Update app.py

app.py

@@ -1,187 +1,187 @@
-import os
-import gradio as gr
-import numpy as np
-from sentence_transformers import SentenceTransformer
-import faiss
-from openai import OpenAI, OpenAIError
-from langchain_community.document_loaders import PyPDFLoader
-from langchain.text_splitter import RecursiveCharacterTextSplitter
-
-# Paths for files generated by build_index.py
-INDEX_FILE  = "r_docs.index"
-CHUNKS_FILE = "r_chunks.npy"
-
-# Check index existence
-if not os.path.exists(INDEX_FILE) or not os.path.exists(CHUNKS_FILE):
-    raise FileNotFoundError(
-        "Index not found. Please run first:\n  python build_index.py"
-    )
-
-# Load FAISS index and chunks
-index  = faiss.read_index(INDEX_FILE)
-chunks = np.load(CHUNKS_FILE, allow_pickle=True)
-
-# Embedding model for context retrieval
-embedding_model = SentenceTransformer('all-MiniLM-L6-v2')
-
-def retrieve_context(query: str, k: int = 4) -> str:
-    q_emb = embedding_model.encode([query], convert_to_numpy=True)
-    _, I = index.search(q_emb, k)
-    return "\n---\n".join(chunks[i] for i in I[0])
-
-# NVIDIA OpenAI-compatible client for chat
-NV_API_KEY = os.getenv(
-    "NV_API_KEY", 
-    "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"
-)
-client = OpenAI(
-    base_url="https://integrate.api.nvidia.com/v1",
-    api_key=NV_API_KEY
-)
-CHAT_MODEL = "meta/llama3-8b-instruct"
-
-# Dialog history stored as list of (user, assistant) tuples
-dialog_history: list[tuple[str, str]] = []
-
-def chatbot(user_input, temperature, top_p, max_tokens):
-    global dialog_history
-    if not user_input:
-        return dialog_history, ""
-
-    # Retrieve context and build system message
-    context = retrieve_context(user_input)
-    system_msg = {
-        "role": "system",
-        "content": (
-            "You are an assistant specialized in R packages. "
-            "Use only the context below to answer. "
-            "If you don't know, state that you don't know."
-            f"\n\n=== Retrieved Context ===\n{context}\n\n"
-        )
-    }
-
-    # Build message list for the API
-    messages = [system_msg]
-    for user_msg, assistant_msg in dialog_history:
-        messages.append({"role": "user", "content": user_msg})
-        messages.append({"role": "assistant", "content": assistant_msg})
-    messages.append({"role": "user", "content": user_input})
-
-    # Call NVIDIA streaming API
-    assistant_reply = ""
-    try:
-        stream = client.chat.completions.create(
-            model=CHAT_MODEL,
-            messages=messages,
-            temperature=temperature,
-            top_p=top_p,
-            max_tokens=max_tokens,
-            stream=True
-        )
-        for chunk in stream:
-            delta = chunk.choices[0].delta
-            if hasattr(delta, "content") and delta.content:
-                assistant_reply += delta.content
-    except OpenAIError as e:
-        assistant_reply = f"⚠️ API Error: {e.__class__.__name__}: {e}"
-
-    dialog_history.append((user_input, assistant_reply))
-    return dialog_history, ""
-
-
-def clear_history():
-    global dialog_history
-    dialog_history = []
-    return [], ""
-
-# Custom CSS for modern, responsive layout
-custom_css = r"""
-:root {
-    --primary: #4a90e2;
-    --secondary: #50e3c2;
-    --background-light: #f9f9f9;
-    --background-dark: #1e1e1e;
-    --text-light: #ffffff;
-    --text-dark: #333333;
-    --radius: 8px;
-    --spacing: 1rem;
-}
-
-body {
-    background-color: var(--background-light);
-    color: var(--text-dark);
-    font-family: 'Helvetica Neue', sans-serif;
-}
-
-#chat-window {
-    height: 60vh;
-    overflow-y: auto;
-    padding: var(--spacing);
-    border: 1px solid #dddddd;
-    border-radius: var(--radius);
-    background-color: #ffffff;
-}
-
-#input-area {
-    margin-top: var(--spacing);
-}
-
-#user-input {
-    flex: 1;
-    padding: 0.5rem;
-    border-radius: var(--radius) 0 0 var(--radius);
-    border: 1px solid #cccccc;
-}
-
-#send-button {
-    border-radius: 0 var(--radius) var(--radius) 0;
