mae

age_estimation/model.py

@@ -2,6 +2,14 @@
 from keras.applications import ResNet50, InceptionResNetV2
 from keras.layers import Dense
 from keras.models import Model
+from keras import backend as K
+
+
+def age_mae(y_true, y_pred):
+    true_age = K.mean(y_true * K.arange(0, 101), axis=-1)
+    pred_age = K.mean(y_pred * K.arange(0, 101), axis=-1)
+    mae = K.mean(K.abs(true_age - pred_age))
+    return mae
 
 
 def get_model(model_name="ResNet50"):

age_estimation/train.py

@@ -4,7 +4,8 @@
 from keras.callbacks import LearningRateScheduler, ModelCheckpoint
 from keras.optimizers import SGD
 from generator import FaceGenerator, ValGenerator
-from model import get_model
+from model import get_model, age_mae
+
 
 class Schedule:
     def __init__(self, nb_epochs):
@@ -56,17 +57,16 @@ def main():
     val_gen = ValGenerator(appa_dir, batch_size=batch_size, image_size=image_size)
     model = get_model(model_name=model_name)
     sgd = SGD(lr=0.1, momentum=0.9, nesterov=True)
-    model.compile(optimizer=sgd, loss="categorical_crossentropy",
-                  metrics=['accuracy'])
+    model.compile(optimizer=sgd, loss="categorical_crossentropy", metrics=[age_mae])
     model.summary()
     output_dir = Path(__file__).resolve().parent.joinpath(args.output_dir)
     output_dir.mkdir(parents=True, exist_ok=True)
     callbacks = [LearningRateScheduler(schedule=Schedule(nb_epochs)),
-                 ModelCheckpoint(str(output_dir) + "/weights.{epoch:03d}-{val_loss:.3f}.hdf5",
-                                 monitor="val_loss",
+                 ModelCheckpoint(str(output_dir) + "/weights.{epoch:03d}-{val_loss:.3f}-{val_age_mae:.3f}.hdf5",
+                                 monitor="val_age_mae",
                                  verbose=1,
                                  save_best_only=True,
-                                 mode="auto")
+                                 mode=min")
                  ]
 
     hist = model.fit_generator(generator=train_gen,