def population_statistics(
fea_des,
dt,
treat,
tar,
threshold,
is_above,
stat_funcs,
):
"""
Inputs -
fea_des - Str():
Description of the data needed for the
calculations of the stat_funcs.
Usge - to filter data by (feature, values) pairs
dt - Dict():
All the data
dt.keys() - Str()
dt.values() - List() :: Order()
treat - Str():
Specific feture of the data
* Must be in data.keys()
tar - Str():
Specific feture of the data
* Must be in data.keys()
threshold - Order():
A value that will be compered to values in data[treat][i]
* Must be able to cmp to data[treat][i]
is_above - Bool():
If true:
creates a list of the members in data[tar],
such that the matching data[treat] > threshold
else:
same but, data[treat] <= threshold.
stat_funcs - List() :: Obj-functions:
List of functions in statistics.py.
Outputs -
None
Main functionality -
Print the stat_funcs of the values in data[tar] while taking into
account the matching value in data[treat].
using data.print_details()
"""
# Not allowed, but much prettier code
# d = reduce(
# lambda ac, val: filter_by_feature(ac, val[0], val[1])[0],
# [dc[i.lower()] for i in fea_des.split(" ") if i.lower() in dc.keys()],
# data,
# )
new_target_data = [
x for x, y in zip(dt[tar], dt[treat]) if (is_above ^ (y <= threshold))
]
dpy.print_details(
{tar: new_target_data},
[tar],
stat_funcs,
)
def population_statistics(feature_description, data, treatment, target, threshold, is_above, statistic_functions):
""" Filters the dictionary to hold only records in which the treatment feature's value is above/equal or below
the threshold supplied (according to is_above argument value).
Prints statistic values of the target feature from the filtered dictionary.
:parameters:
feature_description -- a description of the records in data
data -- the dictionary to filter from
treatment -- a feature to filter by
target -- a feature according to which we'll print the statistic values
threshold -- a value to filter treatment's values by
is_above -- boolean, determines whether to filter the values above threshold the opposite
statistic_functions -- list of statistic functions in which we're interested
:returns: None
"""
data1 = {}
for key in data.keys():
data1[key] = []
for index in range(len(data[treatment])):
if(data[treatment][index] > threshold and is_above) or (data[treatment][index] <= threshold and not is_above):
# copying a record to a new dictionary
for key in data.keys():
data1[key].append(data[key][index])
print(feature_description)
print_details(data1, target, statistic_functions)
def main(argv):
features = (argv[2].split(sep=", "))
the_data = data.load_data(argv[1], features)
statistic_functions = [statistics.sum, statistics.mean, statistics.median]
# seasons = [2,3]
# s = set(seasons)
# data.filter_by_feature(data, "season", s)
#data.print_details(the_data, ["cnt"], statistic_functions)
summer_data = data.filter_by_feature(data, "season", 1)
data.print_details(summer_data, ["hum", "t1", "cnt"], statistic_functions)
def q1(dic):
print("Question1:")
feature = ["hum", "t1", "cnt"]
categories = ["season", "is_holiday"]
names = ["Summer:","Holiday:"]
methods = [statistics.sum, statistics.mean, statistics.median]
for index, category in enumerate(categories):
print(names[index])
data1, data2 = data.filter_by_feature(dic, category, {1})
data.print_details(data1, feature, methods)
print("All:")
data.print_details(dic, feature, methods)
def population_statistics(feature_description, data1, treatment, target,
threshold, is_above, statistic_functions):
print(feature_description)
dic = {}
for key in list(data1.keys()):
dic.setdefault(key, [])
if is_above:
for row, value in enumerate(data1[treatment]):
if value > threshold:
for key in list(data.keys()):
dic[key].append(data1[key][row])
else:
for row, value in enumerate(data1[treatment]):
if value <= threshold:
for key in list(data1.keys()):
dic[key].append(data1[key][row])
data.print_details(dic, [target], statistic_functions)
def Q1(data):
"""
this func prints the requirements for Q1
:param data: dict
:return: none
"""
print("Question 1:")
