Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -13,6 +13,7 @@ from share_btn import community_icon_html, loading_icon_html, share_js, share_bt
|
|
13 |
HF_TOKEN = os.environ.get("HF_TOKEN", None)
|
14 |
|
15 |
API_URL_G = "https://api-inference.huggingface.co/models/ArmelR/starcoder-gradio-v0/"
|
|
|
16 |
|
17 |
with open("./HHH_prompt_short.txt", "r") as f:
|
18 |
HHH_PROMPT = f.read() + "\n\n"
|
@@ -53,6 +54,10 @@ client_g = Client(
|
|
53 |
API_URL_G, headers={"Authorization": f"Bearer {HF_TOKEN}"},
|
54 |
)
|
55 |
|
|
|
|
|
|
|
|
|
56 |
def generate(
|
57 |
prompt,
|
58 |
temperature=0.9,
|
@@ -60,7 +65,7 @@ def generate(
|
|
60 |
top_p=0.95,
|
61 |
repetition_penalty=1.0,
|
62 |
chat_mode="TA prompt",
|
63 |
-
version=
|
64 |
):
|
65 |
|
66 |
temperature = float(temperature)
|
@@ -90,14 +95,19 @@ def generate(
|
|
90 |
chat_prompt = prompt + "\n\nAnswer:"
|
91 |
prompt = base_prompt + chat_prompt
|
92 |
|
93 |
-
|
|
|
|
|
|
|
|
|
|
|
94 |
|
95 |
output = ""
|
96 |
previous_token = ""
|
97 |
|
98 |
for response in stream:
|
99 |
if (
|
100 |
-
(response.token.text in ["
|
101 |
and previous_token in ["\n", "-----"])
|
102 |
or response.token.text == "<|endoftext|>"
|
103 |
):
|
@@ -121,12 +131,17 @@ def bot(
|
|
121 |
top_p=0.95,
|
122 |
repetition_penalty=1.0,
|
123 |
chat_mode=None,
|
124 |
-
version=
|
125 |
):
|
126 |
# concat history of prompts with answers expect for last empty answer only add prompt
|
127 |
-
|
128 |
-
|
129 |
-
|
|
|
|
|
|
|
|
|
|
|
130 |
|
131 |
bot_message = generate(
|
132 |
prompt,
|
@@ -210,12 +225,12 @@ _Note:_ this is an internal chat playground - **please do not share**. The deplo
|
|
210 |
interactive=True,
|
211 |
info="Penalize repeated tokens",
|
212 |
)
|
213 |
-
|
214 |
-
|
215 |
-
|
216 |
-
|
217 |
-
|
218 |
-
|
219 |
with column_1:
|
220 |
# output = gr.Code(elem_id="q-output")
|
221 |
# add visibl=False and update if chat_mode True
|
@@ -251,7 +266,7 @@ _Note:_ this is an internal chat playground - **please do not share**. The deplo
|
|
251 |
user, [instruction, chatbot], [instruction, chatbot], queue=False
|
252 |
).then(
|
253 |
bot,
|
254 |
-
[chatbot, temperature, max_new_tokens, top_p, repetition_penalty, chat_mode],
|
255 |
chatbot,
|
256 |
)
|
257 |
|
@@ -259,7 +274,7 @@ _Note:_ this is an internal chat playground - **please do not share**. The deplo
|
|
259 |
user, [instruction, chatbot], [instruction, chatbot], queue=False
|
260 |
).then(
|
261 |
bot,
|
262 |
-
[chatbot, temperature, max_new_tokens, top_p, repetition_penalty, chat_mode],
|
263 |
chatbot,
|
264 |
)
|
265 |
clear.click(lambda: None, None, chatbot, queue=False)
|
|
|
13 |
HF_TOKEN = os.environ.get("HF_TOKEN", None)
|
14 |
|
15 |
API_URL_G = "https://api-inference.huggingface.co/models/ArmelR/starcoder-gradio-v0/"
|
16 |
+
