I created a neural network which takes images of people as X_training values and their respective genders (binary value) as the Y_train values, where my goal is to predict the relevant gender once a user enters a image. Here is the code where i set the image and the gender values as training data:
from sklearn.model_selection import train_test_split
images_f=np.array(images)
images_f_2=images_f/255
labels_f=np.array(genders)
labels_f:
array([1, 0, 1, ..., 0, 0, 0])
I basically use convolutional layers and since im predicting a binary value(male or female, 0 or 1), i used sigmoid as my final dense layer activation method.
Here is the model code:
# Create a model and add layers
model = Sequential()
model.add(Conv2D(32, (3, 3), padding='same', input_shape=(48, 48, 3),strides=(1, 1),kernel_regularizer=l2(0.001), activation="relu"))
model.add(Conv2D(32, (3, 3), activation="relu"))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Conv2D(64, (3, 3), padding='same', activation="relu"))
model.add(Conv2D(64, (3, 3), activation="relu"))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Conv2D(128, (3, 3), padding='same', activation="relu"))
model.add(Conv2D(128, (3, 3), activation="relu"))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(512, activation="relu"))
model.add(Dropout(0.5))
model.add(Dense(1, activation="sigmoid"))
model.summary()
model.compile(
loss='binary_crossentropy',
optimizer='adam',
metrics=['accuracy']
)
This is where i fit the values code:
# Train the model
model.fit(
X_train,
[Y_train],
batch_size=64,
epochs=30,
validation_data=(X_test, Y_test),
shuffle=True
)
Now once i trained the model, i tried to call the model and set one of my own images and predict.
from keras.preprocessing import image
img = image.load_img("queen.jpg", target_size=(48, 48,3))
# Convert the image to a numpy array
img = image.img_to_array(img)
# Add a forth dimension to the image (since Keras expects a bunch of images, not a single image)
img/=255
img = np.expand_dims(img, axis=0)
result = model.predict(img)
And the result value i get it :
array([[0.06528784]], dtype=float32)
For each and every image i get a float value but im expecting a binary value. Why?