当我转置 dataframe 时，它显示 nan 值

Question

import pymysql
import pandas as pd
import numpy

conn = pymysql.connect(host="localhost",port=3306,db="school",user="root",password="@mit123")
print("Connection established sucessfully")
cursor = conn.cursor()

sql = "SELECT * FROM records"
cursor.execute(sql)
result = cursor.fetchall()

data= result
df = pd.DataFrame(data)
df1=df.T
print(df)
print(df1)

df2 = pd.DataFrame(df1,index=["id","name","rollno.","city"])
print(df2)

The following is the output. What could be causing the problem? Can't I transpose a data frame into another data frame?

Connection established sucessfully
   0       1  2   3       4
0  1    amit  1  92  jorhat
1  2  subham  2  93  jorhat
2  3     ram  3  89   surat
3  4    anil  4  91   delhi
4  5   abdul  5  81  bhopal
5  6  joseph  6  90  sikkim
6  7     Ben  7  94  indore
7  8     tom  8  99     goa
        0       1      2      3       4       5       6    7
0       1       2      3      4       5       6       7    8
1    amit  subham    ram   anil   abdul  joseph     Ben  tom
2       1       2      3      4       5       6       7    8
3      92      93     89     91      81      90      94   99
4  jorhat  jorhat  surat  delhi  bhopal  sikkim  indore  goa
           0    1    2    3    4    5    6    7
id       NaN  NaN  NaN  NaN  NaN  NaN  NaN  NaN
name     NaN  NaN  NaN  NaN  NaN  NaN  NaN  NaN
rollno.  NaN  NaN  NaN  NaN  NaN  NaN  NaN  NaN
city     NaN  NaN  NaN  NaN  NaN  NaN  NaN  NaN

Process finished with exit code 0

This is my sql table:

Also when I use an index in the data frame, it says shape error:

Shape of passed values is (5, 8), indices imply (4, 8)

Answer 1

I could reproduce the NaN error using my database. So I think the reason is that there are no column names there. So you can do following:

import pymysql
import pandas as pd
import numpy

conn = pymysql.connect(host="localhost",
                       port=3306,
                       db="school",
                       user="root",
                       password="@mit123")

print("Connection established sucessfully")

sql = "SELECT * FROM records"

df = pd.read_sql(con=conn,sql=sql)

df1=df.T
print(df)
print(df1)

df2 = pd.DataFrame(df1,index=["id","name","roll_number","city"])
print(df2)

This solves the NaN error. The shape error may be due to the fact that you are not passing the "percentage" column to the index, but I was unable to reproduce this error.