SivilTaram
commented
3 years ago @1973sir I cannot reproduce the errors. I think the installation of transformers library is not correct (we have customized some parts to support the auxiliary task).
Please see here to see how to install the correct dependencies. Hope it helps you.
1973sir
The following parameters will be optimized NORMALLY: linear.weight [ training... ] [ Saving tensorboard logs here: ./tmp/transmitter/tensorboard ] preinitializing pytorch cuda buffer Traceback (most recent call last): File "train_transmitter.py", line 145, in
TrainLoop(opt).train()
File "/disk1/lenny/share/PersonaDG/Persona-Dialogue-Generation/parlai/scripts/train_model.py", line 512, in train
world.parley()
File "/disk1/lenny/share/PersonaDG/Persona-Dialogue-Generation/parlai/core/worlds.py", line 663, in parley
batch_act = self.batch_act(agent_idx, batch_observations[agent_idx])
File "/disk1/lenny/share/PersonaDG/Persona-Dialogue-Generation/parlai/core/worlds.py", line 636, in batch_act
batch_actions = a.batch_act(batch_observation)
File "/disk1/lenny/share/PersonaDG/Persona-Dialogue-Generation/agents/transmitter/transmitter.py", line 857, in batch_act
self.init_cuda_buffer(batchsize)
File "/disk1/lenny/share/PersonaDG/Persona-Dialogue-Generation/agents/transmitter/transmitter.py", line 842, in init_cuda_buffer
sc = self.model(input_dummy, None, None, output_dummy, None)[1]
File "/home/lenny/anaconda3/envs/t18p20/lib/python3.7/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, **kwargs)
File "/disk1/lenny/share/PersonaDG/Persona-Dialogue-Generation/agents/transmitter/gpt/model.py", line 91, in forward
lm_logits, hidden_states = self.transformer_module(input_seq, None, dis_seq)
TypeError: 'NoneType' object is not callable