Add AGA-WebUI

.gitignore

@@ -5,6 +5,7 @@
 .aim/
 AutogluonModels/
 data/
+user_data/
 results/
 cookies.json
 cdk.out/

README.md

@@ -93,3 +93,29 @@ aga ./data ./config --output-filename my_output.csv --config-overrides "autogluo
 ```
 
 `autogluon-assistant-tools` provides more functionality and utilities for benchmarking, wrapped around autogluon-assistant. Please check out the [repo](https://github.com/autogluon/autogluon-assistant-tools/) for more details.
+
+## Autogluon Assistant Web UI
+The Autogluon Assistant Web UI is a user-friendly application that allows users to leverage the capabilities of the Autogluon-Assistant library through an intuitive web interface.
+
+The web UI enables users to upload datasets, configure Autogluon-Assistant runs with customized settings, preview data, monitor execution progress, view and download results, and supports secure, isolated sessions for concurrent users.
+
+#### To run the Autogluon Assistant Web UI:
+Navigate to the project directory and run the app:
+````
+cd src/autogluon_assistant/WebUI && streamlit run app.py
+````
+The Autogluon Assistant Web UI should now be accessible in your web browser at `http://localhost:8501`
+
+#### Add GPT4 Model to the LLM Option:
+If you’d like to add additional GPT4 model to the language model (LLM) dropdown:
+
+1. Navigate to src/autogluon_assistant/WebUI/constants.py
+
+2. Locate the `LLM_OPTIONS` variable, which looks like this:
+````
+LLM_OPTIONS = ["Claude 3.5 with bedrock"]
+````
+3. Add "GPT 4" to the list
+````
+LLM_OPTIONS = ["Claude 3.5 with bedrock", "GPT 4"]
+````

pyproject.toml

@@ -28,7 +28,12 @@ dependencies = [
     "sentence-transformers>=3.1.0",
     "tenacity>=8.2.2,<10.0",
     "gensim>=4.3",
-    "pandas>=2.2"
+    "pandas>=2.2",
+    "stqdm==0.0.5",
+    "streamlit==1.37.1",
+    "streamlit-aggrid==1.0.5",
+    "streamlit-extras==0.4.7",
+    "psutil>=5.9.8",
 ]
 
 [project.optional-dependencies]

src/autogluon_assistant/WebUI/.dockerignore

@@ -0,0 +1,46 @@
+# git
+**/.git
+
+# Byte-compiled / optimized / DLL files
+**/__pycache__/
+**/*.py[cod]
+
+# C extensions
+*.so
+
+# Virtual environment
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+*env/
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+*.egg-info/
+.installed.cfg
+*.egg
+
+user_data/
+local/
+results/
+AutogluonModels/
+.metaflow/
+
+.mypy_cache
+.pytest_cache
+.ruff_cache
+.logs

src/autogluon_assistant/WebUI/.streamlit/config.toml

@@ -0,0 +1,10 @@
+[server]
+maxUploadSize = 4096
+enableStaticServing = true
+[theme]
+primaryColor="#4C7DE7"
+backgroundColor="#FFFFFF"
+secondaryBackgroundColor="#fbfcfc"
+textColor="#404040"
+[client]
+showSidebarNavigation = false

src/autogluon_assistant/WebUI/Dockerfile

@@ -0,0 +1,21 @@
+FROM python:3.10-slim
+
+WORKDIR /app
+
+
+RUN pip install -U pip \
+    && pip install -U setuptools wheel
+
+RUN  apt-get -y update && \
+    apt-get install -y git
+
+COPY . /app
+
+RUN pip install .