features = ["hum", "t1", "cnt"]
summer = (dt.filter_by_feature(data, "season", {1}))[0]
holiday = (dt.filter_by_feature(data, "is_holiday", {1}))[0]
all = data
q1_categories = {"Summer": summer, "Holiday": holiday, "All": all}
for category in q1_categories:
print("%s:" % category)
dt.print_details(q1_categories[category], features,
stats.statistic_functions)
def population_statistics(feature_description, data, treatment, target,
threshold, is_above, statistic_functions):
data1 = {}
for key in data.keys():
data1[key] = []
#seasons=["summer","winter","automn","spring"]
#features=feature_description.split(sep=" ")
#if features[0] in seasons:
#for index in range(len(data["season"])):
# if data[season][index] in values:
# for key in data.keys():
# data1[key].append(data[key][index])
for index in range(len(data[treatment])):
if ((data[treatment][index] > threshold and is_above)
or (data[treatment][index] <= threshold and not is_above)):
for key in data.keys():
data1[key].append(data[key][index])
print_details(data1, target, statistic_functions)
def question1(data):
print("Question 1:")
print("Summer:")
winter, non_winter = filter_by_feature(data, 'season', [1])
features = ['hum', 't1', 'cnt']
statistic_functions = ['sum', 'mean', 'median']
print_details(winter, features, statistic_functions)
print("Holiday:")
holiday, non_holiday = filter_by_feature(data, 'is_holiday', [1])
print_details(holiday, features, statistic_functions)
print("All:")
print_details(data, features, statistic_functions)
def main(argv):
"""
produce answers to Q1, Q2 by user's input
:argument: path to scv file
:parameter: features in csv file for assignment
:return: none
"""
# Q1
print('Question 1:')
features = argv[2].split(
", ") # create a list with continuous variables only
features.remove("season"), features.remove("is_holiday")
stats = [statistics.sum, statistics.mean,
statistics.med] # create a list with statistic functions
data_all = data.load_data(argv[1], argv[2])
data_summer, data_not_summer = data.filter_by_features(
data_all, "season", {1})
print("Summer:")
data.print_details(data_summer, features, stats)
data_holiday, data_not_holiday = data.filter_by_features(
data_all, "is_holiday", {1})
print("Holiday:")
data.print_details(data_holiday, features, stats)
print("All:")
data.print_details(data_all, features, stats)
# Q2
lim_temp = 13
print("\nQuestion 2:")
data_winter, data_not_winter = data.filter_by_features(
data_all, 'season', {3})
data_winter_holiday, data_winter_weekday = data.filter_by_features(
data_winter, 'is_holiday', {1})
features.remove("hum"), features.remove("t1")
stats.remove(statistics.sum)
print("If t1<=13.0, then:") # prints info 13 or under degrees
statistics.population_statistics("winter holiday records",
data_winter_holiday, "t1", "cnt",
lim_temp, 0, stats)
statistics.population_statistics("winter weekday records",
data_winter_weekday, "t1", "cnt",
lim_temp, 0, stats)
print("If t1>13.0, then:") # prints info above 13 degrees
statistics.population_statistics("winter holiday records",
data_winter_holiday, "t1", "cnt",
lim_temp, 1, stats)
statistics.population_statistics("winter weekday records",
data_winter_weekday, "t1", "cnt",
lim_temp, 1, stats)
def main(argv):
"""Main interface for running the program"""
# "london_sample.csv", ['season', 't1', 'is_holiday', 'cnt', 'hum']
# Q1
print(argv)
d = data.load_data(argv[1], argv[2].split(', '))
print("Question 1:")
print("Summer:")
d1, d2 = data.filter_by_feature(d, 'season', [1])
data.print_details(d1, ['hum', 't1', 'cnt'],
[statistics.sum, statistics.mean, statistics.median])
print("Holiday:")
d1, d2 = data.filter_by_feature(d, 'is_holiday', [1])
data.print_details(d1, ['hum', 't1', 'cnt'],
[statistics.sum, statistics.mean, statistics.median])
print("All:")
data.print_details(d, ['hum', 't1', 'cnt'],
[statistics.sum, statistics.mean, statistics.median])
# Q2
print("")
print("Question 2:")
print("If t1<=13.0, then:")
print("Winter holiday records:")
data_main = data.load_data(argv[1], argv[2].split(', '))
