🌴Automating agricultural grading using PyTorch. An end-to-end Feedforward Neural Network (FNN) classifying 7 date fruit varieties with 95.56% test accuracy using image-extracted morphological, color, and wavelet texture features. Implements validation-loss checkpointing.
python data-science machine-learning deep-learning agriculture cuda tabular-data pytorch artificial-intelligence feature-extraction neural-networks classification feedforward-neural-network data-preprocessing multiclass-classification classification-model fnn matplotlb image-features model-checkpointing
-
Updated
Jul 16, 2026 - Jupyter Notebook