Upload app.py

app.py

@@ -0,0 +1,139 @@
+
+
+#Import the necessary Libraries
+import os
+import uuid
+import json
+import gradio as gr
+from openai import OpenAI
+from langchain_community.embeddings.sentence_transformer import SentenceTransformerEmbeddings
+from langchain_community.vectorstores import Chroma
+from huggingface_hub import CommitScheduler
+from pathlib import Path
+from dotenv import load_dotenv
+
+# Create Client
+load_dotenv()
+
+os.environ['OPENAI_API_KEY'] = anyscale_api_key
+
+client = OpenAI(
+    base_url="https://api.endpoints.anyscale.com/v1",
+    api_key=os.environ['OPENAI_API_KEY']
+)
+
+# Define the embedding model and the vectorstore
+embedding_model = SentenceTransformerEmbeddings(model_name='thenlper/gte-large')
+
+# Load the persisted vectorDB
+collection_name = 'picco-2023'
+
+vectorstore_persisted = Chroma(
+    collection_name=collection_name,
+    embedding_function=embedding_model,
+    persist_directory='/content/picco_db'
+)
+
+retriever = vectorstore_persisted.as_retriever(
+    search_type="similarity",
+    search_kwargs={'k': 5},
+)
+
+# Prepare the logging functionality
+log_file = Path("logs/") / f"data_{uuid.uuid4()}.json"
+log_folder = log_file.parent
+
+scheduler = CommitScheduler(
+    repo_id="picco-qna",
+    repo_type="dataset",
+    folder_path=log_folder,
+    path_in_repo="data",
+    every=2
+
+# Define the Q&A system message
+qna_system_message = """
+You are an academic assistant who answers questions regarding pulse integrated continuous cardiac output (PiCCO) for learners preparing for PiCCO workshop .
+User input will have the context required by you to answer user questions.
+This context will begin with the token: ###Context.
+The context contains references to specific portions of a document relevant to the user query.
+
+User questions will begin with the token: ###Question.
+
+Please answer only using the context provided in the input. Do not mention anything about the context in your final answer.
+
+If the answer is not found in the context, respond "I don't know".
+"""
+
+# Define the user message template
+qna_user_message_template = """
+###Context
+Here are some documents that are relevant to the question mentioned below.-
+{context}
+
+###Question
+{question}
+"""
+
+# Define the predict function that runs when 'Submit' is clicked or when an API request is made
+def predict(user_input):
+    #filter = {"source": f"/content/dataset/{company}-10-k-2023.pdf"}
+    relevant_document_chunks = vectorstore_persisted.similarity_search(user_input, k=5)
+
+
+    # Create context_for_query
+    context_list = [f"Page {doc.metadata['page']}: {doc.page_content}" for doc in relevant_document_chunks]
+    context_for_query = ".".join(context_list)
+
+    # Create messages
+    prompt = [
+        {'role': 'system', 'content': qna_system_message},
+        {'role': 'user', 'content': qna_user_message_template.format(context=context_for_query, question=user_input)}
+    ]
+
+    # Get response from the LLM
+    try:
+        response = client.chat.completions.create(
+            model="mlabonne/NeuralHermes-2.5-Mistral-7B",
+            messages=prompt,
+            temperature=0
+        )
+        prediction = response.choices[0].message.content
+    except Exception as e:
+        prediction = f'Sorry, I encountered the following error: \n {e}'
+
+    print(prediction)
+
+    # Log both the inputs and outputs to a local log file
+    with scheduler.lock:
+        with log_file.open("a") as f:
+            f.write(json.dumps({
+                'user_input': user_input,
+                'retrieved_context': context_for_query,
+                'model_response': prediction
+            }))
+            f.write("\n")
+
+    return prediction
+
+# Set-up the Gradio UI
+
+# Create the interface
+demo = gr.Interface(
+    inputs=gr.Textbox(placeholder="Enter your query here"),
+    fn=predict,
+    outputs="text",
+    description="This web API presents an interface to ask questions on PiCCO Technology",
+    article="Note that questions that are not relevant to PiCCO not be answered.",
+    title="KSA/CCSK Hemodynamic Workshop Question & Answer Bot, by Chiko",
+    concurrency_limit=16,
+    examples = [
+        ["State and explain the main scientific principles used in PiCCO technology"],
+        ["What is the relevance of PiCCO in the operating room and critical care unit?"],
+        ["Name and explain the parameters that can be measured using PiCCO, give their reference ranges"],
+        ["State the basic equation that is used to estimate cardiac output using pulse contour analysis, explain each variable in detail"]
+
+    ]
+)
+
+demo.queue()
+demo.launch()