YOLOv8s Stock Market Future Trends Prediction Model

Introduction

The YOLOv8s Stock Market future trends prediction model is an object detection framework designed to identify and classify stock market chart patterns in real-time trading video data. It aids traders and investors by providing timely insights for informed decision-making, leveraging the YOLO (You Only Look Once) framework.

Architecture

The model employs the YOLOv8s architecture, which includes a specialized backbone network, self-attention mechanisms, and features for trends-specific pattern detection. This architecture is optimized for real-time detection of stock market patterns, such as 'Down' and 'Up', in live trading environments.

Training

Training Data

The model is trained on a diverse dataset comprising stock market chart images that reflect various market conditions and scenarios.

Training Procedure

Training is conducted over multiple epochs with extensive computation, aiming to minimize detection loss and enhance performance in pattern detection. The model achieves a mean Average Precision (mAP@0.5) of 0.65, with an overall pattern accuracy of 0.90.

Compute Infrastructure

• Hardware: NVIDIA GeForce RTX 3080
• Software: Trained in a Jupyter Notebook environment

Guide: Running Locally

1. Install Required Packages:

   pip install ultralyticsplus==0.0.28 ultralytics==8.0.43
   

2. Load and Run the Model:

   from ultralyticsplus import YOLO, render_result
   import cv2
   
   # load model
   model = YOLO('foduucom/stockmarket-future-prediction')
   
   # set model parameters
   model.overrides['conf'] = 0.25
   model.overrides['iou'] = 0.45
   model.overrides['agnostic_nms'] = False
   model.overrides['max_det'] = 1000
   
   # set image
   image = '/path/to/your/document/images'
   
   # perform inference
   results = model.predict(image)
   
   # observe results
   print(results[0].boxes)
   render = render_result(model=model, image=image, result=results[0])
   render.show()
   

3. Cloud GPUs: Consider using cloud services like AWS EC2, Google Cloud, or Azure for GPU resources to enhance performance and speed.

License

For inquiries and contributions, contact info@foduu.com. The model card authors are Nehul Agrawal and Rahul Parihar, 2023.