将数百万数据作为datafram从SQL加载到python时出错

Question

我正在尝试从SQL server加载数以百万计的数据，SQL server会不断更新

我使用pymssql并尝试调整块大小。不幸的是，代码不起作用。无法读取数据

QUERY_CF = """
SELECT 
[mf].[ACCTNO],
[mf].[SUBC],
[mf].[COMPANYNO],
[mf].[YYYY-MM],
[mf].[CURRENCY],
[mf].[REVENUE],
[mf].[TOTAL_COST] as COST,
[mf].[GROUP],
[mf].[PARTNUMBER],
[mf].[SUBP],
[mf].[WO_PARTNUMBER],
[mf].[WO_SUBP]

FROM Master_Customer_Financials as mf with (nolock)


"""

conn = pymssql.connect(server=SERVER, user=USER, password=PASSWORD, database=DATABASE, port=PORT)
df1 = pd.read_sql(QUERY_CF, conn, chunksize = 1000000)
df1 = pd.concat(df1)
conn.close()
display(df1.head())
pd.set_option('display.max_columns', 999)

每次我运行代码时juypter笔记本都会崩溃