xkungfu

> I guess you could do this but I would suggest allocating QWidgets on the main thread instead. It seems inefficient and odd that you're creating and using 50000 plots....

updates: I am using qt creator 5.0.0 on windows 10.

1: download qpm.exe to asyncfuture/tests/asyncfutureunittests 2: cd asyncfuture/tests/asyncfutureunittests 3: qpm install [email protected] qpm install [email protected] qpm install [email protected] I have do all the steps above. the error is: TestRunner, QtQuickTest,...

after reinstall qt creator and change to msvc kit resolve the problem. still fail for mingw64 on windows. it seems lost some libs if using mingw64.

没有解决。我无能为力了。你如果找到解决办法了麻烦分享一下，谢谢！你QQ或微信方便加一下吗？

我可能知道是什么原因了，可能是和机器性能有关。需要高配机器。我是4核8G云服务器。训练完可能要好几天或者干脆无法结束。

谢谢。你有高配机器的话，可以试试。 我用了另一个可以跑通，并且作者一直在线回复问题： https://github.com/yongzhuo/Keras-TextClassification

> 试了，能跑通，代码没有问题。 训练好的模型文件可否上传到网盘分享一下呢？谢谢！

> > INFO:tensorflow:Saving checkpoints for 0 into ...../model.ckpt. > > I1107 14:10:38.075445 140053106304832 basic_session_run_hooks.py:606] Saving checkpoints for 0 into ....../model.ckpt. > > 到这就停了。 > > Top命令，也没找到python的进程。 > > 4核cpu. ubuntu18.04....

that is : if I load data from a csv file, all values in it must be number? as you know, all financial data have a column date, and ticker,...