-}
-
-@media (prefers-color-scheme: dark) {
-    body {
-        background-color: var(--background-dark);
-        color: var(--text-light);
-    }
-    #chat-window {
-        background-color: #2a2a2a;
-        border-color: #444444;
-    }
-}
-"""
-
-# Gradio interface
-title = "Search Curriculum Vitae"
-with gr.Blocks(title=title, css=custom_css, theme=gr.themes.Base()) as demo:
-    gr.Markdown(f"## {title}")
-    with gr.Row():
-        # Main column: chat interface
-        with gr.Column(scale=3):
-            chatbot_ui = gr.Chatbot(label="Assistant")
-            with gr.Row(elem_id="input-area"):
-                txt = gr.Textbox(placeholder="Type your question...", lines=2, elem_id="user-input")
-                btn = gr.Button("Send", elem_id="send-button")
-                clr = gr.Button("Clear", elem_id="clear-button")
-        # Sidebar column: advanced controls
-        with gr.Column(scale=1):
-            with gr.Accordion("Advanced Settings", open=False):
-                temperature = gr.Slider(0, 1, value=0.6, label="Temperature")
-                top_p       = gr.Slider(0, 1, value=0.95, label="Top-p")
-                max_tokens  = gr.Slider(64, 2048, value=512, step=64, label="Max Tokens")
-                # Explanation for advanced settings
-                gr.Markdown(
-                    """
-                    **Temperature:** Adjusts the randomness of responses. Lower values make output more deterministic; higher values increase creativity.
-
-                    **Top-p (nucleus sampling):** Limits the next-token selection to the top p percentile of probability mass. Lower values make responses more focused.
-
-                    **Max Tokens:** Sets the maximum length of the assistant's reply.
-                    """
-                )
-
-    # Event bindings
-    btn.click(chatbot, [txt, temperature, top_p, max_tokens], [chatbot_ui, txt])
-    txt.submit(chatbot, [txt, temperature, top_p, max_tokens], [chatbot_ui, txt])
-    clr.click(clear_history, [], [chatbot_ui, txt])
-
-if __name__ == "__main__":
-    demo.launch(server_name="0.0.0.0", server_port=7861)
+import os
+import gradio as gr
+import numpy as np
+from sentence_transformers import SentenceTransformer
+import faiss
+from openai import OpenAI, OpenAIError
+from langchain_community.document_loaders import PyPDFLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+
+# Paths for files generated by build_index.py
+INDEX_FILE  = "r_docs.index"
+CHUNKS_FILE = "r_chunks.npy"
+
+# Check index existence
+if not os.path.exists(INDEX_FILE) or not os.path.exists(CHUNKS_FILE):
+    raise FileNotFoundError(
+        "Index not found. Please run first:\n  python build_index.py"
+    )
+
+# Load FAISS index and chunks
+index  = faiss.read_index(INDEX_FILE)
+chunks = np.load(CHUNKS_FILE, allow_pickle=True)
+
+# Embedding model for context retrieval
+embedding_model = SentenceTransformer('all-MiniLM-L6-v2')
+
+def retrieve_context(query: str, k: int = 4) -> str:
+    q_emb = embedding_model.encode([query], convert_to_numpy=True)
+    _, I = index.search(q_emb, k)
+    return "\n---\n".join(chunks[i] for i in I[0])
+
+# NVIDIA OpenAI-compatible client for chat
+NV_API_KEY = os.getenv(
+    "NV_API_KEY", 
+    "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"
+)
+client = OpenAI(
+    base_url="https://integrate.api.nvidia.com/v1",
+    api_key=NV_API_KEY
+)
+CHAT_MODEL = "meta/llama3-8b-instruct"
+
+# Dialog history stored as list of (user, assistant) tuples
+dialog_history: list[tuple[str, str]] = []
+
+def chatbot(user_input, temperature, top_p, max_tokens):
+    global dialog_history
+    if not user_input:
+        return dialog_history, ""
+
+    # Retrieve context and build system message
+    context = retrieve_context(user_input)
+    system_msg = {
+        "role": "system",
+        "content": (
+            "You are an assistant specialized in R packages. "
+            "Use only the context below to answer. "
+            "If you don't know, state that you don't know."
