Error when trying to decode with discriminative model #15
Description
I am trying to run the following command for decoding with discriminative model:
build/nt-parser/nt-parser --cnn-mem 1700 -x -T train-gen.oracle -p test-gen.oracle -C sample_input_english.txt -P --pretrained_dim 256 -w 256 --lstm_input_dim 128 --hidden_dim 128 -m oracles/lm_0_2_128_384-pid103351.params --alpha 0.8 -s 100 > test-samples.props
but am getting this error:
[cnn] random seed: 125536842
[cnn] allocating memory: 1700MB
[cnn] memory allocation done.
COMMAND LINE: build/nt-parser/nt-parser -x -T train-gen.oracle -p test-gen.oracle -C sample_input_english.txt -P --pretrained_dim 256 -w 256 --lstm_input_dim 128 --hidden_dim 128 -m pretrained-gen-new.params --alpha 0.8 -s 100
PARAMETER FILE: ntparse_pos_0_2_32_128_16_128-pid133.params
Loading top-down oracle from train-gen.oracle [training] ...
Loaded 0 sentences
cumulative action vocab size: 2
cumulative terminal vocab size: 0
cumulative nonterminal vocab size: 0
cumulative pos vocab size: 0
Reading pretrained embeddings from 256 ...
File does not seem to be in word2vec format
Aborted (core dumped)
How can I solve this?