Spaces:
Sleeping
Sleeping
Tuchuanhuhuhu
commited on
Commit
·
079c7eb
1
Parent(s):
4b845f9
改进了在线搜索显示效果
Browse files- chat_func.py +41 -12
chat_func.py
CHANGED
@@ -6,6 +6,7 @@ import logging
|
|
6 |
import json
|
7 |
import os
|
8 |
import requests
|
|
|
9 |
|
10 |
from tqdm import tqdm
|
11 |
import colorama
|
@@ -99,6 +100,7 @@ def stream_predict(
|
|
99 |
top_p,
|
100 |
temperature,
|
101 |
selected_model,
|
|
|
102 |
):
|
103 |
def get_return_value():
|
104 |
return chatbot, history, status_text, all_token_counts
|
@@ -109,7 +111,10 @@ def stream_predict(
|
|
109 |
status_text = "开始实时传输回答……"
|
110 |
history.append(construct_user(inputs))
|
111 |
history.append(construct_assistant(""))
|
112 |
-
|
|
|
|
|
|
|
113 |
user_token_count = 0
|
114 |
if len(all_token_counts) == 0:
|
115 |
system_prompt_token_count = count_token(construct_system(system_prompt))
|
@@ -184,7 +189,7 @@ def stream_predict(
|
|
184 |
yield get_return_value()
|
185 |
break
|
186 |
history[-1] = construct_assistant(partial_words)
|
187 |
-
chatbot[-1] = (
|
188 |
all_token_counts[-1] += 1
|
189 |
yield get_return_value()
|
190 |
|
@@ -199,11 +204,15 @@ def predict_all(
|
|
199 |
top_p,
|
200 |
temperature,
|
201 |
selected_model,
|
|
|
202 |
):
|
203 |
logging.info("一次性回答模式")
|
204 |
history.append(construct_user(inputs))
|
205 |
history.append(construct_assistant(""))
|
206 |
-
|
|
|
|
|
|
|
207 |
all_token_counts.append(count_token(construct_user(inputs)))
|
208 |
try:
|
209 |
response = get_response(
|
@@ -229,7 +238,7 @@ def predict_all(
|
|
229 |
response = json.loads(response.text)
|
230 |
content = response["choices"][0]["message"]["content"]
|
231 |
history[-1] = construct_assistant(content)
|
232 |
-
chatbot[-1] = (
|
233 |
total_token_count = response["usage"]["total_tokens"]
|
234 |
all_token_counts[-1] = total_token_count - sum(all_token_counts)
|
235 |
status_text = construct_token_message(total_token_count)
|
@@ -247,7 +256,7 @@ def predict(
|
|
247 |
temperature,
|
248 |
stream=False,
|
249 |
selected_model=MODELS[0],
|
250 |
-
|
251 |
files = None,
|
252 |
should_check_token_count=True,
|
253 |
): # repetition_penalty, top_k
|
@@ -262,18 +271,24 @@ def predict(
|
|
262 |
history, chatbot, status_text = chat_ai(openai_api_key, index, inputs, history, chatbot)
|
263 |
yield chatbot, history, status_text, all_token_counts
|
264 |
return
|
265 |
-
|
266 |
-
|
|
|
|
|
|
|
|
|
267 |
web_results = []
|
268 |
-
for idx, result in enumerate(
|
269 |
logging.info(f"搜索结果{idx + 1}:{result}")
|
|
|
270 |
web_results.append(f'[{idx+1}]"{result["body"]}"\nURL: {result["href"]}')
|
271 |
-
|
272 |
inputs = (
|
273 |
replace_today(WEBSEARCH_PTOMPT_TEMPLATE)
|
274 |
.replace("{query}", inputs)
|
275 |
-
.replace("{web_results}", web_results)
|
276 |
)
|
|
|
277 |
if len(openai_api_key) != 51:
|
278 |
status_text = standard_error_msg + no_apikey_msg
|
279 |
logging.info(status_text)
|
@@ -286,8 +301,9 @@ def predict(
|
|
286 |
history[-2] = construct_user(inputs)
