Python - 查找在列表中提取相同的值并计算连续值之间的差异

Question

我在dataquest上学习python并试图解决这个问题。

编写一个能够跨年提取相同值的函数，并计算连续值之间的差异，以显示出生数是增加还是减少。 例如，星期六的出生人数在1994年至2003年间每年如何变化？

我试图在Jupyter中解决这个问题。我是python的新手，我不知道如何开始解决这个问题。

此处的数据输入采用CSV格式：US births

    # coding: utf-8

# In[1]:

text_file = open("US_births_1994-2003_CDC_NCHS.csv", "r").read()
line_split = text_file.split("\n")
line_split


# In[2]:

def read_csv(filename):
    text = open(filename, "r").read()
    string_list = text.split('\n')[1:]
    final_list = []
    for row in string_list:
        int_fields = []
        string_fields = row.split(',')
        for item in string_fields:
            int_fields.append(int(item))
        final_list.append(int_fields)
    return(final_list)

cdc_list = read_csv("US_births_1994-2003_CDC_NCHS.csv")
cdc_list[0:10]


# In[3]:

def months_births(filename):
    births_per_month = dict()
    for item in filename:
        num_month = int(item[1])
        num_births = int(item[4])
        if num_month in births_per_month:
            births_per_month[num_month] += num_births
        else:
            births_per_month[num_month] = num_births
    return(births_per_month)

cdc_month_births = months_births(cdc_list)
cdc_month_births


# In[4]:

def dow_births(filename):
    sum_births = dict()
    for item in filename:
        day_week = int(item[3])
        day_birth = int(item[4])
        if day_week in sum_births:
            sum_births[day_week] += day_birth
        else:
            sum_births[day_week] = day_birth
    return(sum_births)

cdc_day_births = dow_births(cdc_list)
cdc_day_births


# In[30]:

def calc_counts(data, column):
    sum_dict = dict()
    for item in data:
        col_num = item[column]
        birth_count = int(item[4])
        if col_num in sum_dict:
            sum_dict[col_num] += birth_count
        else:
            sum_dict[col_num] = birth_count
    return(sum_dict)

cdc_year_births = calc_counts(cdc_list, 0)
cdc_month_births = calc_counts(cdc_list, 1)
cdc_dom_births = calc_counts(cdc_list, 2)
cdc_dow_births = calc_counts(cdc_list, 3)


# In[31]:

cdc_year_births


# In[32]:

cdc_month_births


# In[33]:

cdc_dom_births


# In[34]:

cdc_dow_births


# In[6]:

def min_max_dict(filename, request):
    if request == "max":
        max_value = max(filename, key=filename.get)
        return(filename[max_value])
    else:
        min_value = min(filename, key=filename.get)
        return(filename[min_value])

max_value = min_max_dict(cdc_year_births, "max")
print("max: ",max_value)
min_value = min_max_dict(cdc_year_births, "min")
print("min: ",min_value)


# In[36]:

def diff_in_values(filename):
    final_dict = dict()
    seen_set = set()
    unique_values = list()
    for item in filename:
        year_count = int(item[0])
        birth_count = int(item[4])
        day_of_week = int(item[3])

        if birth_count not in seen_set:
            unique_values.append(birth_count)
            seen_set.add(birth_count)

    return(seen_set)

result = diff_in_values(cdc_list)
result

Answer 1

我也在同一个项目上工作。我已经分享了你需要的部分代码。我在GitHub上有项目的.ipynb文件。您可能还希望看到我的功能结果。干杯！

def read_csv(birth_data_file):
    raw_data = open(birth_data_file, "r").read()
    raw_data = raw_data.split("\n")
    string_list = raw_data[1:]
    final_list = []
    for data in string_list:
        int_fields = []
        string_fields = data.split(",")
        for string_field in string_fields:
            field = int(string_field)
            int_fields.append(field)
        final_list.append(int_fields)
    return(final_list)


def calc_counts(data, column):
    births_counts = {}
    if not column > 0 and column <= 4:
        return("'column' must be either 1, 2, 3, or 4")
    else:
        for instance in data:
            field = instance[column-1]
            births  = instance[4]
            if field in births_counts.keys():
                births_counts[field] += births
            else:
                births_counts[field] = births
        return(births_counts)


# Write a function that extracts the same values across years and calculates the
# differences between consecutive values to show if number of
# births is increasing or decreasing.

def check_birth_growth(birth_data_file):
    cdc_list = read_csv(birth_data_file)
    cdc_year_births = calc_counts(cdc_list, 1)
    previous_year_birth = 0
    previous_birth_diff = 0
    for year, total_births in cdc_year_births.items():
        current_year_birth = int(total_births)
        if previous_year_birth == 0:
            growth_status = "Growth of births in {} not available.".format(year)
            print(growth_status)
            previous_year_birth = current_year_birth
        else:
            if current_year_birth > previous_year_birth:
                growth_status = "Births increased in {}.".format(year)
                print(growth_status)
                previous_year_birth = current_year_birth
            elif current_year_birth < previous_year_birth:
                growth_status = "Births decreased in {}.".format(year)
                print(growth_status)
                previous_year_birth = current_year_birth
            elif current_year_birth == previous_year_birth:
                growth_status = "Births in {} was same as previous year.".format(year)
                print(growth_status)
                previous_year_birth = current_year_birth

Answer 2

def calc_diff(filename, date_one, date_two, column, column_value):
birth_rate_result = []
previous_birth_rate = 0

for row in filename:
    year = row[0]
    current_birth_rate = row[4]
    time_unit = row[column]
    if year in range(date_one,date_two):
        if time_unit is column_value:
            birth_rate_diff = (current_birth_rate - previous_birth_rate)
            if birth_rate_diff > 0:
                growth_status = "increased"
                previous_birth_rate = current_birth_rate
            elif birth_rate_diff < 0:
                growth_status = "decreased"
                previous_birth_rate = current_birth_rate
            elif birth_rate_diff == 0:
                growth_status = "static"
                previous_birth_rate = current_birth_rate

            birth_rate_result.append([birth_rate_diff, growth_status,row[0]])

return birth_rate_result 

我刚刚开始学习python，请您告诉我如何运行这些代码示例（birth_diffrence = calc_diff（cdc_list，4,4,0,4））没有答案。

Answer 3

我相信上述代码仅计算出连续年份和期间之间的出生差异，而没有给出针对不同参数进行计算的机会-例如，在一月或星期日中不同年份的出生差异

在以下功能中，您可以执行以下操作：输入以下内容，查看星期一的出生年份之间的差异：列 = 3和 value = 1

def delta_year(input_list, column, value): # Column to loop over, value to compare over years

    column_year_count = {} # Create dictionary to get attribute yearly count

    for each in input_list:
        attribute = each[column]
        year = each[0]
        births = each[4]
        if attribute == value:
            if year in column_year_count:
                column_year_count[year] += births
            else:
                column_year_count[year] = births

    column_year_count_list = [] # Create list to order years to get consecutive years

    for key, value in column_year_count.items():
        temp = [key,value]
        column_year_count_list.append(temp)

    column_year_delta = {} # Create dictionary to insert ordered years & delta values

    for i in range(0,len(column_year_count_list)): # Loop over list index
        for j in range(0,len(column_year_count_list)): # Loop over second list index
            delta_year = column_year_count_list[j][0] - column_year_count_list[i][0] # Tally delta between years
            if delta_year == 1: # Select consecutive years only
                delta_year_births = column_year_count_list[j][1] - column_year_count_list[i][1]
                column_year_delta[column_year_count_list[j][0]] = delta_year_births
    return column_year_delta