pandera 说它没有属性“validate”

Question

我正在尝试验证 Excel 工作表。这是我的架构和验证代码。根据文档，它应该像调用验证函数一样简单，并且它会打印任何错误，但它对我不起作用。我已确保架构是有效的 json 文件。我不断得到 ”

Traceback (most recent call last):
  File "/home/user/dev_projects/main2.py", line 57, in <module>
    errors = validate_schema(df,schema)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/user/dev_projects/app_functions/validation.py", line 47, in validate_schema
    schema.validate(df, lazy=True)
    ^^^^^^^^^^^^^^^
AttributeError: 'dict' object has no attribute 'validate'
"
validation snipet

for sheet_name in xls.sheet_names:
    print(f'validating {sheet_name}')
    df = pd.read_excel(xls, sheet_name)

    # check if schema exists
    schema_file = (os.path.join(schema_folder, f'{sheet_name}.json'))
    print(f'the schema file is {schema_file}')
    if not os.path.exists(schema_file):
        print('no schema file found')

    # load schema
    try:
        with open(schema_file, "r") as f:
            schema = json.load(f)
    except json.JSONDecodeError as e:
        print(f'Error decoding json in {schema_file}: {e.msg} at line {e.lineno}, column {e.colno}')

    try:
        schema.validate(df, lazy=True)
    except pa.errors.SchemaError as e:
        print(e)

架构.json

{
  "schema_type": "dataframe",
  "version": "0.13.4",
  "columns": {
    "mgmt_vip": {
      "dtype": "str",
      "nullable": false,
      "checks": [
        {
          "type": "ipv4_address",
          "coerce": true,
          "error":  "Value is not a valid IP Address"
        }
      ],
      "unique": true,
      "required": true
    },
    "mgmt_gw": {
      "dtype": "str",
      "nullable": false,
      "checks": [
        {
          "type": "ipv4_address",
          "coerce": true,
          "error":  "Value is not a valid IP Address"
        }
      ],
      "unique": true,
      "required": true
    },
    "ce_name": {
      "dtype": "str",
      "nullable": false,
      "required": true
    },
    "domain": {
      "dtype": "str",
      "nullable": false,
      "checks": [
        {
          "type": "condition",
          "condition": "lamda x: '.' in x and x.count('.') >= 2 and x[0] != '.' and x[-1] != '.'",
          "error": "Value is not an FQDN"
        }
      ],
      "unique": true,
      "required": true
    },
    "Location": {
      "dtype": "str",
      "nullable": false,
      "unique": false,
      "required": true
    },
    "Radius1": {
      "dtype": "str",
      "nullable": false,
      "checks": [
        {
          "type": "ipv4_address",
          "coerce": true,
          "error":  "Value is not a valid IP Address"
        }
      ],
      "unique": true,
      "required": true
    },
    "Radius2": {
      "dtype": "str",
      "nullable": false,
      "checks": [
        {
          "type": "ipv4_address",
          "coerce": true,
          "error":  "Value is not a valid IP Address"
        }
      ],
      "unique": true,
      "required": true
    },
    "Radius_Pw": {
      "dtype": "str",
      "nullable": false,
      "unique": false,
      "required": true
    },
    "routing_engine": {
      "dtype": "str",
      "nullable": false,
      "checks": [
        {
          "type": "isin",
          "isin": ["local", "remote"],
          "error": "Value must be local or remote"
        }
      ]
    },
    "Montitoring_IP": {
      "dtype": "str",
      "nullable": false,
      "checks": [
        {
          "type": "ipv4_address",
          "coerce": true,
          "error":  "Value is not a valid IP Address"
        }
      ]
    },
    "Rest_User": {
      "dtype": "str",
      "nullable": false,
      "unique": false,
      "required": true
    },
    "Rest_Password": {
      "dtype": "str",
      "nullable": false,
      "unique": false,
      "required": true
    }
  }
}

我在这上面花了几个小时，但无法验证架构，它一直给我一个错误，提示没有验证属性

Answer 1

所以文档显示有一个验证方法并且它有效。

import pandas as pd
import pandera as pa

from pandera import Check, Column, DataFrameSchema

df = pd.DataFrame({"column": ["a", "b", "c"]})

schema = pa.DataFrameSchema({"column": Column(int)})
print(type(schema))
schema.validate(df)

OUTPUT
SchemaError: expected series 'column' to have type int64, got object

模式是类型

<class 'pandera.api.pandas.container.DataFrameSchema'>

我正在使用

print(pa.__version__)
0.18.3

那么，如果您输入

module

那么您想要执行什么操作来检查您的模式，那么您就知道您已经导入了模块而不是来自模块的模式。

import altered_scheme
type(altered_schema)
OUTPUT
module

试试这个

from altered_scheme import schema
type(altered_schema)
OUTPUT
pandera.api.pandas.container.DataFrameSchema