|
287 |
yield chatbot, history, status_text, all_token_counts
|
288 |
return
|
289 |
-
|
290 |
-
|
|
|
291 |
if stream:
|
292 |
logging.info("使用流式传输")
|
293 |
iter = stream_predict(
|
@@ -300,6 +316,7 @@ def predict(
|
|
300 |
top_p,
|
301 |
temperature,
|
302 |
selected_model,
|
|
|
303 |
)
|
304 |
for chatbot, history, status_text, all_token_counts in iter:
|
305 |
yield chatbot, history, status_text, all_token_counts
|
@@ -315,8 +332,10 @@ def predict(
|
|
315 |
top_p,
|
316 |
temperature,
|
317 |
selected_model,
|
|
|
318 |
)
|
319 |
yield chatbot, history, status_text, all_token_counts
|
|
|
320 |
logging.info(f"传输完毕。当前token计数为{all_token_counts}")
|
321 |
if len(history) > 1 and history[-1]["content"] != inputs:
|
322 |
logging.info(
|
@@ -325,10 +344,20 @@ def predict(
|
|
325 |
+ f"{history[-1]['content']}"
|
326 |
+ colorama.Style.RESET_ALL
|
327 |
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
328 |
if stream:
|
329 |
max_token = max_token_streaming
|
330 |
else:
|
331 |
max_token = max_token_all
|
|
|
332 |
if sum(all_token_counts) > max_token and should_check_token_count:
|
333 |
status_text = f"精简token中{all_token_counts}/{max_token}"
|
334 |
logging.info(status_text)
|
|
|
6 |
import json
|
7 |
import os
|
8 |
import requests
|
9 |
+
import urllib3
|
10 |
|
11 |
from tqdm import tqdm
|
12 |
import colorama
|
|
|
100 |
top_p,
|
101 |
temperature,
|
102 |
selected_model,
|
103 |
+
fake_input=None
|
104 |
):
|
105 |
def get_return_value():
|
106 |
return chatbot, history, status_text, all_token_counts
|
|
|
111 |
status_text = "开始实时传输回答……"
|
112 |
history.append(construct_user(inputs))
|
113 |
history.append(construct_assistant(""))
|
114 |
+
if fake_input:
|
115 |
+
chatbot.append((parse_text(fake_input), ""))
|
116 |
+
else:
|
117 |
+
chatbot.append((parse_text(inputs), ""))
|
118 |
user_token_count = 0
|
119 |
if len(all_token_counts) == 0:
|
120 |
system_prompt_token_count = count_token(construct_system(system_prompt))
|
|
|
189 |
yield get_return_value()
|
190 |
break
|
191 |
history[-1] = construct_assistant(partial_words)
|
192 |
+
chatbot[-1] = (chatbot[-1][0], parse_text(partial_words))
|
193 |
all_token_counts[-1] += 1
|
194 |
yield get_return_value()
|
195 |
|
|
|
204 |
top_p,
|
205 |
temperature,
|
206 |
selected_model,
|
207 |
+
fake_input=None
|
208 |
):
|
209 |
logging.info("一次性回答模式")
|
210 |
history.append(construct_user(inputs))
|
211 |
history.append(construct_assistant(""))
|
212 |
+
if fake_input:
|
213 |
+
chatbot.append((parse_text(fake_input), ""))
|
214 |
+
else:
|
215 |
+
chatbot.append((parse_text(inputs), ""))
|
216 |
all_token_counts.append(count_token(construct_user(inputs)))
|
217 |
try:
|
218 |
response = get_response(
|
|
|
238 |
response = json.loads(response.text)
|
239 |
content = response["choices"][0]["message"]["content"]
|
240 |
history[-1] = construct_assistant(content)
|
241 |
+
chatbot[-1] = (chatbot[-1][0], parse_text(content))
|
242 |
total_token_count = response["usage"]["total_tokens"]