data_main_1, data_main_2 = data.filter_by_feature(data_main, 'season', [3])
winter_holiday, is_not_holiday = data.filter_by_feature(
data_main_1, 'is_holiday', [1])
statistics.population_statistics('cnt', winter_holiday, 't1', 'cnt', 13.0,
False,
[statistics.mean, statistics.median])
print("Winter weekday records:")
statistics.population_statistics('cnt', is_not_holiday, 't1', 'cnt', 13.0,
False,
[statistics.mean, statistics.median])
print("If t1>13.0, then:")
print("Winter holiday records:")
statistics.population_statistics('cnt', winter_holiday, 't1', 'cnt', 13.0,
True,
[statistics.mean, statistics.median])
print("Winter weekday records:")
statistics.population_statistics('cnt', is_not_holiday, 't1', 'cnt', 13.0,
True,
[statistics.mean, statistics.median])
def main(argv):
# loading the csv, the statistic functions and features for the program
str_input = argv[2]
features: list = str_input.split(", ")
dictionary = data.load_data(argv[1], features)
statistic_functions = [statistics.sum, statistics.mean, statistics.median]
continuous_features = features[:3]
"""
##############################~~QUESTION 1~~###################################
holder are just spare dictionaries to store the second returned dictionary,
they are not being used during the program
"""
data_summer, holder = data.filter_by_feature(dictionary, features[3],
SUMMER)
data_holiday, holder = data.filter_by_feature(dictionary, features[4],
HOLIDAY)
print(f"Question 1:")
print(f"Summer:")
data.print_details(data_summer, continuous_features, statistic_functions)
print(f"Holiday:")
data.print_details(data_holiday, continuous_features, statistic_functions)
print(f"All:")
data.print_details(dictionary, continuous_features, statistic_functions)
"""
############################~~QUESTION 2~~###############################
"""
descriptions = ["Winter holiday records:", "Winter weekday records:"]
data_winter, holder = data.filter_by_feature(dictionary, features[3],
WINTER)
new_data_holiday, data_weekday = data.filter_by_feature(
data_winter, features[4], HOLIDAY)
list_of_dictionaries = [new_data_holiday, data_weekday]
new_statistic_functions = [statistics.mean, statistics.median]
signs_list = ["<=", ">"]
print(f"\nQuestion 2:")
for i in range(SECTIONS):
print(f"If {features[1]}{signs_list[i]}13.0, then:")
for j in range(SECTIONS):
data.population_statistics(descriptions[j],
list_of_dictionaries[j], features[1],
features[2], THRESHOLD, i,
new_statistic_functions)
def main(argv):
print("Question 1:")
csv_path = argv[1]
features = argv[2]
features = list(features.split(", "))
data = data_lib.load_data(csv_path, features)
statistic_functions = [statistics.sum, statistics.mean, statistics.median]
data1, data2 = data_lib.filter_by_feature(data, "season", [1])
print("Summer:")
data_lib.print_details(data1, ["hum", "t1", "cnt"], statistic_functions)
data1, data2 = data_lib.filter_by_feature(data, "is_holiday", [1])
print("Holiday:")
data_lib.print_details(data1, ["hum", "t1", "cnt"], statistic_functions)
print("All:")
data_lib.print_details(data, ["hum", "t1", "cnt"], statistic_functions)
print("Question 2:")
t_threshold = 13
statistic_func = [statistics.mean, statistics.median]
print("If t1<=13.0, then:")
data1, data2 = data_lib.filter_by_feature(data, "season", [3])
data1, data2 = data_lib.filter_by_feature(data1, "is_holiday", [1])
print("Winter holiday records:")
statistics.population_statistics("Winter holiday records", data1, "t1",
["cnt"], t_threshold, False,
statistic_func)
print("Winter weekday records:")
statistics.population_statistics("Winter weekday records", data2, "t1",
["cnt"], t_threshold, False,
statistic_func)
print("If t1>13.0, then:")
print("Winter holiday records:")
statistics.population_statistics("Winter holiday records", data1, "t1",
["cnt"], t_threshold, True,
statistic_func)
print("Winter weekday records:")
statistics.population_statistics("Winter weekday records", data2, "t1",
["cnt"], t_threshold, True,
statistic_func)
def main(argv):
""" Calculates and prints statistic values for features of records from the data
collected in london.csv file.
