search

facebookresearch / end-to-end-negotiator

Deal or No Deal? End-to-End Learning for Negotiation Dialogues
Other
1.38k stars 276 forks source link

Dimension out of range for Softmax in dialog_model.py #16

Closed shivamswarnkar closed 6 years ago

shivamswarnkar commented 6 years ago

I am getting this error when I try to run reinforce.py. Traceback (most recent call last): File "reinforce.py", line 164, in <module> main() File "reinforce.py", line 158, in main reinforce.run() File "reinforce.py", line 57, in run self.dialog.run(ctxs, self.logger) File "/home/ss8464/dl/nego_comm/main/src/dialog.py", line 170, in run out = writer.write() File "/home/ss8464/dl/nego_comm/main/src/agent.py", line 128, in write 100, self.args.temperature) File "/home/ss8464/dl/nego_comm/main/src/models/dialog_model.py", line 300, in write prob = F.softmax(scores) File "/share/apps/pytorch/20171124/python3.6/gnu/lib/python3.6/site-packages/torch/nn/functional.py", line 818, in softmax return torch._C._nn.softmax(input, dim) RuntimeError: dimension out of range (expected to be in range of [-1, 0], but got 1)

denisyarats commented 6 years ago

Hi, could you please share the arguments?