Enhancements to ObjectDetector and Model Training

[devin-ai-integration]

Enhanced ObjectDetector and Model Training

This pull request introduces improvements to the ObjectDetector and model training components of the PikasuBirdAi project.

Changes

• Enhanced ObjectDetector to support batch processing of images
• Added error handling to ObjectDetector
• Improved model_training with adaptive learning rates using exponential decay
• Implemented an early stopping mechanism in model_training
• Added placeholder for future implementation of model compression techniques

Notes

• These changes were implemented without running or testing the code.
• Further testing and validation may be required before merging.

This PR was written by Devin 👼

This Devin run was requested by kasinadhsarma.