Spaces:
Running
Running
File size: 4,056 Bytes
dfbf21d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 |
import gradio as gr
import src.srf_bot as sb
import prompts.system_prompts as sp
from langchain_core.messages import HumanMessage
# Initialize the chatbot
chatbot = sb.SRFChatbot()
# Dictionary to store passages with identifiers
retrieved_passages = {}
# Define the respond function
def respond(query, history):
formatted_query = [HumanMessage(content=query)]
# Invoke the chatbot
result = chatbot.graph.invoke({"messages": formatted_query}, chatbot.config)
# Extract the assistant's response
response = result["messages"][-1].content
# Retrieve passages from your vector database based on the query
# For the example, we'll use dummy passages
passages = [
"This is the full text of Passage 1.",
"This is the full text of Passage 2.",
"This is the full text of Passage 3."
]
# Store passages with identifiers
passage_ids = []
for idx, passage in enumerate(passages):
identifier = f"Passage {idx+1}"
retrieved_passages[identifier] = passage
passage_ids.append(identifier)
# Reference passages in the response
linked_response = f"{response}\n\nReferences:"
for pid in passage_ids:
linked_response += f" [{pid}]"
# Append to history
history.append((query, linked_response))
return history, ""
# Function to get passage content based on selection
def get_passage_content(passage_id):
return retrieved_passages.get(passage_id, "Passage not found.")
# Function to update the system prompt
def update_system_prompt(selected_prompt):
# Update the chatbot's system prompt
chatbot.reset_system_prompt(selected_prompt)
# Update the displayed system prompt text
return sp.system_prompt_templates[selected_prompt]
# Gradio interface
with gr.Blocks() as demo:
gr.Markdown("# SRF Chatbot")
with gr.Row():
with gr.Column(scale=4):
# Chatbot interface
chatbot_output = gr.Chatbot()
user_input = gr.Textbox(placeholder="Type your question here...", label="Your Question")
submit_button = gr.Button("Submit")
with gr.Column(scale=1):
# Dropdown to select system prompts
system_prompt_dropdown = gr.Dropdown(
choices=list(sp.system_prompt_templates.keys()),
label="Select Chatbot Instructions",
value=list(sp.system_prompt_templates.keys())[0]
)
# Display the selected system prompt
system_prompt_display = gr.Textbox(
value=sp.system_prompt_templates[list(sp.system_prompt_templates.keys())[0]],
label="Current Chatbot Instructions",
lines=5,
interactive=False
)
# Update system prompt display when a new prompt is selected
system_prompt_dropdown.change(
fn=update_system_prompt,
inputs=[system_prompt_dropdown],
outputs=[system_prompt_display]
)
# Passage selection and display
gr.Markdown("### References")
passage_selector = gr.Dropdown(label="Select a passage to view", choices=[])
passage_display = gr.Markdown()
# Update the chatbot when the submit button is clicked
submit_button.click(
fn=respond,
inputs=[user_input, chatbot_output],
outputs=[chatbot_output, user_input]
)
# Update the passage selector options when the chatbot output changes
def update_passage_selector(chat_history):
# Get the latest passages
choices = list(retrieved_passages.keys())
return gr.update(choices=choices)
chatbot_output.change(
fn=update_passage_selector,
inputs=[chatbot_output],
outputs=[passage_selector]
)
# Display the selected passage
passage_selector.change(
fn=get_passage_content,
inputs=[passage_selector],
outputs=[passage_display]
)
demo.launch()
|