Created app

README.md

@@ -1,12 +1,12 @@
 ---
-title: Knowledge Center Assistant
-emoji: 📊
-colorFrom: blue
-colorTo: green
+title: Operative Iq
+emoji: 💬
+colorFrom: yellow
+colorTo: purple
 sdk: gradio
-sdk_version: 5.3.0
+sdk_version: 4.36.1
 app_file: app.py
 pinned: false
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+An example chatbot using [Gradio](https://gradio.app), [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/v0.22.2/en/index), and the [Hugging Face Inference API](https://huggingface.co/docs/api-inference/index).

app.py

@@ -0,0 +1,193 @@
+import os
+import boto3
+import gradio as gr
+import math
+import json
+import time
+import re
+from botocore.client import Config
+
+kb_id = os.getenv('KNOWLEDGE_BASE_ID')
+aws_access_key = os.getenv('AWS_ACCESS_KEY_ID')
+aws_secret_key = os.getenv('AWS_SECRET_ACCESS_KEY')
+region = os.getenv('AWS_REGION')
+initial_message = os.getenv('INITIAL_MESSAGE')
+
+amazon_model_id = "amazon.titan-text-premier-v1:0"
+
+bedrock_config = Config(connect_timeout=120, read_timeout=120, retries={'max_attempts': 0})
+bedrock_client = boto3.client(
+    'bedrock-runtime', 
+    region_name=region, 
+    aws_access_key_id=aws_access_key, 
+    aws_secret_access_key=aws_secret_key
+)
+bedrock_agent_client = boto3.client(
+    "bedrock-agent-runtime", 
+    config=bedrock_config, 
+    region_name=region, 
+    aws_access_key_id=aws_access_key, 
+    aws_secret_access_key=aws_secret_key
+)
+
+def retrieve(query, numberOfResults=4):
+    start_time = time.time() 
+    response = bedrock_agent_client.retrieve(
+        retrievalQuery= {
+            'text': query
+        },
+        knowledgeBaseId=kb_id,
+        retrievalConfiguration= {
+            'vectorSearchConfiguration': {
+                'numberOfResults': numberOfResults,
+                'overrideSearchType': "SEMANTIC", #"HYBRID",
+            }
+        }
+    )
+    end_time = time.time()  
+    retrieve_execution_time = end_time - start_time  
+
+    return response['retrievalResults'], retrieve_execution_time
+
+def get_contexts(retrievalResults):
+    contexts = ""
+    for retrievedResult in retrievalResults: 
+        contexts += retrievedResult['content']['text'] + '\n'
+    return contexts
+
+def remove_link(text):
+    pattern = r"\n\nFor more information, follow the link provided: https:\/\/knowledge\.operativeiq\.com\/articles\/\d{9}"
+    cleaned_text = re.sub(pattern, '', text)
+    return cleaned_text
+
+def get_answer(query, history, temperature, top_p, max_token_count):
+    history = remove_link(history)
+    contexts = ""
+    article_url_text = ""    
+    max_words = math.floor(max_token_count*0.75)
+
+    retrievalResults, retrieve_execution_time = retrieve(query)
+    highest_score = retrievalResults[0]['score']
+
+    if highest_score > 0.45:
+        contexts = get_contexts(retrievalResults)
+        article_url = f"https://knowledge.operativeiq.com/articles/{retrievalResults[0]['metadata'].get('article_id')}"
+        article_url_text = f"\n\nFor more information, follow the link provided: {article_url}"
+    
+    PROMPT_TEMPLATE = f""" 
+    System: You are an intelligent assistant helping users understand and navigate website functionalities.
+    Your goal is to provide clear, accurate, and contextually relevant answers based on the information provided.
+
+    Use the information enclosed in the <context> tags and refer to the conversation history in the <history> tags to answer the user's question in the <question> tags. 
+    If you don't know the answer, just say that you don't know, don't try to make up an answer.
+    
+    Your response must:
+    - Be fully formed and grammatically correct without cutting off any sentences.
+    - Complete a logical thought or sentence before stopping, ensuring the response doesn't end mid-sentence.
