[1]] > ordinal_sum_quantity[
ordinal_chosen[0]] and ordinal_sum_quantity[
ordinal_chosen[1]] > ordinal_sum_quantity[ordinal_chosen[2]]:
# 这是一个补丁
if len(category_chosen) == 1:
sentence = f' {data["title"]} of {get_aggre_name([data[major_name][i] for i in category_chosen], category_num)} suddenly increase to {ordinal_sum_quantity[ordinal_chosen[1]]} {data["unit"]} in {get_aggre_name([data[second_name][ordinal_chosen[1]]], 10)} '
else:
sentence = f' there is an unusual rise in {get_aggre_name([data[second_name][ordinal_chosen[1]]], 10)} of {get_aggre_name([data[major_name][i] for i in category_chosen], category_num)}'
sentences.append(
get_sentence_setting('local_trend', sentence, focus_id,
compare_id))
sentence = f' the sum value of {get_aggre_name([data[major_name][i] for i in category_chosen], category_num)} has an unusual rise in {get_aggre_name([data[second_name][ordinal_chosen[1]]], 10)} '
sentences.append(
get_sentence_setting('local_sum_trend', sentence, focus_id,
compare_id))
if ordinal_sum_quantity[ordinal_chosen[1]] < ordinal_sum_quantity[
ordinal_chosen[0]] and ordinal_sum_quantity[
ordinal_chosen[1]] < ordinal_sum_quantity[ordinal_chosen[2]]:
sentence = f'There is an unusual drop in {get_aggre_name([data[second_name][ordinal_chosen[1]]], 10)} of {get_aggre_name([data[major_name][i] for i in category_chosen], category_num)}'
sentences.append(
get_sentence_setting('local_trend', sentence, focus_id,
compare_id))
sentence = f' the sum value of {get_aggre_name([data[major_name][i] for i in category_chosen], category_num)} has an unusual drop in {get_aggre_name([data[second_name][ordinal_chosen[1]]], 10)} '
sentences.append(
get_sentence_setting('local_sum_trend', sentence, focus_id,
compare_id))
return sentences
def cal_ccq_single_sentence(data, focus_id, focus_main_name,
focus_second_name):
sentences = []
value = data['data_array'][focus_id[0]]['q0']
sentence = f'The value of {focus_main_name} in {focus_second_name} is {value}'
sentences.append(get_sentence_setting('absolute_single', sentence))
return sentences
def sentence_compare_trend(data,
focus_id,
compare_id,
major_name='c0',
second_name='o0',
version='English',
fuzzy=True,
quantity_name='q0',
category_name='c0'):
print("compare_trend: ", data)
focus_id, compare_id, focus_category_chosen, focus_ordinal_chosen, compare_category_chosen, compare_ordinal_chosen, new_focus_related_id, new_compare_related_id = modify_compare_trend(
data, focus_id, compare_id)
if len(focus_category_chosen) == 0 or len(
compare_category_chosen
) == 0 or len(focus_ordinal_chosen) < 2 or len(compare_ordinal_chosen) < 2:
return []
max_value = max([datum[quantity_name] for datum in data['data_array']])
focus_name = get_aggre_name(
[data[category_name][i] for i in focus_category_chosen],
len(data[category_name]))
compare_name = get_aggre_name(
[data[category_name][i] for i in compare_category_chosen],
len(data[category_name]))
focus_segment_array, focus_parameter_array_ave, focus_quantity = get_ave_parameter(
data, focus_category_chosen, focus_ordinal_chosen, max_value)
compare_segment_array, compare_parameter_array_ave, compare_quantity = get_ave_parameter(
data, compare_category_chosen, compare_ordinal_chosen, max_value)
if len(focus_category_chosen) > 1:
allow_absolute_value = False
else:
allow_absolute_value = True
focus_sentence = get_sentence(data,
focus_segment_array,
focus_parameter_array_ave,
focus_quantity,
object_name='',
allow_absolute_value=allow_absolute_value)
focus_sentence = fix_sentence_end(focus_sentence)
if len(compare_category_chosen) > 1:
allow_absolute_value = False
else:
allow_absolute_value = True
compare_sentence = get_sentence(data,
compare_segment_array,
compare_parameter_array_ave,
compare_quantity,
object_name='',
allow_absolute_value=allow_absolute_value,
is_compare=True)
sentence = f'{data["title"]} of {focus_name} {focus_sentence}; while {compare_name} {compare_sentence}'
sentences = []
sentences.append(
get_sentence_setting('compare_trend', sentence, new_focus_related_id,
new_compare_related_id))
return sentences
def sentence_sum_trend(data,
focus_id,
compare_id,
major_name='c0',
