Spaces:
Running
on
Zero
Running
on
Zero
Update app.py
Browse files
app.py
CHANGED
@@ -6,7 +6,6 @@ import spaces
|
|
6 |
import torch
|
7 |
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
|
8 |
from typing import List, Dict, Optional, Tuple
|
9 |
-
from http import HTTPStatus
|
10 |
|
11 |
DESCRIPTION = """
|
12 |
# QwQ Distill
|
@@ -88,7 +87,7 @@ def generate(
|
|
88 |
top_p: float = 0.9,
|
89 |
top_k: int = 50,
|
90 |
repetition_penalty: float = 1.2,
|
91 |
-
) -> Iterator[
|
92 |
if query is None:
|
93 |
query = ''
|
94 |
if history is None:
|
@@ -135,11 +134,7 @@ def generate(
|
|
135 |
outputs = []
|
136 |
for text in streamer:
|
137 |
outputs.append(text)
|
138 |
-
|
139 |
-
# Update history with the new response
|
140 |
-
new_messages = messages + [{'role': Role.ASSISTANT, 'content': response}]
|
141 |
-
system, new_history = messages_to_history(new_messages)
|
142 |
-
yield "", new_history, system
|
143 |
|
144 |
|
145 |
demo = gr.ChatInterface(
|
|
|
6 |
import torch
|
7 |
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
|
8 |
from typing import List, Dict, Optional, Tuple
|
|
|
9 |
|
10 |
DESCRIPTION = """
|
11 |
# QwQ Distill
|
|
|
87 |
top_p: float = 0.9,
|
88 |
top_k: int = 50,
|
89 |
repetition_penalty: float = 1.2,
|
90 |
+
) -> Iterator[str]:
|
91 |
if query is None:
|
92 |
query = ''
|
93 |
if history is None:
|
|
|
134 |
outputs = []
|
135 |
for text in streamer:
|
136 |
outputs.append(text)
|
137 |
+
yield "".join(outputs)
|
|
|
|
|
|
|
|
|
138 |
|
139 |
|
140 |
demo = gr.ChatInterface(
|