+
+
+EXPOSE 8501
+
+WORKDIR /app/src
+
+CMD ["streamlit", "run", "app.py"]

src/autogluon_assistant/WebUI/app.py

@@ -0,0 +1,82 @@
+from copy import deepcopy
+
+import streamlit as st
+import streamlit.components.v1 as components
+from constants import DEFAULT_SESSION_VALUES
+from pages.demo import main as demo
+from pages.feature import main as feature
+from pages.nav_bar import nav_bar
+from pages.preview import main as preview
+from pages.task import main as run
+from pages.tutorial import main as tutorial
+
+st.set_page_config(
+    page_title="AutoGluon Assistant",
+    page_icon="https://pbs.twimg.com/profile_images/1373809646046040067/wTG6A_Ct_400x400.png",
+    layout="wide",
+    initial_sidebar_state="collapsed",
+)
+
+# fontawesome
+st.markdown(
+    """
+    <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.0/css/all.min.css">
+    """,
+    unsafe_allow_html=True,
+)
+
+# Bootstrap 4.1.3
+st.markdown(
+    """
+    <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/[email protected]/dist/css/bootstrap.min.css" integrity="sha384-MCw98/SFnGE8fJT3GXwEOngsV7Zt27NXFoaoApmYm81iuXoPkFOJwJ8ERdknLPMO" crossorigin="anonymous">
+    """,
+    unsafe_allow_html=True,
+)
+with open("style.css") as f:
+    st.markdown(f"<style>{f.read()}</style>", unsafe_allow_html=True)
+
+
+reload_warning = """
+<script>
+  window.onbeforeunload = function () {
+
+    return  "Are you sure want to LOGOUT the session ?";
+}; 
+</script>
+"""
+
+components.html(reload_warning, height=0)
+
+
+def initial_session_state():
+    """
+    Initial Session State
+    """
+    for key, default_value in DEFAULT_SESSION_VALUES.items():
+        if key not in st.session_state:
+            st.session_state[key] = (
+                deepcopy(default_value) if isinstance(default_value, (dict, list)) else default_value
+            )
+
+
+def main():
+    initial_session_state()
+    nav_bar()
+    tutorial()
+    demo()
+    feature()
+    run()
+    preview()
+
+    st.markdown(
+        """
+    <script src="https://code.jquery.com/jquery-3.3.1.slim.min.js" integrity="sha384-q8i/X+965DzO0rT7abK41JStQIAqVgRVzpbzo5smXKp4YfRvH+8abtTE1Pi6jizo" crossorigin="anonymous"></script>
+    <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/umd/popper.min.js" integrity="sha384-ZMP7rVo3mIykV+2+9J3UJ46jBk0WLaUAdn689aCwoqbBJiSnjAK/l8WvCWPIPm49" crossorigin="anonymous"></script>
+    <script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/js/bootstrap.min.js" integrity="sha384-ChfqqxuZUCnJSK3+MXmPNIyE6ZbWh2IMqE241rYiqJxyMiZ6OW/JmZQ5stwEULTy" crossorigin="anonymous"></script>
+    """,
+        unsafe_allow_html=True,
+    )
+
+
+if __name__ == "__main__":
+    main()

src/autogluon_assistant/WebUI/constants.py

@@ -0,0 +1,95 @@
+from copy import deepcopy
+
+BASE_DATA_DIR = "./user_data"
+
+
+# Preset configurations
+PRESET_DEFAULT_CONFIG = {
+    "Best Quality": {"time_limit": "4 hrs", "feature_generation": True},
+    "High Quality": {"time_limit": "1 hr", "feature_generation": False},
+    "Medium Quality": {"time_limit": "10 mins", "feature_generation": False},
+}
+DEFAULT_PRESET = "Best Quality"
+
+PRESET_MAPPING = {
+    "Best Quality": "best_quality",
+    "High Quality": "high_quality",
+    "Medium Quality": "medium_quality",
+}
+PRESET_OPTIONS = ["Best Quality", "High Quality", "Medium Quality"]
+
+# Time limit configurations (in seconds)
+TIME_LIMIT_MAPPING = {
+    "1 min": 60,
+    "10 mins": 600,
+    "30 mins": 1800,
+    "1 hr": 3600,
+    "2 hrs": 7200,
+    "4 hrs": 14400,
+}
+
+DEFAULT_TIME_LIMIT = "4 hrs"
+
