Upload 9 files

LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 Luong Huu Thanh
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

README.md

@@ -1,8 +1,8 @@
 ---
-title: Template Final Assignment
-emoji: 🕵🏻‍♂️
-colorFrom: indigo
-colorTo: indigo
+title: GAIA Lite
+emoji: 🤖
+colorFrom: green
+colorTo: blue
 sdk: gradio
 sdk_version: 5.25.2
 app_file: app.py
@@ -12,4 +12,210 @@ hf_oauth: true
 hf_oauth_expiration_minutes: 480
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# 🤖 **GAIA Lite**
+
+## 🌟 **Introduction**
+
+**GAIA Lite** is a small tool-calling agent for learning LangGraph and for the GAIA / Hugging Face Agents Course homework. It keeps a short tool list, a bounded think–act loop, regex answer extraction, and a cached evaluation runner.
+
+## 🚀 **Key Features**
+
+- **🔍 Multi-Modal Search**: Web search, Wikipedia, and arXiv paper search
+- **💻 Code Execution**: Support for Python, Bash, SQL, C, and Java
+- **🖼️ Image Processing**: Analysis, transformation, OCR, and generation
+- **📄 Document Processing**: PDF, CSV, Excel, and text file analysis
+- **📁 File Upload Support**: Handle multiple file types with drag-and-drop
+- **🧮 Mathematical Operations**: Complete set of mathematical tools
+- **💬 Conversational Interface**: Natural chat-based interaction
+- **📊 Evaluation System**: Automated benchmark testing and submission
+
+## 🏗️ **Project Structure**
+
+```
+gaia-lite/
+├── app.py                    # Gradio Q&A chatbot
+├── evaluation_app.py         # GAIA eval + cached submit
+├── agent.py                  # Tools + LangGraph loop
+├── extract_answer.py         # Parse/normalize FINAL ANSWER
+├── files_util.py             # Attachment download + prompt paths
+├── code_interpreter.py       # Python (and other) code execution
+├── image_processing.py       # Optional image helpers (not wired)
+├── system_prompt.txt         # Agent instructions
+├── answers_cache.json        # Local eval cache (gitignored)
+├── requirements.txt
+└── README.md
+```
+
+## 🛠️ **Tools (kept small on purpose)**
+
+The agent binds six tools. Extra math/image-generation helpers were removed from the loop so the model spends less time picking the wrong tool.
+
+- `web_search` — Tavily, up to 3 hits
+- `wiki_search` — Wikipedia, up to 2 pages
+- `execute_python` — calculation, pandas, dates, CSV/Excel via code
+- `download_file_from_url` — save a remote file locally
+- `read_file` — preview text, CSV, Excel, PDF
+- `extract_text_from_image` — Tesseract OCR
+
+Vector-store retrieval is **off by default**. Pass `build_graph(use_retriever=True)` only if you explicitly want similar-question context.
+
+## 🎯 **How to Use**
+
+### **Q&A Chatbot Interface (app.py)**
+
+1. **Start the Chatbot:**
+   ```bash
+   python app.py
+   ```
+
+2. **Access the Interface:**
+   - Open `http://localhost:7860` in your browser
+   - Upload files (images, documents, CSV, etc.) if needed
+   - Ask questions in natural language
+   - Get comprehensive answers with tool usage
+
+3. **Supported Interactions:**
+   - **Text Questions**: "What is the capital of France?"
+   - **Math Problems**: "Calculate the square root of 144"
+   - **Code Requests**: "Write a Python function to sort a list"
+   - **Image Analysis**: Upload an image and ask "What do you see?"
+   - **Data Analysis**: Upload a CSV and ask "What are the trends?"
+   - **Web Search**: "What are the latest AI developments?"
+
+### **Evaluation Runner (evaluation_app.py)**
+
+1. **Run the Evaluation:**
+   ```bash
+   python evaluation_app.py
+   ```
+
+2. **Benchmark Testing:**
+   - Log in with your Hugging Face account
+   - Click "Run Evaluation & Submit All Answers"
+   - Monitor progress as the agent processes GAIA benchmark questions
