I'm AISpaceXDragon, a passionate machine learning enthusiast and developer exploring the frontiers of deep learning, computer vision, and natural language processing. My journey includes building cutting-edge models, tackling challenging datasets, and contributing to impactful projects in domains like healthcare, e-commerce, OCR, and more.
- 🌟 Interests: Deep learning, computer vision, NLP, model fine-tuning, and exploring bias in AI systems.
- 🎓 Learning Goals: Continuously advancing my knowledge in AI, contributing to research, and solving real-world problems with machine learning.
- 🌍 Languages and Tools: Python, PyTorch, TensorFlow, OpenCV, Streamlit, Gradio, and more.
- Optical Character Recognition (OCR): Developing robust multilingual OCR models with advanced caching and deployment strategies.
- Bias in LLMs: Documenting and analyzing bias in large language models using research-backed methodologies.
- Scratch Detection in Images: Training models to classify and detect scratches on text in images with bounding boxes and masks.
- Delivery Prediction: Tackling delayed delivery predictions in shipment logistics using binary classification.
- 💡 Built a multitask model with an innovative loss function for OCR-based tasks.
- 📚 Fine-tuned models like GPT XTTS v2.0 and SpeechT5 for specific use cases.
- ⚙️ Optimized pipelines for large-scale dataset processing and streamlined workflows for audio and image data.
- 🎨 Generated synthetic datasets with real-life variations to improve model robustness.
- Expanding my expertise in model interpretability and fairness.
- Exploring generative AI models for creative and practical applications.
- Collaborating with the community to solve impactful problems in AI.
- 🔗 GitHub Profile
- 📬 Feel free to reach out for collaborations or discussions!
✨ Thanks for visiting! Together, let's push the boundaries of AI and create a smarter future!