search

schelotto / Neural_Speed_Reading_via_Skim-RNN_PyTorch

PyTorch implementation of "Neural Speed Reading via Skim-RNN"
MIT License
18 stars 9 forks source link

RuntimeError: size mismatch at /pytorch/torch/lib/THC/generic/THCTensorMathBlas.cu:243 #7

Open maomaoamaomao opened 6 years ago

maomaoamaomao commented 6 years ago

Loading data...

Parameters: BATCH_SIZE=32 CUDA=True DEVICE=0 DROPOUT=0.5 EMBED_DIM=300 EPOCHS=20 GAMMA=0.1 HIDDEN_LAYER=200 LARGE_CELL_SIZE=200 LOG_INTERVAL=1 LR=0.001 N_CLASS=1997 NUM_LAYERS=1 PREDICT=None SAVE_DIR=snapshot/2018-04-19_16-16-43 SAVE_INTERVAL=500 SHUFFLE=False SMALL_CELL_SIZE=10 SNAPSHOT=None TAU=0.5 TEST=False TEST_INTERVAL=100 VOCAB_SIZE=25208 WORD_DICT=<torchtext.vocab.Vocab object at 0x7f2bc5001390> Traceback (most recent call last): File "/home/nice/maomao/Python/Skim-rnn/main.py", line 111, in logits, h_stack, p_stack = skim_rnn_classifier(sent) File "/home/nice/anaconda3/envs/pytorch0.3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 325, in call result = self.forward(*input, *kwargs) File "/home/nice/maomao/Python/Skim-rnn/skim_rnn.py", line 127, in forward pt = self.linear(torch.cat([embed.contiguous().view(-1, self.embed_dim), h_statel, cl], 1)) File "/home/nice/anaconda3/envs/pytorch0.3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 325, in call result = self.forward(input, **kwargs) File "/home/nice/anaconda3/envs/pytorch0.3/lib/python3.6/site-packages/torch/nn/modules/linear.py", line 55, in forward return F.linear(input, self.weight, self.bias) File "/home/nice/anaconda3/envs/pytorch0.3/lib/python3.6/site-packages/torch/nn/functional.py", line 835, in linear return torch.addmm(bias, input, weight.t()) RuntimeError: size mismatch at /pytorch/torch/lib/THC/generic/THCTensorMathBlas.cu:243

This problem has been bothering me for days. I only run your code and it doesn't change.