rot-bot / app.py
kellydoesstuff's picture
Update app.py
3704f6e verified
import gradio as gr
from sentence_transformers import SentenceTransformer, util
import openai
import os
HF_TOKEN = os.getenv('HF_TOKEN')
hf_writer =gr.HuggingFaceDatasetSaver(HF_TOKEN, "kellydoesstuff/RotBot_Flags")
os.environ["TOKENIZERS_PARALLELISM"] = "false"
# Initialize paths and model identifiers for easy configuration and maintenance
filename = "output_chess_details.txt" # Path to the file storing chess-specific details
retrieval_model_name = 'output/sentence-transformer-finetuned/'
openai.api_key = os.environ["OPENAI_API_KEY"]
# Attempt to load the necessary models and provide feedback on success or failure
try:
retrieval_model = SentenceTransformer(retrieval_model_name)
print("Models loaded successfully.")
except Exception as e:
print(f"Failed to load models: {e}")
def load_and_preprocess_text(filename):
"""
Load and preprocess text from a file, removing empty lines and stripping whitespace.
"""
try:
with open(filename, 'r', encoding='utf-8') as file:
segments = [line.strip() for line in file if line.strip()]
print("Text loaded and preprocessed successfully.")
return segments
except Exception as e:
print(f"Failed to load or preprocess text: {e}")
return []
segments = load_and_preprocess_text(filename)
def find_relevant_segment(user_query, segments):
"""
Find the most relevant text segment for a user's query using cosine similarity among sentence embeddings.
This version finds the best match based on the content of the query.
"""
try:
# Lowercase the query for better matching
lower_query = user_query.lower()
# Encode the query and the segments
query_embedding = retrieval_model.encode(lower_query)
segment_embeddings = retrieval_model.encode(segments)
# Compute cosine similarities between the query and the segments
similarities = util.pytorch_cos_sim(query_embedding, segment_embeddings)[0]
# Find the index of the most similar segment
best_idx = similarities.argmax()
# Return the most relevant segment
return segments[best_idx]
except Exception as e:
print(f"Error in finding relevant segment: {e}")
return ""
def generate_response(user_query, relevant_segment):
"""
Generate a response emphasizing the bot's capability in providing chess information.
"""
try:
# system_message = "You are a chess chatbot specialized in providing information on chess rules, strategies, and terminology."
system_message = "You are a chatbot that is specialized in translating Gen-Z/Gen-Alpha slang into standard English. In addition to translation, you are knowledgable on synonyms and origins of slang words. You not only act as a dictionary/thesaurus for slang words, you can translate sentences with slang words into standard Englsih."
# system_message = """ You are a chatbot that translates slang English, I'm talking sentences with words like rizz and gyatt, into professional standard English.
# We want a translation that sounds like a sentence from a LinkedIn post. You translate sentences with multiple slang words into sentences with comprehensible standard English.
# Not only that, you are also knowledgeable on definitions, synonyms, and origins of slang words if prompted.
# So,if someone asks you to translate a sentence into standard English like, 'What does the sentence She has the biggest gyatt on god! mean?', you would respond with 'This sentence translates to she has the biggest butt I swear to god!'
# Or, if someone asks you to translate the sentence 'What the sigma! She just ghosted me!', you would translate that to 'What the hell! She is ignoring me!'
# If the output is inappropriate warn the user. DO NOT RESPOND WITH THE SAME TEXT PASSED TO YOU. YOU ARE A TRANSLATOR NOT A CONVERSATIONALIST.
# """
user_message = f"Here's the question the user inputted: {relevant_segment}"
messages = [
{"role": "system", "content": system_message},
{"role": "user", "content": user_message}
]
response = openai.ChatCompletion.create(
model="gpt-4o",
messages=messages,
max_tokens=200,
temperature=0.2,
top_p=1,
frequency_penalty=0,
presence_penalty=0
)
return response['choices'][0]['message']['content'].strip()
except Exception as e:
print(f"Error in generating response: {e}")
return f"Error in generating response: {e}"
def query_model(question, history):
"""
Process a question, find relevant information, and generate a response.
"""
# global question_g
# question_g = question
if question == "":
return "What the sigma! You didn't ask a question. Feel free to ask me anything about the topics listed above."
relevant_segment = find_relevant_segment(question, segments)
if not relevant_segment:
return "Unfortunately I cannot answer your question..😔 Try refining your question so I can try again."
# global response_g
response = generate_response(question, relevant_segment)
return response
# Define the welcome message and specific topics the chatbot can provide information about
welcome_message = """
# 🗑️ Welcome to RotBot!
## Your AI assistant for translating slang into standard English!
"""
topics = """
### Feel Free to ask me anything from the topics below!
- Translating slang sentences into standard English
- Defining slang into standard English
- Providing standard English synonyms for slang
"""
# callback = gr.CSVLogger()
# Setup the Gradio Blocks interface with custom layout components
with gr.Blocks(theme='gradio/soft') as demo:
gr.Markdown(welcome_message) # Display the formatted welcome message
with gr.Row():
with gr.Column():
gr.Markdown(topics) # Show the topics on the left side
# with gr.Row():
# with gr.Column():
# question = gr.Textbox(label="Your question", placeholder="What do you want to ask about?")
# answer = gr.Textbox(label="RotBot Response", placeholder="RotBot will respond here...", interactive=False, lines=10)
# submit_button = gr.Button("Submit")
# submit_button.click(fn=query_model, inputs=question, outputs=answer)
# chatbot = gr.ChatInterface(
# fn=query_model,
# examples=["Help me translate this sentence into standard English: Stop glazing him! He isn't even that good. ", "What's a synonym for rizz?", "What is the definition of gyatt?"],
# multimodal=False,
# )
# with gr.Row():
# flag_btn = gr.Button("Flag")
# callback.setup([chatbot], "flagged_data_points")
# flag_btn.click(lambda *args: callback.flag(args), [chatbot], None, preprocess=False)
chatbot = gr.Interface(
fn=query_model,
inputs=gr.Textbox(label="Your question", placeholder="What do you want to ask about?"),
outputs=gr.Textbox(label="RotBot Response", placeholder="RotBot will respond here...", interactive=False, lines=10, show_copy_button = True
),
# outputs = gr.Markdown(label="RotBot Response"),
theme='gradio/soft',
examples=[
["Help me translate this sentence into standard English: Stop glazing him! He isn't even that good."],
["What's a synonym for rizz?"],
["What is the definition of gyatt?"]
],
allow_flagging="manual",
# flagging_dir = "flag",
flagging_options=["Ambiguous", "Wrong", "Other"],
flagging_callback=hf_writer
# show_copy_button = True
)
# Launch the Gradio app to allow user interaction
demo.launch(share=True)