keras中的model.fit(加载模型后未训练)和model.predict的不同输出

时间:2019-06-14 03:23:49

标签: python keras deep-learning

def get_vgg_twoeyes()是我的模型的定义。 我已经加载了预先训练的模型,该模型在同一台计算机上进行了训练,然后我想对模型进行微调。在重新训练模型之前,我设置了model.trainable false,以确保模型的权重固定。训练之前,权重与保存的权重相同。我发现model.fit的输出与model.predict的输出不同。 我假设与model.predict具有相同权重的model.fit应该输出相同的结果,因为model.trainable为false意味着model.fit表现了model.predict。

def get_vgg_twoeyes(optimizer ='adam',model_type ='VGG16',fc1_size = 1024,                     fc2_size = 512,fc3_size = 256):

kern_init = initializers.glorot_normal()
img_input = Input(shape=(36, 60, 3), name='img_input')
headpose_input = Input(shape=(2,), name='headpose_input')

# create the base pre-trained model
if model_type == 'VGG19':
    base_model = VGG19(input_tensor=img_input, weights='imagenet', include_top=False)
elif model_type == 'VGG16':
    base_model = VGG16(input_tensor=img_input, weights='imagenet', include_top=False)
else:
    raise Exception('Unknown model type in get_vgg_twoeyes')

# add a global spatial average pooling layer
x = base_model.output
x = GlobalAveragePooling2D()(x)

# let's add a fully-connected layer
x = Dense(fc1_size, kernel_initializer=kern_init)(x)
x = concatenate([x, headpose_input])
x = BatchNormalization()(x)
x = Activation('relu')(x)

x = Dense(fc2_size, kernel_initializer=kern_init)(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)

gaze_predictions = Dense(2, kernel_initializer=kern_init, name='pred_gaze')(x)

# this is the model we will train
model = Model(inputs=[img_input, headpose_input], outputs=gaze_predictions)
model.compile(optimizer=optimizer, loss=angle_loss, metrics=['accuracy', accuracy_angle])
return model

# fine-tune the model
models=load_model(model_path + "15Fold" + prefix + ''+str(i) + 
    suffix + ".h5",custom_objects={'accuracy_angle':accuracy_angle, 
     'angle_loss': angle_loss}))

model.trainable=False
adam = Adam(lr=0.0001, beta_1=0.9, beta_2=0.95)
model.compile(optimizer=adam, loss=angle_loss, metrics= ['accuracy', accuracy_angle]) 
model.fit({'img_input':cal_images,'headpose_input':cal_headposes},
cal_gazes,shuffle=False,batch_size=32,epochs=1,callbacks= 
  [losshistory()])

predgaze=model.predict({'img_input': cal_images, 'headpose_input': 
   cal_headposes},  batch_size=2,verbose=1)

1 个答案:

答案 0 :(得分:0)

设置model.trainable=False后,您可能必须再次编译模型。另外,您可以像

一样分别手动冻结图层

for l in model.layers: l.trainable=False