Clone model from vinai/bartpho-syllable

.gitattributes

@@ -1,35 +1,29 @@
 *.7z filter=lfs diff=lfs merge=lfs -text
 *.arrow filter=lfs diff=lfs merge=lfs -text
 *.bin filter=lfs diff=lfs merge=lfs -text
+*.bin.* filter=lfs diff=lfs merge=lfs -text
 *.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
 *.ftz filter=lfs diff=lfs merge=lfs -text
 *.gz filter=lfs diff=lfs merge=lfs -text
 *.h5 filter=lfs diff=lfs merge=lfs -text
 *.joblib filter=lfs diff=lfs merge=lfs -text
 *.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
 *.model filter=lfs diff=lfs merge=lfs -text
 *.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
 *.onnx filter=lfs diff=lfs merge=lfs -text
 *.ot filter=lfs diff=lfs merge=lfs -text
 *.parquet filter=lfs diff=lfs merge=lfs -text
 *.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
 *.pt filter=lfs diff=lfs merge=lfs -text
 *.pth filter=lfs diff=lfs merge=lfs -text
 *.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text 
 *.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
 *.tflite filter=lfs diff=lfs merge=lfs -text
 *.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
 *.xz filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
+*.zstandard filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
+model.safetensors filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,3 +1,21 @@
----
-license: mit
----
+---
+license: mit
+---
+# <a name="introduction"></a> BARTpho: Pre-trained Sequence-to-Sequence Models for Vietnamese
+
+
+Two BARTpho versions `BARTpho-syllable` and `BARTpho-word` are the first public large-scale monolingual sequence-to-sequence models pre-trained for Vietnamese. BARTpho uses the "large" architecture and pre-training scheme of the sequence-to-sequence denoising model [BART](https://github.com/pytorch/fairseq/tree/main/examples/bart), thus especially suitable for generative NLP tasks. Experiments on a downstream task of Vietnamese text summarization show that in both automatic and human evaluations, BARTpho outperforms the strong baseline [mBART](https://github.com/pytorch/fairseq/tree/main/examples/mbart) and improves the state-of-the-art.
+
+The general architecture and experimental results of BARTpho can be found in our [paper](https://arxiv.org/abs/2109.09701):
+
+	@article{bartpho,
+	title     = {{BARTpho: Pre-trained Sequence-to-Sequence Models for Vietnamese}},
+	author    = {Nguyen Luong Tran and Duong Minh Le and Dat Quoc Nguyen},
+	journal   = {arXiv preprint},
+	volume    = {arXiv:2109.09701},
+	year      = {2021}
+	}
+
+**Please CITE** our paper when BARTpho is used to help produce published results or incorporated into other software.
+
+For further information or requests, please go to [BARTpho's homepage](https://github.com/VinAIResearch/BARTpho)!

config.json

@@ -0,0 +1,36 @@
+{
+  "activation_dropout": 0.0,
+  "activation_function": "gelu",
+  "architectures": [
+    "MBartModel"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "scale_embedding": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.10.2",
+  "use_cache": true,
+  "vocab_size": 40030,
+  "tokenizer_class": "BartphoTokenizer"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c3ca2ef8f5b24e0f3c71a8736cf207ac9bec18eeae5df39bac9c339721f96f4
+size 1583480280

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eb59e162b7c19001935efa03e0685ebb53b40c8d951286cff76138a87b6ac5e
+size 1583451917

runs/Apr16_16-20-05_gb2t-WS-C621E-SAGE-Series/events.out.tfevents.1713259206.gb2t-WS-C621E-SAGE-Series.60008.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:429e3a33a41c913b63052cf10c3f6fec444f0f47c2d7e42274f41728740cd43a
+size 5075

runs/Apr16_16-23-23_gb2t-WS-C621E-SAGE-Series/events.out.tfevents.1713259405.gb2t-WS-C621E-SAGE-Series.60748.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:447354c2aded4f2703d77adb4dfe5dc8dc27b9808f094b1c83fcfaf14ec50e6f
+size 10518

