Spaces:
Running
Running
jonathanjordan21
commited on
Commit
•
28bce37
1
Parent(s):
ca5271c
Update apis/chat_api.py
Browse files- apis/chat_api.py +34 -28
apis/chat_api.py
CHANGED
@@ -155,32 +155,33 @@ class ChatAPIApp:
|
|
155 |
default=False,
|
156 |
description="(bool) Stream",
|
157 |
)
|
158 |
-
options: dict = Field(
|
159 |
-
|
160 |
-
|
161 |
-
|
162 |
-
|
163 |
-
|
164 |
-
|
165 |
-
|
166 |
-
)
|
167 |
-
# temperature: Union[float, None] = Field(
|
168 |
-
# default=0.5,
|
169 |
-
# description="(float) Temperature",
|
170 |
-
# )
|
171 |
-
# top_p: Union[float, None] = Field(
|
172 |
-
# default=0.95,
|
173 |
-
# description="(float) top p",
|
174 |
-
# )
|
175 |
-
# max_tokens: Union[int, None] = Field(
|
176 |
-
# default=-1,
|
177 |
-
# description="(int) Max tokens",
|
178 |
-
# )
|
179 |
-
# use_cache: bool = Field(
|
180 |
-
# default=False,
|
181 |
-
# description="(bool) Use cache",
|
182 |
# )
|
183 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
184 |
|
185 |
def generate_text(
|
186 |
self, item: GenerateRequest, api_key: str = Depends(extract_api_key)
|
@@ -200,11 +201,16 @@ class ChatAPIApp:
|
|
200 |
streamer = HuggingfaceStreamer(model=item.model)
|
201 |
stream_response = streamer.chat_response(
|
202 |
prompt=item.prompt,
|
203 |
-
temperature=item.
|
204 |
-
top_p=item.
|
205 |
-
max_new_tokens=item.
|
206 |
api_key=api_key,
|
207 |
-
use_cache=item.
|
|
|
|
|
|
|
|
|
|
|
208 |
)
|
209 |
|
210 |
if item.stream:
|
|
|
155 |
default=False,
|
156 |
description="(bool) Stream",
|
157 |
)
|
158 |
+
# options: dict = Field(
|
159 |
+
# default={
|
160 |
+
# "temperature":0.5,
|
161 |
+
# "top_p":0.95,
|
162 |
+
# "max_tokens":-1,
|
163 |
+
# "use_cache":False
|
164 |
+
# },
|
165 |
+
# description="(dict) Options"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
166 |
# )
|
167 |
|
168 |
+
temperature: Union[float, None] = Field(
|
169 |
+
default=0.5,
|
170 |
+
description="(float) Temperature",
|
171 |
+
)
|
172 |
+
top_p: Union[float, None] = Field(
|
173 |
+
default=0.95,
|
174 |
+
description="(float) top p",
|
175 |
+
)
|
176 |
+
max_tokens: Union[int, None] = Field(
|
177 |
+
default=-1,
|
178 |
+
description="(int) Max tokens",
|
179 |
+
)
|
180 |
+
use_cache: bool = Field(
|
181 |
+
default=False,
|
182 |
+
description="(bool) Use cache",
|
183 |
+
)
|
184 |
+
|
185 |
|
186 |
def generate_text(
|
187 |
self, item: GenerateRequest, api_key: str = Depends(extract_api_key)
|
|
|
201 |
streamer = HuggingfaceStreamer(model=item.model)
|
202 |
stream_response = streamer.chat_response(
|
203 |
prompt=item.prompt,
|
204 |
+
temperature=item.temperature,
|
205 |
+
top_p=item.top_p,
|
206 |
+
max_new_tokens=item.max_tokens,
|
207 |
api_key=api_key,
|
208 |
+
use_cache=item.use_cache,
|
209 |
+
# temperature=item.options.get('temperature', 0.6),
|
210 |
+
# top_p=item.options.get('top_p', 0.95),
|
211 |
+
# max_new_tokens=item.options.get('max_new_tokens', -1),
|
212 |
+
# api_key=api_key,
|
213 |
+
# use_cache=item.options.get('use_cache', False),
|
214 |
)
|
215 |
|
216 |
if item.stream:
|