彙總每日數據以計算月平均數

Question

您好，我是Python的新用戶，我遇到了問題，正在做我認爲是相當基本的任務。

我有幾個包含每日積雪深度數據的（> 50）csv文件。我想遍歷csv文件並計算每月的雪深度。數據示例：

Date,SD 
1/1/2000,36 
1/2/2000,36 
1/3/2000,38 
1/4/2000,40 
2/1/2000,48 
2/2/2000,48 

換句話說，我想計算每月積雪深度平均值並將輸出結果寫入新的csv文件。我能夠修改我的數據的一個不同的代碼示例，但我正在接收使用Date作爲我的字典中的鍵值的關鍵錯誤。

有什麼建議嗎？

到目前爲止的代碼：

from __future__ import division 
import csv 
from collections import defaultdict 

def default_factory(): 
    return [0, None, None, 0] 

reader = csv.DictReader(open(r'C:\SandBox\VALIDATION\TestTable.csv')) 

dates = defaultdict(default_factory) 
for row in reader: 
    sd = int(row["SD"]) 
    dates[row["Dates"]][0] += sd 
    max = dates[row["Dates"]][1] 
    dates[row["Dates"]][1] = amount if max is None else amount if amount > max else max 
    min = dates[row["Date"]][2] 
    dates[row["Dates"]][2] = amount if min is None else amount if amount < min else min 
    dates[row["Dates"]][3] += 1 

for date in dates: 
    dates[date][3] = dates[date][0]/dates[date][3] 

writer = csv.writer(open(r'C:\SandBox\VALIDATION\TestAvg.csv', 'w', newline = '')) 
writer.writerow(["Date", "SD", "max", "min", "mean"]) 
writer.writerows([date] + dates[date] for date in dates) 

編輯：只是爲了澄清，我想實現每月總平均，即一月平均，平均日等..不計算平均單個日期。

Answer 1

你可能想使用字典使代碼更易讀。

from __future__ import division 
import csv 
from collections import defaultdict 

def default_factory(): 
    return { "sum": 0, "max": None, "min": None, "count": 0} 

reader = csv.DictReader(open(r'sd.csv')) 

dates = defaultdict(default_factory) 
rows = [] 
for row in reader: 
    date = row["Date"] 
    sd = int(row["Snowdepth"]) 
    rows.append([date, sd]) 
    month = date.split("/")[0] 
    r = dates[month] 
    r["sum"] += sd 
    max = r["max"] 
    r["max"] = sd if max is None else sd if sd > max else max 
    min = r["min"] 
    r["min"] = sd if min is None else sd if sd < min else min 
    r["count"] += 1 

for date in dates: 
    r = dates[date] 
    r["avg"] = r["sum"]/r["count"] 

writer = csv.writer(open(r'TestAvg.csv', 'w')) 
writer.writerow(["Date", "SD", "max", "min", "mean"]) 
for row in rows: 
    r = dates[row[0].split("/")[0]] 
    writer.writerow(row + [r["max"], r["min"], r["avg"]]) 

Answer 2

Someplaces您已經使用Dates作爲列名（例如max = dates[row["Dates"]][1]）和其他地方它是Date（例如min = dates[row["Date"]][2]），從你的示例數據看起來像Date是列名？所以如果你在任何地方使用相同的名字，它應該可以。

s="""Date,Snowdepth 
1/1/2000,36 
1/2/2000,36 
1/3/2000,38 
1/4/2000,40 
2/1/2000,48 
2/2/2000,48""" 

import StringIO 
import csv 
reader = csv.DictReader(StringIO.StringIO(s)) 

for row in reader: 
    print row['Date'] 

輸出：

1/1/2000 
1/2/2000 
1/3/2000 
1/4/2000 
2/1/2000 
2/2/2000 

Answer 3

from __future__ import division 
import csv 
from collections import defaultdict 

def default_factory(): 
    return [0, None, None, 0] 

reader = csv.DictReader(open(r'snow_data.csv')) 

dates = defaultdict(default_factory) 

for row in reader: 
    amount = int(row["Snowdepth"]) 
    dates[row["Date"]][0] += amount 
    max = dates[row["Date"]][1] 
    dates[row["Date"]][1] = amount if max is None else amount if amount > max else max 
    min = dates[row["Date"]][2] 
    dates[row["Date"]][2] = amount if min is None else amoun if amount < min else min 
    dates[row["Date"]][3] += 1 


for date in dates: 
    dates[date][3] = dates[date][0]/dates[date][3] 

writer = csv.writer(open(r'TestAvg.csv', 'w')) 
writer.writerow(["Date", "Snowdepth", "max", "min", "mean"]) 
writer.writerows([date] + dates[date] for date in dates) 

我固定的代碼使用Date和Snowdepth任何地方，那是你的樣本CSV提供什麼。此外，您有一個變量amount，該變量旨在爲sd，否則數量未定義。我到處都是那個amount。

它不會給出非常令人興奮的結果，除非您在單個日期有多個條目。

例如，這裏是從你的樣本CSV輸出：

Date,Snowdepth,max,min,mean 

1/3/2000,38,38,38,38.0 

2/2/2000,48,48,48,48.0 

2/1/2000,48,48,48,48.0 

1/4/2000,40,40,40,40.0 

1/1/2000,36,36,36,36.0 

1/2/2000,36,36,36,36.0