Update app.py
Browse files
app.py
CHANGED
@@ -1,29 +1,27 @@
|
|
1 |
from huggingface_hub import InferenceClient
|
2 |
import gradio as gr
|
3 |
|
4 |
-
|
5 |
-
|
|
|
|
|
6 |
|
7 |
def format_prompt(message, history):
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
|
12 |
-
|
13 |
-
|
14 |
|
15 |
def generate(
|
16 |
prompt, history, temperature=0.9, max_new_tokens=1024, top_p=0.95, repetition_penalty=1.0,
|
17 |
-
model_name=None
|
18 |
):
|
19 |
temperature = float(temperature)
|
20 |
if temperature < 1e-2:
|
21 |
temperature = 1e-2
|
22 |
top_p = float(top_p)
|
23 |
|
24 |
-
# Initialize the InferenceClient with the selected model
|
25 |
-
client = InferenceClient(model_name)
|
26 |
-
|
27 |
generate_kwargs = dict(
|
28 |
temperature=temperature,
|
29 |
max_new_tokens=max_new_tokens,
|
@@ -43,8 +41,8 @@ def generate(
|
|
43 |
yield output
|
44 |
return output
|
45 |
|
|
|
46 |
additional_inputs=[
|
47 |
-
gr.Textbox(label="Model Name", value=initial_model_name, placeholder="Enter model name here"),
|
48 |
gr.Slider(
|
49 |
label="Temperature",
|
50 |
value=0.9,
|
@@ -83,18 +81,9 @@ additional_inputs=[
|
|
83 |
)
|
84 |
]
|
85 |
|
86 |
-
gr.
|
87 |
fn=generate,
|
88 |
-
|
89 |
-
|
90 |
-
|
91 |
-
|
92 |
-
MaxNewTokens=additional_inputs[1],
|
93 |
-
TopP=additional_inputs[2],
|
94 |
-
RepetitionPenalty=additional_inputs[3],
|
95 |
-
ModelName=additional_inputs[4],
|
96 |
-
),
|
97 |
-
outputs="text",
|
98 |
-
title="Interactive Chat with AI Models",
|
99 |
-
description="Type a message and see the AI's response. Adjust parameters to change the behavior.",
|
100 |
-
).launch()
|
|
|
1 |
from huggingface_hub import InferenceClient
|
2 |
import gradio as gr
|
3 |
|
4 |
+
client = InferenceClient(
|
5 |
+
'pthornton614/CodeLama-7b-Instruct'
|
6 |
+
)
|
7 |
+
|
8 |
|
9 |
def format_prompt(message, history):
|
10 |
+
prompt = "<s>"
|
11 |
+
for user_prompt, bot_response in history:
|
12 |
+
prompt += f"[INST] {user_prompt} [/INST]"
|
13 |
+
prompt += f" {bot_response}</s> "
|
14 |
+
prompt += f"[INST] {message} [/INST]"
|
15 |
+
return prompt
|
16 |
|
17 |
def generate(
|
18 |
prompt, history, temperature=0.9, max_new_tokens=1024, top_p=0.95, repetition_penalty=1.0,
|
|
|
19 |
):
|
20 |
temperature = float(temperature)
|
21 |
if temperature < 1e-2:
|
22 |
temperature = 1e-2
|
23 |
top_p = float(top_p)
|
24 |
|
|
|
|
|
|
|
25 |
generate_kwargs = dict(
|
26 |
temperature=temperature,
|
27 |
max_new_tokens=max_new_tokens,
|
|
|
41 |
yield output
|
42 |
return output
|
43 |
|
44 |
+
|
45 |
additional_inputs=[
|
|
|
46 |
gr.Slider(
|
47 |
label="Temperature",
|
48 |
value=0.9,
|
|
|
81 |
)
|
82 |
]
|
83 |
|
84 |
+
gr.ChatInterface(
|
85 |
fn=generate,
|
86 |
+
chatbot=gr.Chatbot(show_label=False, show_share_button=False, show_copy_button=True, layout="panel"),
|
87 |
+
additional_inputs=additional_inputs,
|
88 |
+
title="""Mixtral-8x7B"""
|
89 |
+
).queue().launch(show_api=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|