Open mstc-xqp opened 5 years ago
I use chainer 5.0,while it is written is chainer 1.90 or 3.0?
I use chainer 5.0,while it is written is chainer 1.90 or 3.0?
Chainer 1.10.0 is OK.
The chainer version 1.10.0 requires cuda and cudnn environment, but I don't have a NIVIDA gpu and just want to use default cpu so I installed chainer 3.0.0. I tried to use chainer 3.0.0 or 5.0.0 to run, but there are some error: ValueError: train argument is not supported anymore. Use chainer.using_config. I guess these arguments are used only in chainer version 1, right? So I'm in a deadlock?
I use chainer 5.0,while it is written is chainer 1.90 or 3.0?
Chainer 1.10.0 is OK. 请问你有用cupy吗你用哪个版本呢,你有复现出来吗 ,请问可以加一下微信交流吗,我微信是a409197863
我用CPU版本跑了mnist 要一天,模型能出来,就是可视化那里报错
Traceback (most recent call last):
File "experiment_lenet_mnist.py", line 129, in
File "/usr/lib/python2.7/site-packages/chainer/link.py", line 242, in call out = forward(*args, **kwargs) TypeError: forward() takes exactly 2 arguments (3 given)