+   - View results and scores automatically
+
+## 🔧 **Technical Architecture**
+
+### **LangGraph State Machine**
+```
+START → maybe_example → assistant ⇄ tools
+                          ↑           ↓
+                          └───────────┘
+```
+
+1. **maybe_example**: no-op unless `use_retriever=True`
+2. **assistant**: Groq Qwen with tools; after 8 tool rounds or a repeated identical call, it is forced to emit `FINAL ANSWER`
+3. **tools**: `ToolNode` runs the selected Python functions
+4. **extract_final_answer**: regex parse + light normalization before scoring submit
+
+Evaluation writes `answers_cache.json`, skips successful tasks on rerun, and submits from cache in a separate button. Attachments use `GET /files/{task_id}`, with an optional Hugging Face GAIA dataset fallback.
+
+### **Vector Database Integration**
+- **Supabase Vector Store**: Stores GAIA benchmark Q&A pairs
+- **Semantic Search**: Finds similar questions for context
+- **HuggingFace Embeddings**: sentence-transformers/all-mpnet-base-v2
+
+### **Multi-Modal File Support**
+- **Images**: JPG, PNG, GIF, BMP, WebP
+- **Documents**: PDF, DOC, DOCX, TXT, MD
+- **Data**: CSV, Excel, JSON
+- **Code**: Python, Bash, SQL, C, Java
+
+## ⚙️ **Installation & Setup**
+
+### **1. Clone Repository**
+```bash
+git clone https://github.com/fisherman611/gaia-agent.git gaia-lite
+cd gaia-lite
+```
+
+### **2. Install Dependencies**
+```bash
+pip install -r requirements.txt
+```
+
+### **3. Environment Variables**
+Create a `.env` file with your API keys:
+```env
+SUPABASE_URL=your_supabase_url
+SUPABASE_SERVICE_ROLE_KEY=your_supabase_key
+GROQ_API_KEY=your_groq_api_key
+TAVILY_API_KEY=your_tavily_api_key
+HUGGINGFACEHUB_API_TOKEN=your_hf_token
+LANGSMITH_API_KEY=your_langsmith_key
+
+LANGSMITH_TRACING=true
+LANGSMITH_PROJECT=gaia-lite
+LANGSMITH_ENDPOINT=https://api.smith.langchain.com
+```
+
+### **4. Database Setup (Supabase)**
+Execute this SQL in your Supabase database:
+```sql
+-- Enable pgvector extension
+CREATE EXTENSION IF NOT EXISTS vector;
+
+-- Create match function for documents2 table
+CREATE OR REPLACE FUNCTION public.match_documents_2(
+  query_embedding vector(768)
+)
+RETURNS TABLE(
+  id         bigint,
+  content    text,
+  metadata   jsonb,
+  embedding  vector(768),
+  similarity double precision
+)
+LANGUAGE sql STABLE
+AS $$
+  SELECT
+    id,
+    content,
+    metadata,
+    embedding,
+    1 - (embedding <=> query_embedding) AS similarity
+  FROM public.documents2
+  ORDER BY embedding <=> query_embedding
+  LIMIT 10;
+$$;
+
+-- Grant permissions
+GRANT EXECUTE ON FUNCTION public.match_documents_2(vector) TO anon, authenticated;
+```
+
+## 🚀 **Running the Application**
+
+### **Chatbot Interface**
+```bash
+python app.py
+```
+Access at: `http://localhost:7860`
+
+### **Evaluation Runner**
+```bash
+python evaluation_app.py
+```
+Access at: `http://localhost:7860`
+
+### **Live Demo**
+Upstream template Space: [fisherman611/gaia-agent](https://huggingface.co/spaces/fisherman611/gaia-agent)
+
+## 🔗 **Resources**
+
+- [GAIA Benchmark](https://huggingface.co/spaces/gaia-benchmark/leaderboard)
+- [Hugging Face Agents Course](https://huggingface.co/agents-course)
+- [LangGraph Documentation](https://langchain-ai.github.io/langgraph/)
+- [Supabase Vector Store](https://supabase.com/docs/guides/ai/vector-columns)
+
+## 🤝 **Contributing**
+
+Contributions are welcome! Areas for improvement:
+- **New Tools**: Add specialized tools for specific domains
+- **UI Enhancements**: Improve the chatbot interface
+- **Performance**: Optimize response times and accuracy
+- **Documentation**: Expand examples and use cases
+
+## 📄 **License**
+
+This project is licensed under the [MIT License](https://mit-license.org/).