+            f"\n\n=== Retrieved Context ===\n{context}\n\n"
+        )
+    }
+
+    # Build message list for the API
+    messages = [system_msg]
+    for user_msg, assistant_msg in dialog_history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": assistant_msg})
+    messages.append({"role": "user", "content": user_input})
+
+    # Call NVIDIA streaming API
+    assistant_reply = ""
+    try:
+        stream = client.chat.completions.create(
+            model=CHAT_MODEL,
+            messages=messages,
+            temperature=temperature,
+            top_p=top_p,
+            max_tokens=max_tokens,
+            stream=True
+        )
+        for chunk in stream:
+            delta = chunk.choices[0].delta
+            if hasattr(delta, "content") and delta.content:
+                assistant_reply += delta.content
+    except OpenAIError as e:
+        assistant_reply = f"⚠️ API Error: {e.__class__.__name__}: {e}"
+
+    dialog_history.append((user_input, assistant_reply))
+    return dialog_history, ""
+
+
+def clear_history():
+    global dialog_history
+    dialog_history = []
+    return [], ""
+
+# Custom CSS for modern, responsive layout
+custom_css = r"""
+:root {
+    --primary: #4a90e2;
+    --secondary: #50e3c2;
+    --background-light: #f9f9f9;
+    --background-dark: #1e1e1e;
+    --text-light: #ffffff;
+    --text-dark: #333333;
+    --radius: 8px;
+    --spacing: 1rem;
+}
+
+body {
+    background-color: var(--background-light);
+    color: var(--text-dark);
+    font-family: 'Helvetica Neue', sans-serif;
+}
+
+#chat-window {
+    height: 60vh;
+    overflow-y: auto;
+    padding: var(--spacing);
+    border: 1px solid #dddddd;
+    border-radius: var(--radius);
+    background-color: #ffffff;
+}
+
+#input-area {
+    margin-top: var(--spacing);
+}
+
+#user-input {
+    flex: 1;
+    padding: 0.5rem;
+    border-radius: var(--radius) 0 0 var(--radius);
+    border: 1px solid #cccccc;
+}
+
+#send-button {
+    border-radius: 0 var(--radius) var(--radius) 0;
+}
+
+@media (prefers-color-scheme: dark) {
+    body {
+        background-color: var(--background-dark);
+        color: var(--text-light);
+    }
+    #chat-window {
+        background-color: #2a2a2a;
+        border-color: #444444;
+    }
+}
+"""
+
+# Gradio interface
+title = "Search Curriculum Vitae"
+with gr.Blocks(title=title, css=custom_css, theme=gr.themes.Base()) as demo:
+    gr.Markdown(f"## {title}")
+    with gr.Row():
+        # Main column: chat interface
+        with gr.Column(scale=3):
+            chatbot_ui = gr.Chatbot(label="Assistant")
+            with gr.Row(elem_id="input-area"):
+                txt = gr.Textbox(placeholder="Type your question...", lines=2, elem_id="user-input")
+                btn = gr.Button("Send", elem_id="send-button")
+                clr = gr.Button("Clear", elem_id="clear-button")
+        # Sidebar column: advanced controls
+        with gr.Column(scale=1):
+            with gr.Accordion("Advanced Settings", open=False):
+                temperature = gr.Slider(0, 1, value=0.6, label="Temperature")
+                top_p       = gr.Slider(0, 1, value=0.95, label="Top-p")
+                max_tokens  = gr.Slider(64, 2048, value=512, step=64, label="Max Tokens")
+                # Explanation for advanced settings
+                gr.Markdown(
+                    """
+                    **Temperature:** Adjusts the randomness of responses. Lower values make output more deterministic; higher values increase creativity.
+
+                    **Top-p (nucleus sampling):** Limits the next-token selection to the top p percentile of probability mass. Lower values make responses more focused.
+
+                    **Max Tokens:** Sets the maximum length of the assistant's reply.
+                    """
+                )
+
+    # Event bindings
+    btn.click(chatbot, [txt, temperature, top_p, max_tokens], [chatbot_ui, txt])
+    txt.submit(chatbot, [txt, temperature, top_p, max_tokens], [chatbot_ui, txt])
+    clr.click(clear_history, [], [chatbot_ui, txt])
+
+if __name__ == "__main__":
+    demo.launch(server_name="0.0.0.0", server_port=7860)