问题

我是新手学习Python，尝试计算一些平均值。

我有一个包含以下数据的CSV文件：

网络,结果
网络1,100
网络2,200
网络3,300
网络1,101
网络2,202
网络3,303

我如何计算网络1、网络2和网络3的平均值，并创建另一个包含以下数据的CSV文件：

网络,平均值
网络1,105
网络2,201
网络3,301.5

看起来似乎很简单，但我正在为此苦苦思索。

提前感谢！

Network,Result
Network 1,100
Network 2,200
Network 3,300
Network 1,101
Network 2,202
Network 3,303

Network,Average
Network 1,105
Network 2,201
Network 3,301.5

答案1

得分: 3

这是一种方法：

import pandas as pd
df = pd.read_csv(path_to_csv)

(df 
.groupby('Network')['Result'].mean()
.rename(columns={"Result": "Average"})
.reset_index()
.to_csv("agg.csv", index=False))

     Network  Average
0  Network 1   100.5
1  Network 2   201.0
2  Network 3   301.5

import pandas as pd
df = pd.read_csv(path_to_csv)

(df 
.groupby('Network')['Result'].mean()
.rename(columns={"Result": "Average"})
.reset_index()
.to_csv("agg.csv", index=False))

     Network  Average
0  Network 1   100.5
1  Network 2   201.0
2  Network 3   301.5

答案2

得分: 1

如果你经常与数据库工作，可能会更喜欢编写SQL而不是重新发明轮子。我发现这是最容易理解的：

import pandas as pd
import pandasql as ps

mydata = pd.DataFrame([
['Network 1',100]
,['Network 2',200]
,['Network 3',300]
,['Network 1',101]
,['Network 2',202]
,['Network 3',303]]
,columns=['Network','Result'])

mysql = """
select Network,avg(Result) as avgresult
from mydata
group by Network
"""
print(ps.sqldf(mysql))

返回：

     Network  avgresult
0  Network 1      100.5
1  Network 2      201.0
2  Network 3      301.5

import pandas as pd
import pandasql as ps

mydata = pd.DataFrame([
['Network 1',100]
,['Network 2',200]
,['Network 3',300]
,['Network 1',101]
,['Network 2',202]
,['Network 3',303]]
,columns=['Network','Result'])

mysql = """
select Network,avg(Result) as avgresult
from mydata
group by Network
"""

print(ps.sqldf(mysql))  

     Network  avgresult
0  Network 1      100.5
1  Network 2      201.0
2  Network 3      301.5

答案3

得分: 1

Pandas提供了各种从文件中读取数据的方法。

read_csv()是其中之一。

这是您任务的解决方案。

# 导入Pandas库
import pandas as pd
# 从文件系统中读取CSV文件。
csv = pd.read_csv("data/temp1.csv")
# 您的语句表示您需要独特的网络数据，并对值进行平均处理。
# 对于学习，您可以根据需要在groupby子句中添加多个列名。
# 您还可以在agg函数中使用不同的聚合函数，如count、sum等。再次提醒，这是一个列表，因此可以同时传递多个聚合函数。
newds = csv.groupby(['Network'])['Result'].agg(['mean']).reset_index()
# 最后将您的数据集转换为CSV文件并保存到文件系统中。
newds.to_csv('new_csv.csv', index=False)

# Importing Pandas Library
import pandas as pd
# Reading the CSV file from the File system.
csv = pd.read_csv("data/temp1.csv")
# Your statement says your need unique Network data and average over the values.
# For your learning, you can add multiple column names in groupby clause according to your requirement. 
# You can also use different aggregation like count, sum etc. in agg function. Again its a list hence you can pass multiple aggregations at the same time.
newds = csv.groupby(['Network'])['Result'].agg(['mean']).reset_index()
# Finally converting your dataset to csv file and saving it into the File System. 
newds.to_csv('new_csv.csv', index=False)

答案4

得分: -1

import csv
from collections import defaultdict
with open("data.csv", newline='', mode='r') as csvfile:
    reader = csv.reader(csvfile, delimiter=' ')
    networks = defaultdict(lambda: [0,0])
    for row in reader:
        if len(row) == 1:
            continue
        else:
            network_num, value = row[1].split(",")
            networks[network_num][0] += int(value)
            networks[network_num][1] += 1
    with open("answer.csv", newline='', mode="w") as output:
        writer = csv.writer(output, quoting=csv.QUOTE_NONE)
        writer.writerow(["Network", "Average"])
        for k,v in networks.items():
            average = v[0] / v[1]
            writer.writerow([f'Network {k}', average])

output:

Network,Average
Network 1,100.5
Network 2,201.0
Network 3,301.5

this will output a csv file in the format you want.
defaultdict is a datatype that is similar to a dictionary, except if you attempt to access a key that does not exist yet, a default value is provided for that key so you don't get a key error.

<details>
<summary>英文:</summary>

import csv
from collections import defaultdict
with open("data.csv", newline='', mode='r') as csvfile:
reader = csv.reader(csvfile, delimiter=' ')
networks = defaultdict(lambda: [0,0])
for row in reader:
if len(row) == 1:
continue
else:
network_num, value = row[1].split(",")
networks[network_num][0] += int(value)
networks[network_num][1] += 1
with open("answer.csv", newline='', mode="w") as output:
writer = csv.writer(output, quoting=csv.QUOTE_NONE)
writer.writerow(["Network", "Average"])
for k,v in networks.items():
average = v[0] / v[1]
writer.writerow([f'Network {k}', average])

output:

Network,Average
Network 1,100.5
Network 2,201.0
Network 3,301.5

this will output a csv file in the format you want.
**defaultdict** is a datatype that is similar to a dictionary, except if you attempt to access a key that does not exist yet, a default value is provided for that key so you don't get a key error.



</details>