extract_answer.py

@@ -0,0 +1,41 @@
+import re
+from typing import Optional
+
+_FINAL_ANSWER_RE = re.compile(
+    r"FINAL\s*ANSWER\s*:\s*(.+)$",
+    re.IGNORECASE | re.DOTALL,
+)
+
+
+def extract_final_answer(text: Optional[str]) -> str:
+    """Pull the GAIA-style final answer out of a free-form model reply."""
+    if not text:
+        return ""
+
+    text = str(text).strip()
+    match = _FINAL_ANSWER_RE.search(text)
+    raw = match.group(1).strip() if match else text
+
+    if "FINAL ANSWER:" in raw.upper():
+        raw = re.split(r"FINAL\s*ANSWER\s*:", raw, flags=re.IGNORECASE)[-1].strip()
+
+    first_line = raw.splitlines()[0].strip()
+    first_line = first_line.strip(" `\"'")
+    first_line = re.sub(r"^\*\*?|\*\*?$", "", first_line).strip()
+    return normalize_gaia_answer(first_line)
+
+
+def normalize_gaia_answer(answer: str) -> str:
+    """Light cleanup that matches common GAIA scoring pitfalls."""
+    answer = answer.strip()
+    answer = answer.strip(" `\"'")
+    if answer.endswith("."):
+        answer = answer[:-1].strip()
+
+    compact = answer.replace(" ", "")
+    if re.fullmatch(r"[+-]?\d{1,3}(,\d{3})+(\.\d+)?", answer):
+        return answer.replace(",", "")
+    if re.fullmatch(r"[+-]?\d+(\.\d+)?%?", compact) and "," in answer:
+        return answer.replace(",", "")
+
+    return answer