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

tf_model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1cc15aeacb446d0f6f02c2da354552a4bde7eec757c61d503e0754fa482209bc
+size 1583820760

tokenization_bartpho.py

@@ -0,0 +1,329 @@
+# coding=utf-8
+# Copyright 2021 VinAI Research and the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License
+""" Tokenization classes for BARTpho-syllable model."""
+
+
+import os
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple
+
+import sentencepiece as spm
+
+from transformers.tokenization_utils import AddedToken, PreTrainedTokenizer
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+SPIECE_UNDERLINE = "▁"
+
+VOCAB_FILES_NAMES = {"vocab_file": "sentencepiece.bpe.model", "monolingual_vocab_file": "dict.txt"}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+    "vocab_file": {
+        "vinai/bartpho-syllable": "https://huggingface.co/vinai/bartpho-syllable/resolve/main/sentencepiece.bpe.model",
+    },
+    "monolingual_vocab_file": {
+        "vinai/bartpho-syllable": "https://huggingface.co/vinai/bartpho-syllable/resolve/main/dict.txt",
+    },
+}
+
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {"vinai/bartpho-syllable": 1024}
+
+
+class BartphoTokenizer(PreTrainedTokenizer):
+    """
+    Adapted from [`XLMRobertaTokenizer`]. Based on [SentencePiece](https://github.com/google/sentencepiece).
+
+    This tokenizer inherits from [`PreTrainedTokenizer`] which contains most of the main methods. Users should refer to
+    this superclass for more information regarding those methods.
+
+    Args:
+        vocab_file (`str`):
+            Path to the vocabulary file. This vocabulary is the pre-trained SentencePiece model available from the
+            multilingual XLM-RoBERTa, also used in mBART, consisting of 250K types.
+        monolingual_vocab_file (`str`):
+            Path to the monolingual vocabulary file. This monolingual vocabulary consists of Vietnamese-specialized
+            types extracted from the multilingual vocabulary vocab_file of 250K types.
+        bos_token (`str`, *optional*, defaults to `"<s>"`):
+            The beginning of sequence token that was used during pretraining. Can be used a sequence classifier token.
+
+            <Tip>
+
+            When building a sequence using special tokens, this is not the token that is used for the beginning of
+            sequence. The token used is the `cls_token`.
+
+            </Tip>
+
+        eos_token (`str`, *optional*, defaults to `"</s>"`):
+            The end of sequence token.
+
+            <Tip>
+
+            When building a sequence using special tokens, this is not the token that is used for the end of sequence.
+            The token used is the `sep_token`.
+
+            </Tip>
+
+        sep_token (`str`, *optional*, defaults to `"</s>"`):
+            The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for
+            sequence classification or for a text and a question for question answering. It is also used as the last
+            token of a sequence built with special tokens.
+        cls_token (`str`, *optional*, defaults to `"<s>"`):
+            The classifier token which is used when doing sequence classification (classification of the whole sequence
+            instead of per-token classification). It is the first token of the sequence when built with special tokens.
+        unk_token (`str`, *optional*, defaults to `"<unk>"`):
+            The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
+            token instead.
+        pad_token (`str`, *optional*, defaults to `"<pad>"`):
+            The token used for padding, for example when batching sequences of different lengths.
+        mask_token (`str`, *optional*, defaults to `"<mask>"`):
+            The token used for masking values. This is the token used when training this model with masked language
+            modeling. This is the token which the model will try to predict.
+        additional_special_tokens (`List[str]`, *optional*, defaults to `["<s>NOTUSED", "</s>NOTUSED"]`):
+            Additional special tokens used by the tokenizer.
+        sp_model_kwargs (`dict`, *optional*):
+            Will be passed to the `SentencePieceProcessor.__init__()` method. The [Python wrapper for
+            SentencePiece](https://github.com/google/sentencepiece/tree/master/python) can be used, among other things,
+            to set:
+
+            - `enable_sampling`: Enable subword regularization.
+            - `nbest_size`: Sampling parameters for unigram. Invalid for BPE-Dropout.
+
+              - `nbest_size = {0,1}`: No sampling is performed.
+              - `nbest_size > 1`: samples from the nbest_size results.
+              - `nbest_size < 0`: assuming that nbest_size is infinite and samples from the all hypothesis (lattice)
+                using forward-filtering-and-backward-sampling algorithm.
+
+            - `alpha`: Smoothing parameter for unigram sampling, and dropout probability of merge operations for
+              BPE-dropout.
+
+    Attributes:
+        sp_model (`SentencePieceProcessor`):
+            The *SentencePiece* processor that is used for every conversion (string, tokens and IDs).
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+    model_input_names = ["input_ids", "attention_mask"]
+
+    def __init__(
+        self,
+        vocab_file,
+        monolingual_vocab_file,
+        bos_token="<s>",
+        eos_token="</s>",
+        sep_token="</s>",
+        cls_token="<s>",
+        unk_token="<unk>",
+        pad_token="<pad>",
+        mask_token="<mask>",
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        **kwargs
+    ) -> None:
+        # Mask token behave like a normal word, i.e. include the space before it
+        mask_token = AddedToken(mask_token, lstrip=True, rstrip=False) if isinstance(mask_token, str) else mask_token
+
+        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+
+        super().__init__(
+            bos_token=bos_token,
+            eos_token=eos_token,
+            unk_token=unk_token,
+            sep_token=sep_token,
+            cls_token=cls_token,
+            pad_token=pad_token,
+            mask_token=mask_token,
+            sp_model_kwargs=self.sp_model_kwargs,
+            **kwargs,
+        )
+
+        self.vocab_file = vocab_file
+        self.monolingual_vocab_file = monolingual_vocab_file
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(str(vocab_file))
+
+        # Load the reduced vocab
+
+        # Keep order of special tokens for backward compatibility
+        self.fairseq_tokens_to_ids = {}
+        cnt = 0
+        for token in [bos_token, pad_token, eos_token, unk_token, sep_token, cls_token]:
+            if str(token) not in self.fairseq_tokens_to_ids:
+                self.fairseq_tokens_to_ids[str(token)] = cnt
+                cnt += 1
+        with open(monolingual_vocab_file, "r", encoding="utf-8") as f:
+            for line in f.readlines():
+                token = line.strip().split()[0]
+                self.fairseq_tokens_to_ids[token] = len(self.fairseq_tokens_to_ids)
+        if str(mask_token) not in self.fairseq_tokens_to_ids:
+            self.fairseq_tokens_to_ids[str(mask_token)] = len(self.fairseq_tokens_to_ids)
+
+        self.fairseq_ids_to_tokens = {v: k for k, v in self.fairseq_tokens_to_ids.items()}
+
+    def __getstate__(self):
+        state = self.__dict__.copy()
+        state["sp_model"] = None
+        state["sp_model_proto"] = self.sp_model.serialized_model_proto()
+        return state
+
+    def __setstate__(self, d):
+        self.__dict__ = d
+
+        # for backward compatibility
+        if not hasattr(self, "sp_model_kwargs"):
+            self.sp_model_kwargs = {}
+
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.LoadFromSerializedProto(self.sp_model_proto)
+
+    def build_inputs_with_special_tokens(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
+        adding special tokens. An BARTPho sequence has the following format:
+
+        - single sequence: `<s> X </s>`
+        - pair of sequences: `<s> A </s></s> B </s>`
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs to which the special tokens will be added.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of [input IDs](../glossary#input-ids) with the appropriate special tokens.
+        """
+
+        if token_ids_1 is None:
+            return [self.cls_token_id] + token_ids_0 + [self.sep_token_id]
+        cls = [self.cls_token_id]
+        sep = [self.sep_token_id]
+        return cls + token_ids_0 + sep + sep + token_ids_1 + sep
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer `prepare_for_model` method.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        if token_ids_1 is None:
+            return [1] + ([0] * len(token_ids_0)) + [1]
+        return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Create a mask from the two sequences passed to be used in a sequence-pair classification task. BARTPho does not
+        make use of token type ids, therefore a list of zeros is returned.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of zeros.
+
+        """
+
+        sep = [self.sep_token_id]
+        cls = [self.cls_token_id]
+
+        if token_ids_1 is None:
+            return len(cls + token_ids_0 + sep) * [0]
+        return len(cls + token_ids_0 + sep + sep + token_ids_1 + sep) * [0]
+
+    @property
+    def vocab_size(self):
+        return len(self.fairseq_ids_to_tokens)
+
+    def get_vocab(self):
+        vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def _tokenize(self, text: str) -> List[str]:
+        return self.sp_model.encode(text, out_type=str)
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        if token in self.fairseq_tokens_to_ids:
+            return self.fairseq_tokens_to_ids[token]
+        else:
+            return self.unk_token_id
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        return self.fairseq_ids_to_tokens[index]
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (strings for sub-words) in a single string."""
+        out_string = "".join(tokens).replace(SPIECE_UNDERLINE, " ").strip()
+        return out_string
+
+    def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+            return
+        out_vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+        out_monolingual_vocab_file = os.path.join(
+            save_directory,
+            (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["monolingual_vocab_file"],
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+        elif not os.path.isfile(self.vocab_file):
+            with open(out_vocab_file, "wb") as fi:
+                content_spiece_model = self.sp_model.serialized_model_proto()
+                fi.write(content_spiece_model)
+
+        if os.path.abspath(self.monolingual_vocab_file) != os.path.abspath(
+            out_monolingual_vocab_file
+        ) and os.path.isfile(self.monolingual_vocab_file):
+            copyfile(self.monolingual_vocab_file, out_monolingual_vocab_file)
+        elif not os.path.isfile(self.monolingual_vocab_file):
+            with open(out_monolingual_vocab_file, "w", encoding="utf-8") as fp:
+                for token in self.fairseq_tokens_to_ids:
+                    if token not in self.all_special_tokens:
+                        fp.write(f"{str(token)} \n")
+
+        return out_vocab_file, out_monolingual_vocab_file