+TIME_LIMIT_OPTIONS = ["1 min", "10 mins", "30 mins", "1 hr", "2 hrs", "4 hrs"]
+
+# LLM configurations
+LLM_MAPPING = {
+    "Claude 3.5 with bedrock": "anthropic.claude-3-5-sonnet-20241022-v2:0",
+    "GPT 4": "gpt-4o-mini-2024-07-18",
+}
+
+LLM_OPTIONS = ["Claude 3.5 with bedrock"]
+
+# Provider configuration
+PROVIDER_MAPPING = {"Claude 3.5 with bedrock": "bedrock", "GPT 4": "openai"}
+
+
+API_KEY_LOCATION = {"Claude 3.5 with bedrock": "BEDROCK_API_KEY", "GPT 4": "OPENAI_API_KEY"}
+
+INITIAL_STAGE = {
+    "Task Understanding": [],
+    "Feature Generation": [],
+    "Model Training": [],
+    "Prediction": [],
+}
+# Initial Session state
+DEFAULT_SESSION_VALUES = {
+    "config_overrides": [],
+    "preset": DEFAULT_PRESET,
+    "time_limit": DEFAULT_TIME_LIMIT,
+    "llm": None,
+    "pid": None,
+    "logs": "",
+    "process": None,
+    "clicked": False,
+    "task_running": False,
+    "output_file": None,
+    "output_filename": None,
+    "task_description": "",
+    "sample_description": "",
+    "return_code": None,
+    "task_canceled": False,
+    "uploaded_files": {},
+    "sample_files": {},
+    "selected_dataset": None,
+    "sample_dataset_dir": None,
+    "description_uploader_key": 0,
+    "sample_dataset_selector": None,
+    "current_stage": None,
+    "feature_generation": True,
+    "stage_status": {},
+    "show_remaining_time": False,
+    "model_path": None,
+    "increment_time": 0,
+    "progress_bar": None,
+    "increment": 2,
+    "zip_path": None,
+    "stage_container": deepcopy(INITIAL_STAGE),
+}
+
+
+# DataSet Options
+DATASET_OPTIONS = ["Sample Dataset", "Upload Dataset"]
+
+# Captions under DataSet Options
+CAPTIONS = ["Run with sample dataset", "Upload Train, Test and Output (Optional) Dataset"]

src/autogluon_assistant/WebUI/file_uploader.py

@@ -0,0 +1,57 @@
+import os
+
+import pandas as pd
+import streamlit as st
+from utils import get_user_data_dir
+
+
+def save_description_file(description):
+    """
+    Save the task description to a file in the user's data directory.
+
+    Args:
+        description (str): The task description to save.
+    """
+    try:
+        user_data_dir = get_user_data_dir()
+        description_file = os.path.join(user_data_dir, "description.txt")
+        with open(description_file, "w") as f:
+            f.write(description)
+    except Exception as e:
+        print(f"Error saving file: {str(e)}")
+
+
+def description_file_uploader():
+    """
+    Handle Description file uploads
+    """
+    uploaded_file = st.file_uploader(
+        "Upload task description file",
+        type="txt",
+        key=st.session_state.description_uploader_key,
+        help="Accepted file format: .txt",
+        label_visibility="collapsed",
+    )
+    if uploaded_file:
+        task_description = uploaded_file.read().decode("utf-8")
+        st.session_state.task_description = task_description
+        save_description_file(st.session_state.task_description)
+        st.session_state.description_uploader_key += 1
+        st.rerun()
+
+
+def file_uploader():
+    """
+    Handle file uploads
+    """
+    st.markdown("#### Upload Dataset")
+    uploaded_files = st.file_uploader(
+        "Select the dataset", accept_multiple_files=True, label_visibility="collapsed", type=["csv", "xlsx"]
+    )
+    st.session_state.uploaded_files = {}
+    for file in uploaded_files:
+        if file.name.endswith(".csv"):
+            df = pd.read_csv(file)
+        elif file.name.endswith(".xlsx"):
+            df = pd.read_excel(file)
+        st.session_state.uploaded_files[file.name] = {"file": file, "df": df}