imran3180 / pytorch-nli Goto Github PK

How do I put character level embeddings for the text field

Hi
I want to train a basic NLI model, for I followed the README and just change the dataset into "multinli".
Here is my command
"python train.py --dataset=multinli --model=bilstm"
However, the error message appeared
"Traceback (most recent call last):
File "train.py", line 99, in
task = Train()
File "train.py", line 29, in init
self.dataset = datasets.dictself.args.dataset
File "/net/setagaya/disk6/ztc/nlifile/pytorch-nli/datasets/multinli.py", line 30, in multinli
return MultiNLI(options)
File "/net/setagaya/disk6/ztc/nlifile/pytorch-nli/datasets/multinli.py", line 12, in init
self.inputs = data.Field(lower=True, tokenize='spacy', batch_first = True)
File "/home/ztc/centos6/pyenv/pyenv/versions/3.6.2/lib/python3.6/site-packages/torchtext/data/field.py", line 163, in init
self.tokenize = get_tokenizer(tokenize, tokenizer_language)
File "/home/ztc/centos6/pyenv/pyenv/versions/3.6.2/lib/python3.6/site-packages/torchtext/data/utils.py", line 114, in get_tokenizer
spacy = spacy.load(language)
File "/home/ztc/centos6/pyenv/pyenv/versions/3.6.2/lib/python3.6/site-packages/spacy/init.py", line 30, in load
return util.load_model(name, **overrides)
File "/home/ztc/centos6/pyenv/pyenv/versions/3.6.2/lib/python3.6/site-packages/spacy/util.py", line 169, in load_model
raise IOError(Errors.E050.format(name=name))
OSError: [E050] Can't find model 'en'. It doesn't seem to be a shortcut link, a Python package or a valid path to a data directory."

How can I solve it?
Thank you in advance!

Hi, does your code support the TSV format dataset? Or any suggestion to run your system for the TSV format datasets?