Closed simon-p-j-r closed 1 year ago
Hi, did you follow our commands in README to run generate_data_degree_eae.py
/generate_data_degree_e2e.py
/generate_data_degree_ed.py
to generate the .pkl files first?
Thanks!!! It's my carelessness.
but when I train degree(end2end), an error occured:
[2023-03-10 12:44:46] - data - Loaded 292257 instances from ./data/degree_e2e_ace05ep/train_all.pkl
[2023-03-10 12:44:47] - data - Loaded 13929 instances from ./data/degree_e2e_ace05ep/dev_all.pkl
[2023-03-10 12:44:47] - data - Loaded 10535 instances from ./data/degree_e2e_ace05ep/test_all.pkl
[2023-03-10 12:44:47] - model - Loading pre-trained model ./bart-large
[2023-03-10 12:45:40] - main - Start training ...
[2023-03-10 12:45:40] - main - ./output/degree_e2e_ace05ep/20230310_124441/train.log
[2023-03-10 12:45:40] - main - Epoch 1
Train 1: 0%| | 0/9134 [00:00<?, ?it/s]Traceback (most recent call last):
File "degree/train_degree_e2e.py", line 106, in <module>
loss = model(batch)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, kwargs)
File "/home/net/pjr/DEGREE-master/DEGREE-master/degree/model.py", line 20, in forward
outputs = self.model(input_ids=batch.enc_idxs,
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, *kwargs)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/transformers/modeling_bart.py", line 1066, in forward
outputs = self.model(
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(input, kwargs)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/transformers/modeling_bart.py", line 914, in forward
encoder_outputs = self.encoder(
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, kwargs)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/transformers/modeling_bart.py", line 352, in forward
x, attn = encoder_layer(x, attention_mask, output_attentions=output_attentions)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, *kwargs)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/transformers/modeling_bart.py", line 253, in forward
x, attn_weights = self.self_attn(
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(input, kwargs)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/transformers/modeling_bart.py", line 672, in forward
q = self.q_proj(query) self.scaling
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(input, **kwargs)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/linear.py", line 94, in forward
return F.linear(input, self.weight, self.bias)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/functional.py", line 1753, in linear
return torch._C._nn.linear(input, weight, bias)
RuntimeError: CUDA error: CUBLAS_STATUS_INTERNAL_ERROR when calling cublasCreate(handle)
Train 1: 0%| | 0/9134 [00:01<?, ?it/s]
Did you have the error when you run the code? My virtual environment is exactly the same as Readme. Thanks for your reply!!!!
------------------ 原始邮件 ------------------ 发件人: "PlusLabNLP/DEGREE" @.>; 发送时间: 2023年3月10日(星期五) 中午12:31 @.>; @.**@.>; 主题: Re: [PlusLabNLP/DEGREE] ".pkl"files (Issue #7)
Hi, did you follow our commands in README to run generate_data_degree_eae.py/generate_data_degree_e2e.py/generate_data_degree_ed.py/ to generate the .pkl files first?
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>
Thanks!!! It's my carelessness.
but when I train degree(end2end), an error occured:
[2023-03-10 12:44:46] - data - Loaded 292257 instances from ./data/degree_e2e_ace05ep/train_all.pkl
[2023-03-10 12:44:47] - data - Loaded 13929 instances from ./data/degree_e2e_ace05ep/dev_all.pkl
[2023-03-10 12:44:47] - data - Loaded 10535 instances from ./data/degree_e2e_ace05ep/test_all.pkl
[2023-03-10 12:44:47] - model - Loading pre-trained model ./bart-large
[2023-03-10 12:45:40] - main - Start training ...
[2023-03-10 12:45:40] - main - ./output/degree_e2e_ace05ep/20230310_124441/train.log
[2023-03-10 12:45:40] - main - Epoch 1
Train 1: 0%| | 0/9134 [00:00<?, ?it/s]Traceback (most recent call last):
File "degree/train_degree_e2e.py", line 106, in <module>
loss = model(batch)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, kwargs)
File "/home/net/pjr/DEGREE-master/DEGREE-master/degree/model.py", line 20, in forward
outputs = self.model(input_ids=batch.enc_idxs,
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, *kwargs)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/transformers/modeling_bart.py", line 1066, in forward
outputs = self.model(
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(input, kwargs)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/transformers/modeling_bart.py", line 914, in forward
encoder_outputs = self.encoder(
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, kwargs)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/transformers/modeling_bart.py", line 352, in forward
x, attn = encoder_layer(x, attention_mask, output_attentions=output_attentions)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(*input, *kwargs)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/transformers/modeling_bart.py", line 253, in forward
x, attn_weights = self.self_attn(
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(input, kwargs)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/transformers/modeling_bart.py", line 672, in forward
q = self.q_proj(query) self.scaling
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
result = self.forward(input, **kwargs)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/modules/linear.py", line 94, in forward
return F.linear(input, self.weight, self.bias)
File "/home/net/anaconda3/envs/pjrDEGREE/lib/python3.8/site-packages/torch/nn/functional.py", line 1753, in linear
return torch._C._nn.linear(input, weight, bias)
RuntimeError: CUDA error: CUBLAS_STATUS_INTERNAL_ERROR when calling cublasCreate(handle)
Train 1: 0%| | 0/9134 [00:01<?, ?it/s]
Did you have the error when you run the code? My virtual environment is exactly the same as Readme. Thanks for your reply!!!!
I fix it, by install torch==1.7.1. Thanks for your patient so much, I will close the issue. Thanks again
------------------ 原始邮件 ------------------ 发件人: "PlusLabNLP/DEGREE" @.>; 发送时间: 2023年3月10日(星期五) 中午12:31 @.>; @.**@.>; 主题: Re: [PlusLabNLP/DEGREE] ".pkl"files (Issue #7)
Hi, did you follow our commands in README to run generate_data_degree_eae.py/generate_data_degree_e2e.py/generate_data_degree_ed.py/ to generate the .pkl files first?
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>
hello, dear author. Where can I get train_all.pkl, dev_all.pkl and test_all.pkl files?