cnn-vit-ensemble

A streamlined deep learning framework that fuses advanced image preprocessing, grid search-based hyperparameter tuning, CNN/Transformer fine-tuning, and ensemble strategies for high-precision herb classification.

A streamlined deep learning framework that fuses advanced image preprocessing, grid search-based hyperparameter tuning, CNN/Transformer fine-tuning, and ensemble strategies for high-precision soil classification along with adaptive fuzzy logic-based crop recommendations.

EMORA is a multimodal AI conversational agent that integrates text, audio, and visual signals to understand emotion-related patterns. It combines MFN-based emotion recognition, conversational memory, and adaptive response strategies with Llama 3-MAGneT. Built to explore context-aware, supportive conversations for behaviour-change support.

A reproducible cross‑framework study comparing CNN and CNN‑ViT hybrid architectures for EuroSAT satellite crop classification using aligned Keras and PyTorch implementations.

Binary classification of satellite imagery into agricultural and non-agricultural land using custom CNNs and a CNN-ViT hybrid pipeline, implemented in both Keras and PyTorch.