Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Bugfix/utils get and tokenize dataset #4

Open
wants to merge 4 commits into
base: master
Choose a base branch
from
Open

Bugfix/utils get and tokenize dataset #4

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
3ce6b2b
Fixed issue: invalid index while using dataset_map['labels']
sahilsharma05 Sep 6, 2019
cb56df7
Fixed issue:
sahilsharma05 Sep 6, 2019
ac9f44b
delete extra added code
sahilsharma05 Sep 6, 2019
ec5e9bb
added unit test case for get_and_tokenize_dataset method
sahilsharma05 Sep 6, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 13 additions & 6 deletions utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@
from ignite.contrib.handlers import ProgressBar
from ignite.contrib.handlers.tensorboard_logger import OptimizerParamsHandler, OutputHandler, TensorboardLogger

from pytorch_pretrained_bert import cached_path
from pytorch_pretrained_bert import cached_path, BertTokenizer

DATASETS_URL = {
'wikitext-2': {'train': "https://s3.amazonaws.com/datasets.huggingface.co/wikitext-2/train.txt",
Expand All @@ -25,16 +25,16 @@
'simplebooks-92-raw': {'train': "https://s3.amazonaws.com/datasets.huggingface.co/simplebooks-92-raw/train.txt",
'valid': "https://s3.amazonaws.com/datasets.huggingface.co/simplebooks-92-raw/valid.txt"},
'imdb': {'train': "https://s3.amazonaws.com/datasets.huggingface.co/aclImdb/train.txt",
'test': "https://s3.amazonaws.com/datasets.huggingface.co/aclImdb/test.txt"},
'valid': "https://s3.amazonaws.com/datasets.huggingface.co/aclImdb/valid.txt"},
'trec': {'train': "https://s3.amazonaws.com/datasets.huggingface.co/trec/train.txt",
'test': "https://s3.amazonaws.com/datasets.huggingface.co/trec/test.txt"},
'valid': "https://s3.amazonaws.com/datasets.huggingface.co/trec/test.txt"},
}

DATASETS_LABELS_URL = {
'imdb': {'train': "https://s3.amazonaws.com/datasets.huggingface.co/aclImdb/train.labels.txt",
'test': "https://s3.amazonaws.com/datasets.huggingface.co/aclImdb/test.labels.txt"},
'valid': "https://s3.amazonaws.com/datasets.huggingface.co/aclImdb/valid.labels.txt"},
'trec': {'train': "https://s3.amazonaws.com/datasets.huggingface.co/trec/train.labels.txt",
'test': "https://s3.amazonaws.com/datasets.huggingface.co/trec/test.labels.txt"},
'valid': "https://s3.amazonaws.com/datasets.huggingface.co/trec/test.labels.txt"},
}

DATASETS_LABELS_CONVERSION = {
Expand Down Expand Up @@ -124,7 +124,7 @@ def get_and_tokenize_dataset(tokenizer, dataset_dir='wikitext-103', dataset_cach
if with_labels:
label_conversion_map = DATASETS_LABELS_CONVERSION[dataset_dir]
for split_name in DATASETS_LABELS_URL[dataset_dir]:
dataset_file = cached_path(dataset_map['labels'][split_name])
dataset_file = cached_path(DATASETS_LABELS_URL[dataset_dir][split_name])
with open(dataset_file, "r", encoding="utf-8") as f:
all_lines = f.readlines()
labels[split_name] = [label_conversion_map[line.strip()] for line in tqdm(all_lines)]
Expand Down Expand Up @@ -154,3 +154,10 @@ def encode(obj):
torch.save(encoded_dataset, dataset_cache)

return encoded_dataset


if __name__ == '__main__':
tokenizer = BertTokenizer.from_pretrained('bert-base-cased', do_lower_case=False)
get_and_tokenize_dataset(tokenizer, dataset_dir='imdb', dataset_cache=None, with_labels=True)