Skip to content

Modeling markup lm include tags dic into tokenizer config #30

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

Merged
NielsRogge merged 6 commits into from
Jan 11, 2022
Merged

Modeling markup lm include tags dic into tokenizer config #30

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
c6be98d
replace `microsoft/markuplm-` by `SaulLu/markuplm-`
SaulLu Jan 10, 2022
bc47cb0
remove `tags_dict` from files lists
SaulLu Jan 10, 2022
06a1fae
handle `tag_dict` as a dict object and not as a path
SaulLu Jan 10, 2022
f8a3044
adapt test
SaulLu Jan 10, 2022
b38cdf4
fix test
SaulLu Jan 10, 2022
3a8d067
Revert "replace `microsoft/markuplm-` by `SaulLu/markuplm-`"
SaulLu Jan 11, 2022
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
18 changes: 3 additions & 15 deletions src/transformers/models/markuplm/tokenization_markuplm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,7 +48,6 @@
VOCAB_FILES_NAMES = {
"vocab_file": "vocab.json",
"merges_file": "merges.txt",
"tags_dict": "tags_dict.json",
}

PRETRAINED_VOCAB_FILES_MAP = {
Expand All @@ -60,10 +59,6 @@
"microsoft/markuplm-base": "https://huggingface.co/microsoft/markuplm-base/resolve/main/merges.txt",
"microsoft/markuplm-large": "https://huggingface.co/microsoft/markuplm-large/resolve/main/merges.txt",
},
"tags_dict": {
"microsoft/markuplm-base": "https://huggingface.co/microsoft/markuplm-base/resolve/main/tags_dict.json",
"microsoft/markuplm-large": "https://huggingface.co/microsoft/markuplm-large/resolve/main/tags_dict.json",
},
}


Expand Down Expand Up @@ -261,8 +256,8 @@ def __init__(

with open(vocab_file, encoding="utf-8") as vocab_handle:
self.encoder = json.load(vocab_handle)
with open(tags_dict, encoding="utf-8") as tags_dict_handle:
self.tags_dict = json.load(tags_dict_handle)

self.tags_dict = tags_dict
self.decoder = {v: k for k, v in self.encoder.items()}
self.errors = errors # how to handle errors in decoding
self.byte_encoder = bytes_to_unicode()
Expand Down Expand Up @@ -425,9 +420,6 @@ def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] =
merge_file = os.path.join(
save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["merges_file"]
)
tags_dict_file = os.path.join(
save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["tags_dict"]
)

# save vocab_file
with open(vocab_file, "w", encoding="utf-8") as f:
Expand All @@ -447,11 +439,7 @@ def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] =
writer.write(" ".join(bpe_tokens) + "\n")
index += 1

# save tags_dict_file
with open(tags_dict_file, "w", encoding="utf-8") as f:
f.write(json.dumps(self.tags_dict, ensure_ascii=False))

return vocab_file, merge_file, tags_dict_file
return vocab_file, merge_file

def prepare_for_tokenization(self, text, is_split_into_words=False, **kwargs):
add_prefix_space = kwargs.pop("add_prefix_space", self.add_prefix_space)
Expand Down
15 changes: 3 additions & 12 deletions src/transformers/models/markuplm/tokenization_markuplm_fast.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,7 +53,6 @@
"vocab_file": "vocab.json",
"merges_file": "merges.txt",
"tokenizer_file": "tokenizer.json",
"tags_dict": "tags_dict.json",
}

PRETRAINED_VOCAB_FILES_MAP = {
Expand All @@ -65,10 +64,6 @@
"microsoft/markuplm-base": "https://huggingface.co/microsoft/markuplm-base/resolve/main/merges.txt",
"microsoft/markuplm-large": "https://huggingface.co/microsoft/markuplm-large/resolve/main/merges.txt",
},
"tags_dict": {
"microsoft/markuplm-base": "https://huggingface.co/microsoft/markuplm-base/resolve/main/tags_dict.json",
"microsoft/markuplm-large": "https://huggingface.co/microsoft/markuplm-large/resolve/main/tags_dict.json",
},
}

PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {
Expand Down Expand Up @@ -179,8 +174,8 @@ def __init__(
self.backend_tokenizer.pre_tokenizer = pre_tok_class(**pre_tok_state)

self.add_prefix_space = add_prefix_space
with open(tags_dict, encoding="utf-8") as tags_dict_handle:
self.tags_dict = json.load(tags_dict_handle)

self.tags_dict = tags_dict

tokenizer_component = "post_processor"
tokenizer_component_instance = getattr(self.backend_tokenizer, tokenizer_component, None)
Expand Down Expand Up @@ -727,8 +722,4 @@ def create_token_type_ids_from_sequences(
def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
files = self._tokenizer.model.save(save_directory, name=filename_prefix)

tags_dict_file = os.path.join(
save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["tags_dict"]
)

return tuple(files) + (tags_dict_file,)
return tuple(files)
9 changes: 5 additions & 4 deletions tests/test_tokenization_markuplm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -71,18 +71,19 @@ def setUp(self):
]
vocab_tokens = dict(zip(vocab, range(len(vocab))))
merges = ["#version: 0.2", "\u0120 l", "\u0120l o", "\u0120lo w", "e r", ""]
tags_dict = {"a": 0, "abbr": 1, "acronym": 2, "address": 3}
self.tags_dict = {"a": 0, "abbr": 1, "acronym": 2, "address": 3}
self.special_tokens_map = {"unk_token": "<unk>"}

self.vocab_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["vocab_file"])
self.merges_file = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["merges_file"])
self.tags_dict = os.path.join(self.tmpdirname, VOCAB_FILES_NAMES["tags_dict"])
self.tokenizer_config_file = os.path.join(self.tmpdirname, "tokenizer_config.json")

with open(self.vocab_file, "w", encoding="utf-8") as fp:
fp.write(json.dumps(vocab_tokens) + "\n")
with open(self.merges_file, "w", encoding="utf-8") as fp:
fp.write("\n".join(merges))
with open(self.tags_dict, "w", encoding="utf-8") as fp:
fp.write(json.dumps(tags_dict) + "\n")
with open(self.tokenizer_config_file, "w", encoding="utf-8") as fp:
fp.write(json.dumps({"tags_dict": self.tags_dict}))

# def get_clean_sequence(self, tokenizer):
# html_string = "<html> hello world </html>"
Expand Down