malaria-detection-mobilenetv2

Malaria Detection - MobileNetV2 Transfer Learning

Model Description

This model uses MobileNetV2 (pre-trained on ImageNet) with transfer learning to detect malaria parasites in cell images. The model classifies blood cell images as either Parasitized or Uninfected.

Model Architecture

• Base Model: MobileNetV2 (pre-trained on ImageNet, frozen)
• Global Average Pooling layer
• Dropout layer (0.2)
• Dense layer with sigmoid activation for binary classification

Training Details

• Dataset: Cell Images for Detecting Malaria
• Input Size: 150x150x3
• Optimizer: Adam (learning_rate=0.0001)
• Loss Function: Binary Crossentropy
• Epochs: 10
• Validation Split: 20%
• Transfer Learning: Base model frozen

Performance

• Validation Accuracy: {history_mobilenet.history['val_accuracy'][-1]:.4f}
• Validation Loss: {history_mobilenet.history['val_loss'][-1]:.4f}