|
243 |
all_token_counts[-1] = total_token_count - sum(all_token_counts)
|
244 |
status_text = construct_token_message(total_token_count)
|
|
|
256 |
temperature,
|
257 |
stream=False,
|
258 |
selected_model=MODELS[0],
|
259 |
+
use_websearch=False,
|
260 |
files = None,
|
261 |
should_check_token_count=True,
|
262 |
): # repetition_penalty, top_k
|
|
|
271 |
history, chatbot, status_text = chat_ai(openai_api_key, index, inputs, history, chatbot)
|
272 |
yield chatbot, history, status_text, all_token_counts
|
273 |
return
|
274 |
+
|
275 |
+
old_inputs = ""
|
276 |
+
link_references = []
|
277 |
+
if use_websearch:
|
278 |
+
search_results = ddg(inputs, max_results=5)
|
279 |
+
old_inputs = inputs
|
280 |
web_results = []
|
281 |
+
for idx, result in enumerate(search_results):
|
282 |
logging.info(f"搜索结果{idx + 1}:{result}")
|
283 |
+
domain_name = urllib3.util.parse_url(result["href"]).host
|
284 |
web_results.append(f'[{idx+1}]"{result["body"]}"\nURL: {result["href"]}')
|
285 |
+
link_references.append(f"[{idx+1}]: [{domain_name}]({result['href']})")
|
286 |
inputs = (
|
287 |
replace_today(WEBSEARCH_PTOMPT_TEMPLATE)
|
288 |
.replace("{query}", inputs)
|
289 |
+
.replace("{web_results}", "\n\n".join(web_results))
|
290 |
)
|
291 |
+
|
292 |
if len(openai_api_key) != 51:
|
293 |
status_text = standard_error_msg + no_apikey_msg
|
294 |
logging.info(status_text)
|
|
|
301 |
history[-2] = construct_user(inputs)
|
302 |
yield chatbot, history, status_text, all_token_counts
|
303 |
return
|
304 |
+
|
305 |
+
yield chatbot, history, "开始生成回答……", all_token_counts
|
306 |
+
|
307 |
if stream:
|
308 |
logging.info("使用流式传输")
|
309 |
iter = stream_predict(
|
|
|
316 |
top_p,
|
317 |
temperature,
|
318 |
selected_model,
|
319 |
+
fake_input=old_inputs
|
320 |
)
|
321 |
for chatbot, history, status_text, all_token_counts in iter:
|
322 |
yield chatbot, history, status_text, all_token_counts
|
|
|
332 |
top_p,
|
333 |
temperature,
|
334 |
selected_model,
|
335 |
+
fake_input=old_inputs
|
336 |
)
|
337 |
yield chatbot, history, status_text, all_token_counts
|
338 |
+
|
339 |
logging.info(f"传输完毕。当前token计数为{all_token_counts}")
|
340 |
if len(history) > 1 and history[-1]["content"] != inputs:
|
341 |
logging.info(
|
|
|
344 |
+ f"{history[-1]['content']}"
|
345 |
+ colorama.Style.RESET_ALL
|
346 |
)
|
347 |
+
|
348 |
+
if use_websearch:
|
349 |
+
response = history[-1]['content']
|
350 |
+
response += "\n\n" + "\n".join(link_references)
|
351 |
+
logging.info(f"Added link references.")
|
352 |
+
logging.info(response)
|
353 |
+
chatbot[-1] = (parse_text(old_inputs), response)
|
354 |
+
yield chatbot, history, status_text, all_token_counts
|
355 |
+
|
356 |
if stream:
|
357 |
max_token = max_token_streaming
|
358 |
else:
|
359 |
max_token = max_token_all
|
360 |
+
|
361 |
if sum(all_token_counts) > max_token and should_check_token_count:
|
362 |
status_text = f"精简token中{all_token_counts}/{max_token}"
|
363 |
logging.info(status_text)
|