python - Keras 的 fit_generator 额外训练值

Question

train_datagen = ImageDataGenerator(
                    rescale=1./255,
                    shear_range=0.1,
                    zoom_range=0.1,
                    rotation_range=5.,
                    width_shift_range=0.1,
                    height_shift_range=0.1)

val_datagen = ImageDataGenerator(rescale=1./255)

train_generator = train_datagen.flow_from_directory(
                    train_data_dir,
                    target_size = (img_width, img_height),
                    batch_size = 20,
                    shuffle = True,
                    classes = TYPES,
                class_mode = 'categorical')

validation_generator = val_datagen.flow_from_directory(
                    val_data_dir,
                    target_size=(img_width, img_height),
                    batch_size=20,
                    shuffle = True,
                    classes = TYPES,
                    class_mode = 'categorical')

model.fit_generator(
                train_generator,
                samples_per_epoch = 2000,
                nb_epoch = 20
            )

Epoch 14/50
 480/2000 [======>.......................] - ETA: 128s - loss: 0.8708

Epoch 13/50
2021/2000 [==============================] - 171s - loss: 0.7973 - acc: 0.7041 

我的 ImageGenerators 从文件夹中读取 2261 个训练图像和 567 个测试图像。我正在尝试使用 2000 个 samples_per_epoch 和 20 个 batch_size 来训练我的模型。 batch_size 对于 samples_per_epoch 是可分割的，但不知何故它增加了额外的值(value)并显示警告:

( UserWarning: Epoch comprised more than samples_per_epoch samples, which might affect learning results. Set samples_per_epoch correctly to avoid this warning).

它适用于 Single-Gpu 但如果我尝试训练使用 Multi-Gpus 时会出现该错误:

InvalidArgumentError (see above for traceback): Incompatible shapes: [21] vs. [20] [[Node: Equal = Equal[T=DT_INT64, _device="/job:localhost/replica:0/task:0/gpu:0"](ArgMax, ArgMax_1)]] [[Node: gradients/concat_25_grad/Slice_1/_10811 = _Recvclient_terminated=false, recv_device="/job:localhost/replica:0/task:0/gpu:1", send_device="/job:localhost/replica:0/task:0/cpu:0", send_device_incarnation=1, tensor_name="edge_101540_gradients/concat_25_grad/Slice_1", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/gpu:1"]]

我正在使用 code对于模型并行化:

感谢您的帮助...

Answer 1

训练数据的数量必须等于samples_per_epoch ×∀batch_size。请减少1条数据，使训练数据条数达到2260条。steps_per_epoch=113batch_size=20