MAY 🤸♂️ SAKIT 🤸♂️ BA 🤸♂️ BESHY 🤸♂️ KO? 🤸♂️
I'm thrilled to announce that I've just made a powerful (wow powerful, what an adjective HAHAHA) deep learning model for crop species disease classification! 🚀🌿
Using Microsoft's ResNet50 architecture and leveraging the incredible ImageNet weights, I've incorporated transfer learning to enhance the model's performance. Now, it's capable of detecting and classifying crop diseases into an impressive 38 distinct categories, encompassing various crop species and their respective health statuses. 🍃📷
My journey began by collecting a massive dataset of 54,306 images, featuring 14 different crops (I got the dataset from PlantVillage's Github Repo but it's now unavailable ><). With careful attention to detail, I split this dataset into an 80% training set and a 20% validation set to ensure robust training and evaluation. 💻📊
Harnessing the power of a Google Colaboratory GPU, I trained the ResNet50 model via transfer learning, capitalizing on the pre-trained weights to expedite the training process. 🎛️⚙️
And guess what? The validation accuracy I achieved is an incredible 98.47%! Akalain niyo yun, 98.47% WTF 🎉✅
With this groundbreaking model (I cannot with these adjectives), I can now accurately differentiate between 38 diverse classes, comprising various crop species and their health statuses. 🌾🌱
If you're interested in exploring the potential of deep learning for plant disease detection and classification, reach out to me! I'm more than happy to share my source code. 🌍🌿💪
BTW, this took me a LOT of time, blood, sweat, and tears to make (even if I benchmarked some of it lol). Hindi ko lang siya ma-deploy kasi may problem net namin but it will be up soon!
#deeplearning #PlantDiseaseDetection #transferlearning #aiforagriculture