小编Jen*_*ein的帖子

Hi\xe2\x80\xa6 I\xe2\x80\x99m 在我的 P3 AWS 机器中运行 mnist 代码，与我之前的 P2 机器相比，初始化过程似乎很长（尽管 P3>P2）

Train on 60000 samples, validate on 10000 samples\nEpoch 1/10\n60000/60000 [==============================] - 265s 4ms/step - loss: 0.2674 - acc: 0.9175 - val_loss: 0.0602 - val_acc: 0.9811\nEpoch 2/10\n60000/60000 [==============================] - 3s 51us/step - loss: 0.0860 - acc: 0.9742 - val_loss: 0.0393 - val_acc: 0.9866\nEpoch 3/10\n60000/60000 [==============================] - 3s 50us/step - loss: 0.0647 - acc: 0.9808 - val_loss: 0.0338 - val_acc: 0.9884\nEpoch 4/10\n60000/60000 [==============================] - 3s 50us/step - loss: 0.0542 - …