tokenization_bartpho_fast.py

@@ -0,0 +1,334 @@
+# coding=utf-8
+# Copyright 2021 VinAI Research and the HuggingFace Inc. team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License
+""" Tokenization classes for BARTpho-syllable model."""
+
+import os
+from collections import defaultdict
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple, Union
+
+from transformers.tokenization_utils import AddedToken
+from transformers.tokenization_utils_base import EncodingFast
+from transformers.tokenization_utils_fast import PreTrainedTokenizerFast
+from transformers.utils import is_sentencepiece_available, logging
+
+
+if is_sentencepiece_available():
+    from .tokenization_bartpho import BartphoTokenizer
+else:
+    BartphoTokenizer = None
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {
+    "vocab_file": "sentencepiece.bpe.model",
+    "monolingual_vocab_file": "dict.txt",
+    "tokenizer_file": "tokenizer.json",
+}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+    "vocab_file": {
+        "vinai/bartpho-syllable": "https://huggingface.co/vinai/bartpho-syllable/resolve/main/sentencepiece.bpe.model",
+    },
+    "monolingual_vocab_file": {
+        "vinai/bartpho-syllable": "https://huggingface.co/vinai/bartpho-syllable/resolve/main/dict.txt",
+    },
+    "tokenizer_file": {
+        "vinai/bartpho-syllable": "https://huggingface.co/vinai/bartpho-syllable/resolve/main/tokenizer.json",
+    },
+}
+
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {"vinai/bartpho-syllable": 1024}
+
+
+class BartphoTokenizerFast(PreTrainedTokenizerFast):
+    """
+    Construct a "fast" BARTpho tokenizer (backed by HuggingFace's *tokenizers* library). Adapted from
+    [`XLMRobertaTokenizerFast`]. Based on [SentencePiece](https://github.com/google/sentencepiece).
+
+    This tokenizer inherits from [`PreTrainedTokenizerFast`] which contains most of the main methods. Users should
+    refer to this superclass for more information regarding those methods.
+
+    Args:
+        vocab_file (`str`):
+            Path to the vocabulary file.
+        bos_token (`str`, *optional*, defaults to `"<s>"`):
+            The beginning of sequence token that was used during pretraining. Can be used a sequence classifier token.
+
+            <Tip>
+
+            When building a sequence using special tokens, this is not the token that is used for the beginning of
+            sequence. The token used is the `cls_token`.
+
+            </Tip>
+
+        eos_token (`str`, *optional*, defaults to `"</s>"`):
+            The end of sequence token.
+
+            <Tip>
+
+            When building a sequence using special tokens, this is not the token that is used for the end of sequence.
+            The token used is the `sep_token`.
+
+            </Tip>
+
+        sep_token (`str`, *optional*, defaults to `"</s>"`):
+            The separator token, which is used when building a sequence from multiple sequences, e.g. two sequences for
+            sequence classification or for a text and a question for question answering. It is also used as the last
+            token of a sequence built with special tokens.
+        cls_token (`str`, *optional*, defaults to `"<s>"`):
+            The classifier token which is used when doing sequence classification (classification of the whole sequence
+            instead of per-token classification). It is the first token of the sequence when built with special tokens.
+        unk_token (`str`, *optional*, defaults to `"<unk>"`):
+            The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
+            token instead.
+        pad_token (`str`, *optional*, defaults to `"<pad>"`):
+            The token used for padding, for example when batching sequences of different lengths.
