Spaces:
Sleeping
Sleeping
Upload 4 files
Browse files- batch_query.py +38 -0
- sdoc_request.py +43 -0
- streamlit_sdoc_analysis.py +70 -0
- tianshu.py +64 -0
batch_query.py
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from sdoc_request import get_tianshu_response
|
2 |
+
import pandas as pd
|
3 |
+
import time
|
4 |
+
import json
|
5 |
+
|
6 |
+
# 问题数据导入
|
7 |
+
sdoc_df = pd.read_excel("sdoc直播观众提问和回复.xlsx")
|
8 |
+
questions = sdoc_df["描述"].to_list()
|
9 |
+
|
10 |
+
# AI智能分析
|
11 |
+
categories = []
|
12 |
+
modules = []
|
13 |
+
summaries = []
|
14 |
+
|
15 |
+
for qs in questions:
|
16 |
+
if qs and len(qs)>0:
|
17 |
+
time.sleep(1)
|
18 |
+
params = {"HOA_USERINPUT": qs}
|
19 |
+
temp_result = get_tianshu_response(params)
|
20 |
+
try:
|
21 |
+
result = json.loads(temp_result)
|
22 |
+
except json.JSONDecodeError:
|
23 |
+
result = {"类别":'', "模块":'', "问题归类": ''}
|
24 |
+
print("not json!")
|
25 |
+
else:
|
26 |
+
result = {"类别":'', "模块":'', "问题归类": ''}
|
27 |
+
print("question:", qs)
|
28 |
+
print("answer:", result)
|
29 |
+
categories.append(result["类别"])
|
30 |
+
modules.append(result["模块"])
|
31 |
+
summaries.append(result["问题归类"])
|
32 |
+
|
33 |
+
# 结果数据整理
|
34 |
+
sdoc_df['类别'] = categories
|
35 |
+
sdoc_df['模块'] = modules
|
36 |
+
sdoc_df['问题归类'] = summaries
|
37 |
+
|
38 |
+
sdoc_df.to_excel("smart_analysis.xlsx", index=0)
|
sdoc_request.py
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from tianshu import AiHelperClient, generate_serial_number
|
2 |
+
|
3 |
+
def get_tianshu_response(params):
|
4 |
+
# 请求地址
|
5 |
+
base_url = "https://api-aihelper.sheincorp.cn"
|
6 |
+
# 流水号,非必填,请求幂等处理,方便后续请求跟踪
|
7 |
+
out_id = generate_serial_number()
|
8 |
+
# 应用秘钥,必填
|
9 |
+
app_secret = "120a6f8685f8652ace71f7a2b0f2d395"
|
10 |
+
# 应用id,必填
|
11 |
+
scene_id = 7180
|
12 |
+
# 业务编码,对应系统编码或工号,必填
|
13 |
+
biz_code = "10270178"
|
14 |
+
# 构建请求客户端
|
15 |
+
client = AiHelperClient(base_url, app_secret, scene_id)
|
16 |
+
# 构建请求参数
|
17 |
+
# params = {"HOA_USERINPUT": "在线表格可以 @人吗"}
|
18 |
+
|
19 |
+
|
20 |
+
response = client.post("/open/v1/chat", {
|
21 |
+
# "out_id": out_id,
|
22 |
+
"scene_id": scene_id,
|
23 |
+
"biz_code": biz_code,
|
24 |
+
"params": params
|
25 |
+
})
|
26 |
+
result = response["info"].get("info")
