Your input ran out of data； interrupting training. Make sure that your dataset or generator can ....

《Python 深度学习》第5章  P114 在运行以下代码有个报错：

history = model.fit_generator(
      train_generator,
      steps_per_epoch=100,
      epochs=100,
      validation_data=validation_generator,
      validation_steps=50)

Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches (in this case, 10000 batches). You may need to use the repeat() function when building your dataset.

翻译：您的输入数据不足；中断训练。确保您的数据集或生成器至少可以生成“steps_per_epoch*epochs”批。（在本例中为10000批）。在构建数据集时，可能需要使用repeat（）函数。

报错产生原因：

我们训练图像有2000张，验证图像1000张，和1000张测试图像。

 steps_per_epoch=100，batch_size=32，如此数据应该是3200张，很明显输入训练数据不够。
 validation_steps=50，batch_size=32，如此数据应该是1600张，很明显验证数据不够。

因此，改为steps_per_epoch=2000/32≈63，validation_steps=1000/32≈32。

history = model.fit_generator(
      train_generator,
      steps_per_epoch=63,
      epochs=100,
      validation_data=validation_generator,
      validation_steps=32)

 可以运行！但是可能还是有些纰漏。

---

参考链接：

WARNING:tensorflow:Your input ran out of data； interrupting training. Make sure that your dataset or

Your input ran out of data； interrupting training. Make sure that your dataset or generat