Open mithunpaul08 opened 5 years ago
Hi, I checked my env, it should be pytorch 0.3.1
On Fri, 22 Mar 2019 at 01:18, Mithun Paul notifications@github.com wrote:
Hey, i keep running into this error. I googled it up and looks like your code is fine. Hence makes me suspect pytorch might have changed something in the newer versions. do you remember what version of pytorch and python you were using? am also pasting below my pytorch versions, if that helps.
Traceback (most recent call last): File "train_baseline_snli.py", line 352, in
train(args) File "train_baseline_snli.py", line 154, in train if m.bias: RuntimeError: bool value of Tensor with more than one value is ambiguous pycparser 2.19 py37_0 python 3.7.2 h0371630_0 pytorch 1.0.1 py3.7_cuda8.0.61_cudnn7.1.2_2 pytorch
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/libowen2121/SNLI-decomposable-attention/issues/6, or mute the thread https://github.com/notifications/unsubscribe-auth/ATC5BAiRft1YnG5YvTNBK3QEsFzxL0IFks5vZC-CgaJpZM4cCqZr .
-- Bowen Li
School of Informatics The University of Edinburgh
I have the same problem. "if m.bias is not None:" works for me
Hey, i keep running into this error. I googled it up and looks like your code is fine. Hence makes me suspect pytorch might have changed something in the newer versions. do you remember what version of pytorch and python you were using? am also pasting below my pytorch versions, if that helps.