|
27 |
+
|
28 |
+
return result
|
29 |
+
|
30 |
+
|
31 |
+
# if response is not None:
|
32 |
+
# result = response["info"].get("info")
|
33 |
+
# try:
|
34 |
+
# json_result = json.loads(result)
|
35 |
+
# print(json_result["类别"])
|
36 |
+
# print(json_result["模块"])
|
37 |
+
# except json.JSONDecodeError:
|
38 |
+
# print("not json!")
|
39 |
+
# else:
|
40 |
+
# print("request error!")
|
41 |
+
|
42 |
+
params = {"question": "SDOC的表格,我已经建了个在线表格,我要插入一个本地EX表,从企微文档迁移过来是在的,但我在上面操作,上传不了新的"}
|
43 |
+
get_tianshu_response(params)
|
streamlit_sdoc_analysis.py
ADDED
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import streamlit as st
|
2 |
+
from sdoc_request import get_tianshu_response
|
3 |
+
import pandas as pd
|
4 |
+
import time
|
5 |
+
import json
|
6 |
+
|
7 |
+
# 设置页面标题
|
8 |
+
st.title("SDoc 问题智能分析")
|
9 |
+
|
10 |
+
# 创建文件上传组件
|
11 |
+
st.text("1. 少量问题,请访问链接提问。")
|
12 |
+
st.markdown("https://aiapi.sheincorp.cn/llm/#/scene/general_app/7180")
|
13 |
+
|
14 |
+
uploaded_file = st.file_uploader("2. 大量问题,请上传 Excel 文件,必须含有【问题描述】字段。", type=["xlsx"])
|
15 |
+
|
16 |
+
if uploaded_file is not None:
|
17 |
+
# 读取上传的 Excel 文件
|
18 |
+
sdoc_df = pd.read_excel(uploaded_file)
|
19 |
+
questions = sdoc_df["问题描述"].to_list()
|
20 |
+
|
21 |
+
# AI 智能分析
|
22 |
+
categories = []
|
23 |
+
modules = []
|
24 |
+
summaries = []
|
25 |
+
|
26 |
+
# 创建进度条
|
27 |
+
progress_bar = st.progress(0)
|
28 |
+
total_questions = len(questions)
|
29 |
+
|
30 |
+
for i, qs in enumerate(questions):
|
31 |
+
if qs and len(qs) > 0:
|
32 |
+
time.sleep(1)
|
33 |
+
params = {"HOA_USERINPUT": qs}
|
34 |
+
temp_result = get_tianshu_response(params)
|
35 |
+
try:
|
36 |
+
result = json.loads(temp_result)
|
37 |
+
except json.JSONDecodeError:
|
38 |
+
result = {"类别": '', "模块": '', "问题归类": ''}
|
39 |
+
st.warning(f"问题 '{qs}' 的返回结果不是有效的 JSON 格式。")
|
40 |
+
else:
|
41 |
+
result = {"类别": '', "模块": '', "问题归类": ''}
|
42 |
+
|
43 |
+
st.write(f"问题{i+1}: {qs}")
|
44 |
+
st.write(f"答案: {result}")
|
45 |
+
|
46 |
+
categories.append(result.get("类别"))
|
47 |
+
modules.append(result.get("模块"))
|
48 |
+
summaries.append(result.get("问题归类"))
|
49 |
+
|
50 |
+
# 更新进度条
|
51 |
+
progress_bar.progress((i + 1) / total_questions)
|
52 |
+
|
53 |
+
# 结果数据整理
|
54 |
+
sdoc_df['类别'] = categories
|
55 |
+
sdoc_df['模块'] = modules
|
56 |
+
sdoc_df['问题归类'] = summaries
|
57 |
+
|
58 |
+
# 将结果保存为 Excel 文件
|
59 |
+
output_file = "smart_analysis.xlsx"
|
60 |
+
sdoc_df.to_excel(output_file, index=0)
|
61 |
+
|
62 |
+
# 创建下载链接
|
63 |
+
with open(output_file, "rb") as file:
|
64 |
+
btn = st.download_button(
|
65 |
+
label="下载分析结果",
|
66 |
+
data=file,
|
67 |
+
file_name="smart_analysis.xlsx",
|
68 |
+
mime="application/vnd.openxmlformats-officedocument.spreadsheetml.sheet"
|
69 |
+
)
|
70 |
+
|
tianshu.py
ADDED
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import requests
|
2 |
+
import hashlib
|
3 |
+
import hmac
|
4 |
+
import base64
|
5 |
+
import time
|
6 |
+
import uuid
|
7 |
+
import random
|
8 |
+
|
9 |
+
# 生成流水号
|
10 |
+
def generate_serial_number():
|
11 |
+
uid = uuid.uuid4()
|
12 |
+
serial_number = uid.hex[:32]
|
13 |
+
return serial_number
|
14 |
+
|
15 |
+
# 获取随机码
|
16 |
+
def generate_random_code(length=5):
|
17 |
+
characters = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789"
|
18 |
+
code = ""
|
19 |
+
for _ in range(length):
|
20 |
+
code += random.choice(characters)
|
21 |
+
return code
|
22 |
+
|
23 |
+
# 生成签名和时间戳
|
24 |
+
def generate_signature(scene_id, request_path, app_secret, random_key):
|
25 |
+
timestamp = str(int(time.time() * 1000))
|
26 |
+
sign_string = f"{scene_id}&{timestamp}&{request_path}"
|
27 |
+
secret_key = app_secret + random_key
|
28 |
+
hmac_sha256 = hmac.new(
|
29 |
+
secret_key.encode(),
|
30 |
+
sign_string.encode(),
|
31 |
+
hashlib.sha256
|
32 |
+
).hexdigest().encode()
|
33 |
+
signature = random_key + base64.b64encode(hmac_sha256).decode()
|
34 |
+
return signature, timestamp
|
35 |
+
|
36 |
+
class AiHelperClient:
|
37 |
+
def __init__(self, base_url, app_secret, scene_id):
|
38 |
+
self.base_url = base_url
|
39 |
+
self.app_secret = app_secret
|
40 |
+
self.random_key = generate_random_code()
|
41 |
+
self.scene_id = scene_id
|
42 |
+
self.session = requests.Session()
|
43 |
+
|
44 |
+
def post(self, endpoint, data):
|
45 |
+
signature, timestamp = generate_signature(
|
46 |
+
self.scene_id,
|
47 |
+
endpoint,
|
48 |
+
self.app_secret,
|
49 |
+
self.random_key
|
50 |
+
)
|
51 |
+
|
52 |
+
url = f"{self.base_url}{endpoint}"
|
53 |
+
headers = {
|
54 |
+
'Content-Type': 'application/json',
|
55 |
+
'x-signature': signature,
|
56 |
+
'x-timestamp': timestamp
|
57 |
+
}
|
58 |
+
|
59 |
+
try:
|
60 |
+
response = self.session.post(url, json=data, headers=headers)
|
61 |
+
response.raise_for_status()
|
62 |
+
return response.json()
|
63 |
+
except requests.RequestException as e:
|
64 |
+
return None
|