autogluon

Create a fully-functional mini-application using the 'autogluon' Python package that demonstrates its capability to quickly build machine learning models with high accuracy. Your task is to develop a simple yet effective tool that predicts housing prices based on various attributes like square footage, number of bedrooms, bathrooms, etc. The application should follow these steps:

1. Data Collection: Gather a dataset from a reliable source such as Kaggle, UCI Machine Learning Repository, or another trusted provider that includes housing price data along with relevant attributes.
2. Preprocessing: Clean the dataset to handle missing values, outliers, and ensure all data types are correctly formatted for analysis.
3. Model Training: Utilize Autogluon's capabilities to automatically train multiple machine learning models on the preprocessed data. Focus on leveraging Autogluon's ability to tune hyperparameters and select the best model without requiring extensive manual tuning.
4. Evaluation: Evaluate the performance of the trained models using appropriate metrics such as RMSE (Root Mean Square Error) or MAE (Mean Absolute Error). Provide visualizations if possible to compare different models' performances.
5. Prediction: Implement a feature where users can input new housing attributes, and the app will predict the house price using the best performing model identified during training.

Suggested Features:
- A user-friendly interface for data input and results display.
- Option to save and load trained models for future use.
- Detailed logs or reports about the training process and model evaluation.
- Visualization of feature importance to help understand which attributes significantly influence house prices.

Utilize Autogluon's core functionalities to streamline the entire process from data ingestion to prediction, showcasing its efficiency and ease-of-use in practical applications.