second_name='q0',
version='English',
fuzzy=True,
ordinal_name='o0',
category_name='c0'):
focus_id, compare_id, category_chosen, ordinal_chosen, new_focus_related_id = modify_sum_trend(
data, focus_id, compare_id)
if len(ordinal_chosen) < 2:
return []
ordinal_sum_quantity = get_sum_quantity_array(data, category_chosen)
max_value = max(ordinal_sum_quantity)
segment_array, segment_parameter_array = get_segment_parameter(
ordinal_chosen, ordinal_sum_quantity, max_value)
# 只有一个没办法称为sum
if len(category_chosen) == 1:
object_name = f'{data["title"]} of {get_aggre_name([data[category_name][i] for i in category_chosen], len(data[category_name]))}'
else:
object_name = f'The sum of {data["title"]} {get_aggre_name([data[category_name][i] for i in category_chosen], len(data[category_name]))}'
sentence = get_sentence(data,
segment_array,
segment_parameter_array,
ordinal_sum_quantity,
object_name=object_name)
sentences = []
# sentence = str(segment_parameter_array)
sentences.append(
get_sentence_setting('sum_trend', sentence, focus_id, compare_id))
if len(category_chosen) > 1:
object_name = f'The value of {get_aggre_name([data[category_name][i] for i in category_chosen], len(data[category_name]))}'
sentence = get_sentence(data,
segment_array,
segment_parameter_array,
ordinal_sum_quantity,
object_name=object_name,
allow_absolute_value=False)
sentences.append(
get_sentence_setting('all_trend',
sentence,
new_focus_related_id,
compare_id,
sure=False))
return sentences
def compare_1d(data,
focus_data,
compare_data,
focus_name,
compare_name,
share_name,
main_dimension,
version='English'):
sentences = []
focus_array = cal_compare_1d(focus_data)
compare_array = cal_compare_1d(compare_data)
relation, degree, range, ratio = compare(focus_array, compare_array)
unit = data['unit']
if relation == 'higher' or relation == 'lower':
relation = relation + " than"
else:
degree = "almost"
relation = "the same as"
if len(focus_array) == 1:
sentence = f'In the category of {share_name}, {data["title"]} of {focus_name} is {focus_array[0]} {unit}, which is {degree} {relation} {compare_name}'
sentences.append(get_sentence_setting('compare', sentence))
sentences.append(get_sentence_setting('compare_ave', sentence))
# # print(focus_array, compare_array)
if range == "all" and len(focus_array) + len(compare_array) > 2 and len(
data[main_dimension]) == len(focus_array) + len(compare_array):
extreme = "highest"
if ratio < 1:
extreme = "lowest"
if len(focus_array) > 1:
sentence = f"{focus_name} has the {extreme} {len(focus_array)} value in {share_name}"
else:
sentence = f"{focus_name} is the {extreme} value in the categery of {share_name}"
# sentence = "the value in {} is the {} {} value in the category of {}".format(focus_name, extreme, len(focus_array), share_name)
# # print(' '.join(sentence.split()))
sentences.append(get_sentence_setting('compare', sentence))
sentence = "the value in {} is {} {} {} in the category of {}".format(
focus_name, degree, relation, compare_name, share_name)
# print(sentence)
sentences.append(get_sentence_setting('compare', sentence))
return sentences
def get_count(data, focus_value, compare_value, focus_aggre_name,
compare_aggre_name):
# print('into count')
sentences = []
V_all = [datum['q0'] for datum in data['data_array']]
V_total_max = max(V_all)
V_total_min = min(V_all)
V_min = min(focus_value)
V_max = max(focus_value)
count = len(focus_value)
if V_max == V_total_max:
sentence = f'There are {count} categeries that higher than {V_min}'
sentences.append(get_sentence_setting('count', sentence))
if V_min == V_total_min:
sentence = f'There are {count} categeries that lower than {V_max}'
sentences.append(get_sentence_setting('count', sentence))
return sentences
def cal_ccq_sum_sentence(data, focus_id, focus_main_dim, focus_second_dim,
focus_main_name, focus_second_name):
sum = 0
for i in focus_id:
sum = sum + data['data_array'][i]['q0']
sentences = []
unit = data['unit']
sentence = f'The sum value of {focus_main_name} in {focus_second_name} is {sum} {unit}'
sentences.append(get_sentence_setting('absolute_sum', sentence))
return sentences
def cal_ccq_average_sentence(data, focus_id, focus_main_dim, focus_second_dim,