:parameter:
argv -- list of arguments supplied (file to run, csv file address, features to get from the csv)
:returns: None
"""
# Question 1
# Saves the features given in a list
features = (argv[2].split(sep=", "))
the_data = data.load_data(argv[1], features)
statistic_functions = [sum, mean, median]
# Saves the relevant records
summer_data, not_summer = data.filter_by_feature(the_data, "season", [1])
holiday_data, not_holiday = data.filter_by_feature(the_data, "is_holiday",
[1])
print("Question 1:")
print("Summer:")
data.print_details(summer_data, ["hum", "t1", "cnt"], statistic_functions)
print("Holiday:")
data.print_details(holiday_data, ["hum", "t1", "cnt"], statistic_functions)
print("All:")
data.print_details(the_data, ["hum", "t1", "cnt"], statistic_functions)
# Question 2
print("\nQuestion 2")
print("If t1<=13.0, then:")
# Saves the relevant records
winter_data, not_winter = data.filter_by_feature(the_data, "season", [3])
w_h_data, not_w_h_data = data.filter_by_feature(winter_data, "is_holiday",
[1])
population_statistics("Winter holiday records:", w_h_data, "t1", ["cnt"],
THRESHOLD, 0, statistic_functions[1:])
population_statistics("Winter weekday records:", not_w_h_data, "t1",
["cnt"], THRESHOLD, 0, statistic_functions[1:])
print("If t1>13.0, then:")
population_statistics("Winter holiday records:", w_h_data, "t1", ["cnt"],
THRESHOLD, 1, statistic_functions[1:])
population_statistics("Winter weekday records:", not_w_h_data, "t1",
["cnt"], THRESHOLD, 1, statistic_functions[1:])
def main(argv):
"""
Main function of the program
:param argv: argv[0] = /home/student/your_path/main.py, argv[1] = /home/student/your_path/ london.csv,
argv[2] = "hum, t1, cnt, season, is_holiday"
:return: none
"""
# question 1
data_dict = data.load_data(argv[1], argv[2])
statistic_functions = [
statistics.sum_f, statistics.mean, statistics.median
]
print("Question 1:")
feature_list = ["hum", "t1", "cnt"]
data_name, data_not_name = data.filter_by_value(data_dict, 'season', {1})
data.print_details(data_name, feature_list, statistic_functions, 'Summer')
data_name, data_not_name = data.filter_by_value(data_dict, 'is_holiday',
{1})
data.print_details(data_name, feature_list, statistic_functions, 'Holiday')
data.print_details(data_dict, feature_list, statistic_functions, 'All')
print() # before question 2
# question 2
print("Question 2:")
is_above = 0
title_list = ["Winter holiday records:", "Winter weekday records:"]
print("If t1<=13.0, then:")
data_winter, data_not_winter = data.filter_by_value(
data_dict, 'season', {3})
data_name, data_not_name = data.filter_by_value(data_winter, 'is_holiday',
{1})
dict_list = [data_name, data_not_name]
for m, k in enumerate(title_list):
statistics.population_statistics(k, dict_list[m], 't1', 'cnt', 13.0,
is_above, statistic_functions)
is_above = 1
print("If t1>13.0, then:")
for n, l in enumerate(title_list):
statistics.population_statistics(l, dict_list[n], 't1', 'cnt', 13.0,
is_above, statistic_functions)
def print_details(self, features, statistic_functions):
data.print_details(self.dataset.data,features,statistic_functions)
def main(argv):
data = dt.load_data(argv[1], argv[2])
dc = {
"spring": ("season", [0]),
"summer": ("season", [1]),
"autumn": ("season", [2]),
"winter": ("season", [3]),
"holiday": ("is_holiday", [1]),
"weekend": ("is_weekend", [1]),
}
print("Question 1:")
for mode in ["Summer", "Holiday", "All"]:
print(f"{mode}:")
if mode == "All":
dt.print_details(data, ["hum", "t1", "cnt"],
[st.sum, st.mean, st.median])
else:
val = dc[mode.lower()]
dt.print_details(
dt.filter_by_feature(data, val[0], val[1])[0],
["hum", "t1", "cnt"],
[st.sum, st.mean, st.median],
)
print("\nQuestion 2:")
threshold = 13.0
winter_data, _ = dt.filter_by_feature(data, dc["winter"][0],
dc["winter"][1])
holiday_data, weekday_data = dt.filter_by_feature(winter_data,
dc["holiday"][0],
dc["holiday"][1])
print(f"If t1<={threshold}, then:")
print("Winter holiday records:")
st.population_statistics(
"Winter holiday records",
holiday_data,
"t1",
"cnt",
threshold,
False,
[st.mean, st.median],
)
print("Winter weekday records:")
st.population_statistics(
"Winter weekday records",
weekday_data,
"t1",
"cnt",
threshold,
False,
[st.mean, st.median],
)
print(f"If t1>{threshold}, then:")
print("Winter holiday records:")
st.population_statistics(
"Winter holiday records",
holiday_data,
"t1",
"cnt",
threshold,
True,
[st.mean, st.median],
)
print("Winter weekday records:")
st.population_statistics(
"Winter weekday records",
weekday_data,
"t1",
"cnt",
threshold,
True,
[st.mean, st.median],
)
def population_statistics(features_description, data, treatment, target,
threshold, is_above, statistic_functions):
from data import print_details # non in-function import will cause circular import
print(features_description)
print_details(data, target, statistic_functions)