+    - Be clear, easy to understand, and succinct, not exceeding {max_words} words.
+    - Refer specifically to website features or actions when relevant to the user's question.
+    - Avoid providing URL links or external references.
+    
+    <history>
+    {history}
+    </history>
+
+    <context>
+    {contexts}
+    </context>
+
+    <question>
+    {query}
+    </question>
+
+    Provide a detailed, concise response that fully answers the user's question.
+    Make sure all sentences of your reponse are completely formed and grammatically correct.
+    If necessary, reduce the amount of detail provided to keep the response within the word limit but still complete.
+
+    Assistant:
+    """
+
+    body = json.dumps({
+        "inputText": PROMPT_TEMPLATE,
+        "textGenerationConfig": {
+            "maxTokenCount": max_token_count,
+            "temperature": temperature,
+            "topP": top_p
+        }
+    })
+
+    kwargs = {
+        "modelId": "amazon.titan-text-premier-v1:0",
+        "contentType": "application/json",
+        "accept": "*/*",
+        "body": body
+    }
+    
+    start_time = time.time()  
+
+    response = bedrock_client.invoke_model(**kwargs)
+    
+    end_time = time.time()  
+    invoke_model_time = end_time - start_time
+
+    response_body = json.loads(response.get('body').read())
+    response_text = response_body['results'][0]['outputText']
+    
+    response_text += article_url_text
+
+    prompt_and_time = f"""
+    Prompt:
+    {PROMPT_TEMPLATE}
+
+    Retrieve execution time: {retrieve_execution_time} seconds
+    Invoke model execution time: {invoke_model_time} seconds
+    """
+    return response_text, prompt_and_time
+
+def format_chat_history(chat_history):
+    prompt = ""
+    for turn in chat_history:
+        user_message, bot_message = turn
+        prompt = f"{prompt}User: {user_message}\nAssistant: {bot_message}\n"
+    return prompt
+
+def respond(message, chat_history, temperature=0.9, top_p=0.6, max_token_count=512):
+    formatted_history = format_chat_history(chat_history)
+    chat_history.append([message, ""])
+
+    stream, prompt_and_time = get_answer(message, formatted_history, temperature, top_p, max_token_count)
+    
+    for idx, text_token in enumerate(stream):
+        if idx == 0 and text_token.startswith(" "):
+            text_token = text_token[1:]
+
+        chat_history[-1][1] += text_token
+        yield "", chat_history, prompt_and_time
+    
+def clear_chat_history():
+    return '', []
+
+def main():
+    with gr.Blocks() as demo:
+        chatbot = gr.Chatbot([[None, initial_message]], height=550)
+
+        msg = gr.Textbox(label="Question")
+        
+        with gr.Accordion(label="Advanced options", open=False):
+            temperature = gr.Slider(label="Temperature", minimum=0.1, maximum=1, value=0.6, step=0.1)
+            top_p = gr.Slider(label="Top P", minimum=0.1, maximum=1, value=0.5, step=0.1)
+            max_token_count = gr.Slider(label="Max token count", minimum=1, maximum=1024, value=400, step=10)
+            prompt_and_time = gr.Textbox(label="Prompt and Time", interactive=False)
+        
+        btn = gr.Button("Submit")
+        clear = gr.Button("Clear console")
+
+        btn.click(respond, inputs=[msg, chatbot, temperature, top_p, max_token_count], outputs=[msg, chatbot, prompt_and_time])
+        msg.submit(respond, inputs=[msg, chatbot, temperature, top_p, max_token_count], outputs=[msg, chatbot, prompt_and_time])
+
+        clear.click(clear_chat_history, outputs=[msg, chatbot, prompt_and_time])
+    
+    demo.queue().launch()
+
+if __name__ == "__main__":
+    main()

gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

requirements.txt

@@ -0,0 +1,4 @@
+huggingface_hub==0.22.2
+boto3==1.34.161
+gradio==4.41.0
+botocore==1.34.161