focus_main_name, focus_second_name):
sum = 0
for i in focus_id:
sum = sum + data['data_array'][i]['q0']
average = sum / len(focus_id)
sentences = []
unit = data['unit']
sentence = f'The average value of {focus_main_name} in {focus_second_name} is {average} {unit}'
sentences.append(get_sentence_setting('absolute_average', sentence))
return sentences
def get_range(data, focus_value, compare_value, focus_aggre_name,
compare_aggre_name):
# type = absolute_range/ absolute_higher/
sentences = []
V_all = [datum['q0'] for datum in data['data_array']]
V_total_max = max(V_all)
V_total_min = min(V_all)
V_min = min(focus_value)
V_max = max(focus_value)
count = len(focus_value)
if len(focus_value) == 1:
sentence = f'The value of {focus_aggre_name} is {focus_value[0]}'
sentences.append(get_sentence_setting('absolute_single', sentence))
return sentences
if len(focus_value) == len(data['data_array']):
focus_aggre_name = 'all categories'
sentence = f'The value of {focus_aggre_name} are higher than {V_min}'
sentences.append(get_sentence_setting('absolute_higher', sentence))
sentence = f'The value of {focus_aggre_name} are lower than {V_max}'
sentences.append(get_sentence_setting('absolute_lower', sentence))
sentence = f'The value of {focus_aggre_name} ranges from {V_min} to {V_max}'
sentences.append(get_sentence_setting('absolute_range', sentence))
return sentences
def compare_max(data,
focus_data,
compare_data,
focus_name,
compare_name,
share_name,
main_dimension,
version='English'):
focus_array, share_name_array = cal_compare_max(focus_data)
compare_array, share_name_array = cal_compare_max(compare_data)
# 当在第一个数组中出现的部分不在第二个数组中出现。
focus_array = list(set(focus_array))
compare_array = list(set(compare_array))
focus_other_array = [
name for name in share_name_array if name not in focus_array
]
compare_other_array = [
name for name in share_name_array if name not in compare_array
]
# print(share_name)
focal_max_name = get_aggre_name(focus_array, len(share_name), version)
focal_other_name = get_aggre_name(focus_other_array, 100, version)
compare_max_name = get_aggre_name(compare_array, len(share_name), version)
compare_other_name = get_aggre_name(compare_other_array, len(share_name),
version)
ret = [i for i in focus_array if i in compare_array]
# # print(focal_max_name)
# # print(compare_max_name)
sentences = []
if len(ret) == 0 and len(focus_array) == 1 and len(compare_array) == 1:
if len(compare_other_array) < 2:
sentence = f"{focal_max_name} is higher than {focal_other_name} in {focus_name}, while {compare_max_name} is higher than {compare_other_name} in {compare_name}"
else:
sentence = f"{focal_max_name} is the maximum in {focus_name}, while {compare_max_name} is the maximum in {compare_name}"
sentences.append(get_sentence_setting('compare_max', sentence))
return sentences
def compare_min(data,
focus_data,
compare_data,
focus_name,
compare_name,
share_name,
main_dimension,
version='English'):
focus_array = cal_compare_min(focus_data)
compare_array = cal_compare_min(compare_data)
# 当在第一个数组中出现的部分不在第二个数组中出现。
focus_array = list(set(focus_array))
compare_array = list(set(compare_array))
# print(share_name)
focal_max_name = get_aggre_name(focus_array, len(share_name), version)
compare_max_name = get_aggre_name(compare_array, len(share_name), version)
ret = [i for i in focus_array if i in compare_array]
# # print(focal_max_name)
# # print(compare_max_name)
sentences = []
if len(ret) == 0 and len(focus_array) == 1 and len(compare_array) == 1:
sentence = f"{focal_max_name} is the minimum of {focus_name}, while {compare_max_name} is minimum of {compare_name} "
sentences.append(get_sentence_setting('compare_min', sentence))
return sentences
def get_same(data, focus_value, compare_value, focus_aggre_name,
compare_aggre_name):
sentences = []
sentence = f'The value of {focus_aggre_name} is almost the same, around {int(numpy.mean(focus_value))} {data["unit"]}'
sentences.append(get_sentence_setting('same', sentence))
return sentences
def compare_sum(data,
focus_data,
compare_data,
focus_name,
compare_name,
share_name,
main_dimension,
version='English'):
