Question

我正在尝试运行神经网络算法，但出现错误“ ValueError：生成器已在执行”。我读过我可以通过使worker = 1来解决此问题，但这没有帮助。

我已经看过this post，但不确定如何实现它，或者我是否有必要。我也尝试设置use_multiprocessing = True，但这给出了另一个错误。我已经附上了我认为相关的Python代码和完整的错误代码。在Anaconda / Spyder中从2.2.4降级后，我正在使用Keras 2.0.8。

import os
os.environ["CUDA_VISIBLE_DEVICES"] = ""

from keras.optimizers import Adam
from keras.callbacks import EarlyStopping, TensorBoard, ModelCheckpoint, ReduceLROnPlateau
import matplotlib.pyplot as plt
from DataGenerator import generator
import datetime #, os
import models
import numpy as np 
from metrics import metrics_callback, weighted_categorical_crossentropy
from multiprocessing import Pool

batch_size = 16
n_epochs = 150

windowsize = 1024
data_gen = generator(folder=datafolder, batch_size=batch_size, windowsize=windowsize)

train_gen,train_batches = data_gen.train_gen_online()
test_gen,test_batches = data_gen.test_gen()

train_gen=train_gen()
test_gen=test_gen()

.....

##ERROR HAPPENS ON LINE BELOW
model.fit_generator(generator=train_gen,
                    steps_per_epoch=train_batches,
                    epochs=1,
                    validation_data=train_gen,
                    validation_steps = 100,
                    max_queue_size=40,
                    workers=1,
                    use_multiprocessing=False
                    )
test_x, test_y = next(test_gen) 
test_pred = model.predict(test_x)

由于没有多重处理的事实，我希望代码能够运行并提供神经网络作为输出，尽管非常慢。


Epoch 1/1
10/11 [==========================>...] - ETA: 42s - loss: 6246.2208 - mean_squared_error: 6246.2208 Exception in thread Thread-8:
Traceback (most recent call last):
  File "C:\Users\s164286\AppData\Local\Continuum\anaconda3\lib\threading.py", line 916, in _bootstrap_inner
    self.run()
  File "C:\Users\s164286\AppData\Local\Continuum\anaconda3\lib\threading.py", line 864, in run
    self._target(*self._args, **self._kwargs)
  File "C:\Users\s164286\AppData\Local\Continuum\anaconda3\lib\site-packages\keras\utils\data_utils.py", line 568, in data_generator_task
    generator_output = next(self._generator)
ValueError: generator already executing

如何修复“ ValueError：生成器已在执行”

0 个答案: