Python：将excel文件转换为JSON格式

Question

我正在创建一个 ML 模型，它将使用 JSON 文件来理解模式和响应格式。由于我的数据为 excel 格式，因此我在 python 中将其转换为 JSON。

这是代码：

import xlrd
from collections import OrderedDict
import simplejson as json
# Open the workbook and select the first worksheet
wb = xlrd.open_workbook('D:\\android\\testdata2.xlsx')
sh = wb.sheet_by_index(0)
# List to hold dictionaries
data_list = []
# Iterate through each row in worksheet and fetch values into dict
for rownum in range(1, sh.nrows):
    data = OrderedDict()
    row_values = sh.row_values(rownum)
    data['pattern'] = row_values[0]
    data['response'] = row_values[1]
    data_list.append(data)
# Serialize the list of dicts to JSON
j = json.dumps(data_list)
# Write to file
with open('data1.json', 'w') as f:
    f.write(j)

我得到的输出为：

[{
    "pattern": "WALLSTENT NON COUVERTE ",
    "response": "ENDOPROTHESE STENT  VASCULAIRE "
}, {
    "pattern": "PRIMEADVANCED SURSCAN MRI ",
    "response": "NEUROSTIMULATEUR NERF VAGUE GAUCHE "
}, {
    "pattern": "AVASTIN  FLACON DE",
    "response": "BEVACIZUMAB"
}, {
    "pattern": "PERJETA SOLUTION A DILUER POUR PERFUSION",
    "response": "BRENTUXIMAB VEDOTIN"
}]

我正在寻找的所需输出是这样的：

{
    "intents": [{
        "pattern": ["WALLSTENT, NON, COUVERTE "],
        "response": ["ENDOPROTHESE STENT  VASCULAIRE] "
    }, {
        "pattern": ["PRIMEADVANCED ,SURSCAN ,MRI"] ,
        "response": ["NEUROSTIMULATEUR NERF VAGUE GAUCHE "]
    }, {
        "pattern": ["AVASTIN , FLACON ,DE"],
        "response": ["BEVACIZUMAB"]
    }, {
        "pattern": ["PERJETA, SOLUTION, A, DILUER, POUR ,PERFUSION"],
        "response": ["BRENTUXIMAB VEDOTIN"]
    }]
}

我可以在我的函数中做哪些修改来获得我正在寻找的输出。

Answer 1

应该这样做：

import xlrd
from collections import OrderedDict
import simplejson as json
# Open the workbook and select the first worksheet
wb = xlrd.open_workbook('D:\\android\\testdata2.xlsx')
sh = wb.sheet_by_index(0)
# List to hold dictionaries
data_list = []
# Iterate through each row in worksheet and fetch values into dict
for rownum in range(1, sh.nrows):
    data = OrderedDict()
    row_values = sh.row_values(rownum)
    data['pattern'] = row_values[0]
    data['response'] = row_values[1]
    data_list.append(data)
data_list = {'intents': data_list} # Added line
# Serialize the list of dicts to JSON
j = json.dumps(data_list)
# Write to file
with open('data1.json', 'w') as f:
    f.write(j)

注意添加的data_list = {'intents': data_list}.