ngram: 3
project_name: temp_model
dataset: r8
trainable_edges: 1
trainable edges
load data helper.
new model.
/data/yuhai/tc/TextLevelGCN/pmi.py:38: RuntimeWarning: invalid value encountered in double_scalars
pair_count_matrix[i, j] / (word_count[i] word_count[j])
/data/yuhai/tc/TextLevelGCN/pmi.py:38: RuntimeWarning: divide by zero encountered in log
pair_count_matrix[i, j] / (word_count[i] word_count[j])
1380208
torch.Size([1380208, 1])
Traceback (most recent call last):
File "/home/yuhai/workspace/tc/TextLevelGCN/train.py", line 237, in
model = train(args.ngram, args.name, bar, args.dropout, dataset=args.dataset, is_cuda=True, edges=edges)
File "/home/yuhai/workspace/tc/TextLevelGCN/train.py", line 107, in train
trainable_edges=edges, pmi=edges_weights, cuda=is_cuda)
File "/data/yuhai/tc/TextLevelGCN/model.py", line 63, in init
self.nodehidden.weight.data.copy(torch.tensor(self.load_word2vec('glove.6B.200d.vec.txt')))
File "/data/yuhai/tc/TextLevelGCN/model.py", line 91, in load_word2vec
model = word2vec.load(word2vec_file)
File "/home/yuhai/workspace/anaconda3/envs/torch1.3/lib/python3.7/site-packages/word2vec/io.py", line 18, in load
return word2vec.WordVectors.from_text(fname, *args, **kwargs)
File "/home/yuhai/workspace/anaconda3/envs/torch1.3/lib/python3.7/site-packages/word2vec/wordvectors.py", line 262, in from_text
vectors[i] = unitvec(vector)
ValueError: could not broadcast input array from shape (14) into shape (200)
ngram: 3 project_name: temp_model dataset: r8 trainable_edges: 1 trainable edges load data helper. new model. /data/yuhai/tc/TextLevelGCN/pmi.py:38: RuntimeWarning: invalid value encountered in double_scalars pair_count_matrix[i, j] / (word_count[i] word_count[j]) /data/yuhai/tc/TextLevelGCN/pmi.py:38: RuntimeWarning: divide by zero encountered in log pair_count_matrix[i, j] / (word_count[i] word_count[j]) 1380208 torch.Size([1380208, 1]) Traceback (most recent call last): File "/home/yuhai/workspace/tc/TextLevelGCN/train.py", line 237, in
model = train(args.ngram, args.name, bar, args.dropout, dataset=args.dataset, is_cuda=True, edges=edges)
File "/home/yuhai/workspace/tc/TextLevelGCN/train.py", line 107, in train
trainable_edges=edges, pmi=edges_weights, cuda=is_cuda)
File "/data/yuhai/tc/TextLevelGCN/model.py", line 63, in init
self.nodehidden.weight.data.copy(torch.tensor(self.load_word2vec('glove.6B.200d.vec.txt')))
File "/data/yuhai/tc/TextLevelGCN/model.py", line 91, in load_word2vec
model = word2vec.load(word2vec_file)
File "/home/yuhai/workspace/anaconda3/envs/torch1.3/lib/python3.7/site-packages/word2vec/io.py", line 18, in load
return word2vec.WordVectors.from_text(fname, *args, **kwargs)
File "/home/yuhai/workspace/anaconda3/envs/torch1.3/lib/python3.7/site-packages/word2vec/wordvectors.py", line 262, in from_text
vectors[i] = unitvec(vector)
ValueError: could not broadcast input array from shape (14) into shape (200)
Process finished with exit code 1