Update app.py
Browse files
app.py
CHANGED
@@ -4,7 +4,8 @@ from huggingface_hub import InferenceClient
|
|
4 |
"""
|
5 |
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
|
6 |
"""
|
7 |
-
client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
|
|
|
8 |
|
9 |
|
10 |
def respond(
|
@@ -27,17 +28,49 @@ def respond(
|
|
27 |
|
28 |
response = ""
|
29 |
|
30 |
-
for message in client.chat_completion(
|
31 |
-
messages,
|
32 |
-
max_tokens=max_tokens,
|
33 |
-
stream=True,
|
34 |
-
temperature=temperature,
|
35 |
-
top_p=top_p,
|
36 |
-
):
|
37 |
-
token = message.choices[0].delta.content
|
38 |
|
39 |
-
|
40 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
41 |
|
42 |
"""
|
43 |
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
|
@@ -45,7 +78,7 @@ For information on how to customize the ChatInterface, peruse the gradio docs: h
|
|
45 |
demo = gr.ChatInterface(
|
46 |
respond,
|
47 |
additional_inputs=[
|
48 |
-
gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
|
49 |
gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
|
50 |
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
|
51 |
gr.Slider(
|
|
|
4 |
"""
|
5 |
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
|
6 |
"""
|
7 |
+
#client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
|
8 |
+
client = InferenceClient("unsloth/Llama-3.2-1B-Instruct")
|
9 |
|
10 |
|
11 |
def respond(
|
|
|
28 |
|
29 |
response = ""
|
30 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
31 |
|
32 |
+
try:
|
33 |
+
for message in client.chat_completion(
|
34 |
+
messages,
|
35 |
+
max_tokens=max_tokens,
|
36 |
+
stream=True,
|
37 |
+
temperature=temperature,
|
38 |
+
top_p=top_p,
|
39 |
+
):
|
40 |
+
# Ensure the message has a valid structure
|
41 |
+
if not message or not isinstance(message, dict):
|
42 |
+
continue
|
43 |
+
|
44 |
+
try:
|
45 |
+
# Extract content and finish reason
|
46 |
+
content = message.choices[0].delta.content
|
47 |
+
finish_reason = message.choices[0].finish_reason
|
48 |
+
|
49 |
+
# Check if the content is empty
|
50 |
+
if content.strip() == "":
|
51 |
+
# If the finish reason is 'stop', it's expected and we can break the loop
|
52 |
+
if finish_reason == "stop":
|
53 |
+
print("Stream ended normally.")
|
54 |
+
break
|
55 |
+
else:
|
56 |
+
print("Received unexpected empty content, skipping...")
|
57 |
+
continue
|
58 |
+
|
59 |
+
response += content
|
60 |
+
yield response
|
61 |
+
|
62 |
+
except (AttributeError, IndexError, KeyError) as e:
|
63 |
+
print(f"Error processing message: {e}")
|
64 |
+
continue
|
65 |
+
|
66 |
+
except Exception as e:
|
67 |
+
print(f"Unexpected error: {e}")
|
68 |
+
yield "An error occurred while generating the response."
|
69 |
+
|
70 |
+
# Final check if the response is empty
|
71 |
+
if response.strip() == "":
|
72 |
+
yield "No response generated. Please try again or adjust the settings."
|
73 |
+
|
74 |
|
75 |
"""
|
76 |
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
|
|
|
78 |
demo = gr.ChatInterface(
|
79 |
respond,
|
80 |
additional_inputs=[
|
81 |
+
gr.Textbox(value="You are a friendly Chatbot. Your name is Juninho.", label="System message"),
|
82 |
gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
|
83 |
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
|
84 |
gr.Slider(
|