Alpaca / Not Alpaca Binary Classification (MobileNetV2)

You can upload your own image and see what the model says about it.

We can improve this model by changing the model architecture, and adding more training data. At this level I wanted to showcase my work towards implementation and deployment of Deep Learning model.

Try to pick squared image | Max size: 2 Mb | File types: png, jpg, jpeg

Development summary

High-level architecture of the application

Model summary

Training accuracy: 100% Validation accuracy: 95% 
                Model: "model_3"
_________________________________________________________________
 Layer (type)                Output Shape              Param #   
=================================================================
 input_11 (InputLayer)       [(None, 160, 160, 3)]     0         
                                                                 
 sequential_6 (Sequential)   (None, 160, 160, 3)       0         
                                                                 
 tf.math.truediv_3 (TFOpLam  (None, 160, 160, 3)       0         
 bda)                                                            
                                                                 
 tf.math.subtract_3 (TFOpLa  (None, 160, 160, 3)       0         
 mbda)                                                           
                                                                 
 mobilenetv2_1.00_160 (Func  (None, 5, 5, 1280)        2257984   
 tional)                                                         
                                                                 
 global_average_pooling2d_6  (None, 1280)              0         
  (GlobalAveragePooling2D)                                       
                                                                 
 dropout_3 (Dropout)         (None, 1280)              0         
                                                                 
 dense_3 (Dense)             (None, 1)                 1281      
                                                                 
=================================================================
Total params: 2259265 (8.62 MB)
Trainable params: 2225153 (8.49 MB)
Non-trainable params: 34112 (133.25 KB)
_________________________________________________________________