focus_array = cal_compare_sum(focus_data)
compare_array = cal_compare_sum(compare_data)
# print(focus_data.shape)
# # print(focus_array, compare_array)
sentences = []
relation, degree, range, ratio = compare(focus_array, compare_array)
if range == "all" and len(focus_array) == 1 and len(
data[main_dimension]) == len(focus_array) + len(compare_array):
extreme = "highest"
if ratio < 1:
extreme = "lowest"
if len(focus_array) > 1:
if len(compare_array) + len(focus_array) > 2:
sentence = f"{focus_name} has the {extreme} {len(focus_array)} sum value of {share_name}"
sentences.append(get_sentence_setting('compare_sum', sentence))
sentence = f"{focus_name} has the {extreme} {len(focus_array)} average value of {share_name}"
sentences.append(get_sentence_setting('compare_ave', sentence))
sentence = f"{focus_name} has the {extreme} {len(focus_array)} value of {share_name}"
sentences.append(get_sentence_setting('compare_val', sentence))
else:
sentence = f"The sum value of {share_name} in {focus_name} is {degree} {relation} {compare_name}"
sentences.append(get_sentence_setting('compare_sum', sentence))
sentence = f"{data['title']} of {share_name} in {focus_name} is {degree} {relation} {compare_name}"
sentences.append(get_sentence_setting('compare_ave', sentence))
sentence = f"The sum value of {share_name} in {focus_name} is {degree} {relation} {compare_name}"
sentences.append(get_sentence_setting('compare_val', sentence))
else:
if len(compare_array) > 1:
sentence = f"{focus_name} has the {extreme} sum value of {share_name}"
sentences.append(get_sentence_setting('compare_sum', sentence))
sentence = f"{focus_name} has the {extreme} average value of {share_name}"
sentences.append(get_sentence_setting('compare_ave', sentence))
sentence = f"{focus_name} has the {extreme} value of {share_name}"
sentences.append(get_sentence_setting('compare_val', sentence))
else:
sentence = f"The sum value of {share_name} in {focus_name} is {degree} {relation} {compare_name}"
sentences.append(get_sentence_setting('compare_sum', sentence))
sentence = f"The value of {share_name} in {focus_name} is {degree} {relation} {compare_name}"
sentences.append(get_sentence_setting('compare_ave', sentence))
sentence = f"The sum value of {share_name} in {focus_name} is {degree} {relation} {compare_name}"
sentences.append(get_sentence_setting('compare_val', sentence))
if version == 'Chinese':
if extreme == 'highest':
extreme = "大"
else:
extreme = "小"
sentence = f"平均而言，{focus_name}是{share_name}中最{extreme}的"
sentences.append(get_sentence_setting('compare_sum', sentence))
if len(focus_array) == 1:
sentence = f"{focus_name} has sum value of {focus_array[0]} and it's {extreme}"
sentences.append(get_sentence_setting('compare_sum', sentence))
sentence = f"{focus_name} has average value of {focus_array[0]/len(focus_data[0])} and it's {extreme}"
sentences.append(get_sentence_setting('compare_ave', sentence))
sentence = f"{focus_name} has average value of {focus_array[0]/len(focus_data[0])} and it's {extreme}"
sentences.append(get_sentence_setting('compare_ave', sentence))
if relation == 'higher' or relation == 'lower':
relation = relation + " than"
else:
degree = "almost"
relation = "the same as"
if len(data[main_dimension]) == len(focus_array) + len(compare_array):
if len(compare_array) > 1:
sentence = f'The value of {focus_name} is {degree} {relation} others in {share_name}'
else:
sentence = f'The value of {focus_name} is {degree} {relation} {compare_name} in {share_name}'
sentences.append(
get_sentence_setting('compare_ave', sentence, sure=False))
sentence = f"The sum value of {share_name} in {focus_name} is {degree} {relation} {compare_name}"
if version == 'Chinese':
sentence = f"{share_name}的总和在{focus_name}中比{compare_name}{relation_chinese[relation]}{degree_chinese[degree]}"
# shenme = 11
# sentence = f'the sum value in a is {shenme}'
sentences.append(get_sentence_setting('compare_sum', sentence))
sentence = f"{data['title']} of {focus_name} is {degree} {relation} {compare_name} in the category of {share_name}"
sentences.append(get_sentence_setting('compare_ave', sentence, sure=False))
return sentences