API_URL_S = "https://api-inference.huggingface.co/models/HuggingFaceH4/starcoderbase-finetuned-oasst1"
|
17 |
|
18 |
with open("./HHH_prompt_short.txt", "r") as f:
|
19 |
HHH_PROMPT = f.read() + "\n\n"
|
|
|
54 |
API_URL_G, headers={"Authorization": f"Bearer {HF_TOKEN}"},
|
55 |
)
|
56 |
|
57 |
+
client_starchat = Client(
|
58 |
+
API_URL_S, headers={"Authorization": f"Bearer {HF_TOKEN}"},
|
59 |
+
)
|
60 |
+
|
61 |
def generate(
|
62 |
prompt,
|
63 |
temperature=0.9,
|
|
|
65 |
top_p=0.95,
|
66 |
repetition_penalty=1.0,
|
67 |
chat_mode="TA prompt",
|
68 |
+
version="StarChat-alpha",
|
69 |
):
|
70 |
|
71 |
temperature = float(temperature)
|
|
|
95 |
chat_prompt = prompt + "\n\nAnswer:"
|
96 |
prompt = base_prompt + chat_prompt
|
97 |
|
98 |
+
if version == "StarCoder-gradio" :
|
99 |
+
stream = client_g.generate_stream(prompt, **generate_kwargs)
|
100 |
+
elif version == "StarChat-alpha" :
|
101 |
+
stream = client_s.generate_stream(prompt, **generate_kwargs)
|
102 |
+
else :
|
103 |
+
pass
|
104 |
|
105 |
output = ""
|
106 |
previous_token = ""
|
107 |
|
108 |
for response in stream:
|
109 |
if (
|
110 |
+
(response.token.text in ["Human", "-----", "Question:"]
|
111 |
and previous_token in ["\n", "-----"])
|
112 |
or response.token.text == "<|endoftext|>"
|
113 |
):
|
|
|
131 |
top_p=0.95,
|
132 |
repetition_penalty=1.0,
|
133 |
chat_mode=None,
|
134 |
+
version="starchat-alpha",
|
135 |
):
|
136 |
# concat history of prompts with answers expect for last empty answer only add prompt
|
137 |
+
if version == "StarCoder-gradio"
|
138 |
+
prompt = "\n".join(
|
139 |
+
[f"Question: {prompt}\n\nAnswer: {answer}" for prompt, answer in history[:-1]] + [f"\nQuestion: {history[-1][0]}"]
|
140 |
+
)
|
141 |
+
else :
|
142 |
+
prompt = "\n".join(
|
143 |
+
[f"Human: {prompt}\n\nAssistant: {answer}" for prompt, answer in history[:-1]] + [f"\nHuman: {history[-1][0]}"]
|
144 |
+
)
|
145 |
|
146 |
bot_message = generate(
|
147 |
prompt,
|
|
|
225 |
interactive=True,
|
226 |
info="Penalize repeated tokens",
|
227 |
)
|
228 |
+
version = gr.Dropdown(
|
229 |
+
["StarCoder-gradio", "StarChat-alpha"],
|
230 |
+
value="StarCoderBase",
|
231 |
+
label="Version",
|
232 |
+
info="",
|
233 |
+
)
|
234 |
with column_1:
|
235 |
# output = gr.Code(elem_id="q-output")
|
236 |
# add visibl=False and update if chat_mode True
|
|
|
266 |
user, [instruction, chatbot], [instruction, chatbot], queue=False
|
267 |
).then(
|
268 |
bot,
|
269 |
+
[chatbot, temperature, max_new_tokens, top_p, repetition_penalty, chat_mode, version],
|
270 |
chatbot,
|
271 |
)
|
272 |
|
|
|
274 |
user, [instruction, chatbot], [instruction, chatbot], queue=False
|
275 |
).then(
|
276 |
bot,
|
277 |
+
[chatbot, temperature, max_new_tokens, top_p, repetition_penalty, chat_mode, version],
|
278 |
chatbot,
|
279 |
)
|
280 |
clear.click(lambda: None, None, chatbot, queue=False)
|