https://github.com/pablodieaco/quant-drl-web
quant-drl-web is a Streamlit-based interactive dashboard for visualizing and interacting with DRL-driven portfolio management models.
https://github.com/pablodieaco/quant-drl-web
deep-reinforcement-learning quantitative-finance quantitative-trading streamlit yfinance
Last synced: about 2 months ago
JSON representation
quant-drl-web is a Streamlit-based interactive dashboard for visualizing and interacting with DRL-driven portfolio management models.
- Host: GitHub
- URL: https://github.com/pablodieaco/quant-drl-web
- Owner: pablodieaco
- License: mit
- Created: 2025-04-03T16:59:57.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-04-16T11:40:26.000Z (about 1 year ago)
- Last Synced: 2025-05-25T06:14:00.129Z (about 1 year ago)
- Topics: deep-reinforcement-learning, quantitative-finance, quantitative-trading, streamlit, yfinance
- Language: Python
- Homepage:
- Size: 17.4 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# 📊 quant-drl-web
**quant-drl-web** is a Streamlit-based interactive dashboard for visualizing, evaluating, and simulating Deep Reinforcement Learning (DRL) models applied to portfolio management.
It serves as the frontend for the [`quant-drl-core`](https://github.com/pablodieaco/quant-drl-core) research framework, providing a user-friendly interface for financial data exploration, model comparison, and portfolio decision analysis.
> 🎓 Developed as part of my **Final Degree Project (TFG)** in the **Double Degree in Mathematics and Computer Engineering** at the **University of Seville**.
---
## ✨ Features
- 📈 **Model evaluation & metrics explorer**
- 💡 **Portfolio simulation and strategy testing**
- 🧠 **Deep RL integration** (PPO, SAC)
- 🔌 **Database-backed** via PostgreSQL
- 🎨 **Streamlit** multipage navigation
- 🐳 Docker & Docker Compose for deployment
- 📂 Examples and demo models included
---
## 🖼️ App Preview
### 📊 Financial Information
An overview of each company's economic and financial data.
---
### 📝 Evaluations
#### 🧪 Batch Evaluation
Evaluate multiple simulation at once, and analyse the performance of your model.
#### 🔍 Individual Evaluation
Dive deep into the analysis of a single simulation.
---
### 📈 Metrics Analysis
Explore key performance indicators of your trained models.
## 📁 Project Structure
```bash
quant-drl-web/
├── web/ # Streamlit app (modular)
│ ├── app.py # Main entry point
│ ├── pages/ # UI Pages (Streamlit multipage)
│ ├── commons/ # Session & layout
│ └── db/ # DB scripts and seeders
├── examples/ # Sample models and evaluation results
│ ├── models/ # Pretrained DRL models
│ └── results/ # evaluation_results.csv
├── docker-compose.yml # Launch web + PostgreSQL
├── Dockerfile # Web app container
├── build.sh # Script to copy core + build
└── requirements.txt # Python dependencies
```
---
## ⚙️ Requirements
- Docker & Docker Compose
- Python 3.10+ (optional for local runs)
- `quant-drl-core` available locally (used during build) (TODO: It will not be necessary in future releases)
---
## 🚀 Quickstart
### 1. Clone the repo
```bash
git clone https://github.com/pablodieaco/quant-drl-web.git
cd quant-drl-web
```
### 2. Create a `.env` file using the sample file `.env.sample`
```bash
cp .env.sample .env
```
Then edit the `.env` file:
```env
POSTGRES_USER=
POSTGRES_PASSWORD=
POSTGRES_DB=
RESULTS_CSV=examples/results/evaluation_results.csv
MODEL_HIERARCHY=examples/models/metadata/hierarchy.json
```
### 3. Run using the latest Docker image (recommended)
This will pull the latest published image from GitHub Container Registry and run the app.
```bash
make run
```
Once running, the app will be available at http://localhost:8501
### 4. Local development mode (with local quant-drl-core)
If you're modifying `quant-drl-web` and want to test changes locally:
```bash
make build
make run-local
```
This will
- Build the Docker image with your local changes
- Launch the app using the new built image
Once running, the app will also be available at http://localhost:8501
---
## 🧪 Example Models
We’ve included pretrained models and example evaluation results inside the `examples/` folder:
- `examples/models/` → PPO and SAC agents
- `examples/results/evaluation_results.csv`→ model evaluation scores
---
## 🧠 About this project
This dashboard is part of a broader research framework:
👉 Check out `quant-drl-core` for:
- DRL training pipelines (PPO, SAC)
- Custom Gym environments
- Evaluation and visualization logic
- Experiment reproducibility
## 📝 License
This project is licensed under the **MIT License**.
See the [LICENSE](./LICENSE) file for details.
---
## 🙋♂️ Author
Made with ❤️ by **Pablo Diego Acosta**
- 💼 LinkedIn: [linkedin.com/in/pablodiegoacosta](https://www.linkedin.com/in/pablodiegoacosta)
---
## ✨ Todo / Ideas
- [ ] Embed live charts from financial APIs
- [ ] Auto-refresh evaluation dashboard with scheduled updates
- [ ] Live inference from trained DRL models
- [ ] Add functionalities for managing saved portfolios