files_util.py

@@ -0,0 +1,82 @@
+import os
+import shutil
+from typing import List, Optional
+
+import requests
+
+DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
+
+
+def format_user_message(question: str, file_paths: Optional[List[str]] = None) -> str:
+    question = (question or "").strip()
+    if not file_paths:
+        return question
+
+    lines = [question, "", "Attached local files (use tools with these paths):"]
+    for path in file_paths:
+        if not path:
+            continue
+        name = os.path.basename(path)
+        exists = os.path.exists(path)
+        size = os.path.getsize(path) if exists else 0
+        lines.append(f"- {name}: {path} (exists={exists}, bytes={size})")
+    return "\n".join(lines)
+
+
+def download_scoring_file(
+    task_id: str,
+    file_name: Optional[str] = None,
+    dest_dir: str = ".",
+    api_url: str = DEFAULT_API_URL,
+    timeout: int = 60,
+) -> Optional[str]:
+    """Download a GAIA attachment from the course scoring API."""
+    if not file_name:
+        return None
+
+    os.makedirs(dest_dir, exist_ok=True)
+    dest_path = os.path.join(dest_dir, os.path.basename(file_name))
+    url = f"{api_url.rstrip('/')}/files/{task_id}"
+
+    try:
+        response = requests.get(url, timeout=timeout)
+        if response.status_code != 200:
+            print(f"Scoring API file {task_id} -> HTTP {response.status_code}: {response.text[:200]}")
+            return download_gaia_dataset_file(file_name, dest_path)
+        with open(dest_path, "wb") as handle:
+            handle.write(response.content)
+        print(f"Saved attachment {file_name} -> {dest_path}")
+        return dest_path
+    except requests.RequestException as exc:
+        print(f"Scoring API file download failed for {task_id}: {exc}")
+        return download_gaia_dataset_file(file_name, dest_path)
+
+
+def download_gaia_dataset_file(file_name: str, dest_path: str) -> Optional[str]:
+    """Fallback when /files/{task_id} is 404: try the gated GAIA dataset on the Hub."""
+    try:
+        from huggingface_hub import hf_hub_download
+    except ImportError:
+        return None
+
+    token = os.getenv("HUGGINGFACEHUB_API_TOKEN") or os.getenv("HF_TOKEN")
+    candidates = [
+        f"2023/validation/{file_name}",
+        f"2023/test/{file_name}",
+        file_name,
+    ]
+    for filename in candidates:
+        try:
+            cached = hf_hub_download(
+                repo_id="gaia-benchmark/GAIA",
+                filename=filename,
+                repo_type="dataset",
+                token=token,
+            )
+            os.makedirs(os.path.dirname(dest_path) or ".", exist_ok=True)
+            shutil.copy2(cached, dest_path)
+            print(f"Copied GAIA dataset file {filename} -> {dest_path}")
+            return dest_path
+        except Exception as exc:
+            print(f"Hub download missed {filename}: {exc}")
+    return None

image_processing.py

@@ -0,0 +1,26 @@
+import os
+import io
+import base64
+import uuid
+from PIL import Image
+
+# Helper functions for image processing
+def encode_image(image_path: str) -> str:
+    """Convert an image file to base64 string."""
+    with open(image_path, "rb") as image_file:
+        return base64.b64encode(image_file.read()).decode("utf-8")
+
+
+def decode_image(base64_string: str) -> Image.Image:
+    """Convert a base64 string to a PIL Image."""
+    image_data = base64.b64decode(base64_string)
+    return Image.open(io.BytesIO(image_data))
+
+
+def save_image(image: Image.Image, directory: str = "image_outputs") -> str:
+    """Save a PIL Image to disk and return the path."""
+    os.makedirs(directory, exist_ok=True)
+    image_id = str(uuid.uuid4())
+    image_path = os.path.join(directory, f"{image_id}.png")
+    image.save(image_path)
+    return image_path

requirements.txt

@@ -1,2 +1,21 @@
 gradio
-requests
+requests
+langchain
+langchain-community
+langchain-core
+langchain-google-genai
+langchain-huggingface
+langchain-groq
+langchain-tavily
+langchain-chroma
+langgraph
+huggingface_hub
+supabase
+arxiv
+pymupdf
+wikipedia
+pgvector
+python-dotenv
+pytesseract
+matplotlib
+sentence_transformers

system_prompt.txt

@@ -0,0 +1,17 @@
+You are GAIA Lite, a tool-using assistant. Solve questions with tools, then give a scorable short answer.
+
+How to work:
+- Use tools when you need facts, files, calculation, or code. Prefer Python for math, tables, dates, and string processing.
+- If a file path is given, read or analyze that local path. Do not invent file contents.
+- If a search fails, change the query once. Do not call the same tool with the same arguments twice.
+- After you have enough evidence, stop using tools.
+
+Final message format:
+- You may reason in earlier messages.
+- Your LAST message must be exactly one line:
+FINAL ANSWER: <answer>
+- The answer must be a number, a few words, or a comma-separated list.
+- Numbers: no thousand separators, no units (no $, %, °) unless the question asks for them.
+- Strings: no leading articles, no abbreviations for cities/countries, digits in plain text unless the question wants numerals.
+- Lists: exactly one space after each comma.
+- If unsure, still output the best short guess in this format. Never refuse with a long explanation.