+        mask_token (`str`, *optional*, defaults to `"<mask>"`):
+            The token used for masking values. This is the token used when training this model with masked language
+            modeling. This is the token which the model will try to predict.
+        additional_special_tokens (`List[str]`, *optional*, defaults to `["<s>NOTUSED", "</s>NOTUSED"]`):
+            Additional special tokens used by the tokenizer.
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+    model_input_names = ["input_ids", "attention_mask"]
+    slow_tokenizer_class = BartphoTokenizer
+
+    def __init__(
+        self,
+        vocab_file=None,
+        monolingual_vocab_file=None,
+        tokenizer_file=None,
+        bos_token="<s>",
+        eos_token="</s>",
+        sep_token="</s>",
+        cls_token="<s>",
+        unk_token="<unk>",
+        pad_token="<pad>",
+        mask_token="<mask>",
+        **kwargs
+    ):
+        # Mask token behave like a normal word, i.e. include the space before it
+        mask_token = AddedToken(mask_token, lstrip=True, rstrip=False) if isinstance(mask_token, str) else mask_token
+
+        super().__init__(
+            vocab_file,
+            monolingual_vocab_file,
+            tokenizer_file=tokenizer_file,
+            bos_token=bos_token,
+            eos_token=eos_token,
+            sep_token=sep_token,
+            cls_token=cls_token,
+            unk_token=unk_token,
+            pad_token=pad_token,
+            mask_token=mask_token,
+            **kwargs,
+        )
+
+        self.vocab_file = vocab_file
+        self.monolingual_vocab_file = monolingual_vocab_file
+        self.can_save_slow_tokenizer = False if not self.vocab_file else True
+
+    def get_added_vocab_hacking(self):
+        """
+        Returns the added tokens in the vocabulary as a dictionary of token to index.
+
+        Returns:
+            `Dict[str, int], Dict[int, int]`: The added tokens, and their original and new ids
+        """
+        base_vocab_size = self._tokenizer.get_vocab_size(with_added_tokens=False)
+        full_vocab_size = self._tokenizer.get_vocab_size(with_added_tokens=True)
+        if full_vocab_size == base_vocab_size:
+            return {}, {}
+
+        # Tokens in added_vocab should have ids that are equal to or larger than the size of base_vocab
+        added_vocab = dict(
+            (self._tokenizer.id_to_token(index), index + 1 - base_vocab_size + self.mask_token_id)
+            for index in range(base_vocab_size, full_vocab_size)
+        )
+
+        id_mapping = dict((index, self._tokenizer.token_to_id(tok)) for tok, index in added_vocab.items())
+
+        return added_vocab, id_mapping
+
+    def _decode(
+        self,
+        token_ids: Union[int, List[int]],
+        skip_special_tokens: bool = False,
+        clean_up_tokenization_spaces: bool = True,
+        **kwargs
+    ) -> str:
+        self._decode_use_source_tokenizer = kwargs.pop("use_source_tokenizer", False)
+
+        if isinstance(token_ids, int):
+            token_ids = [token_ids]
+
+        # Mapping ids into their original values
+        _, id_mapping = self.get_added_vocab_hacking()
+        if len(id_mapping) > 0:
+            token_ids = [id_mapping[id] if id in id_mapping else id for id in token_ids]
+
+        text = self._tokenizer.decode(token_ids, skip_special_tokens=skip_special_tokens)
+
+        if clean_up_tokenization_spaces:
+            clean_text = self.clean_up_tokenization(text)
+            return clean_text
+        else:
+            return text
+
+    def _convert_encoding(
+        self,
+        encoding: EncodingFast,
+        return_token_type_ids: Optional[bool] = None,
+        return_attention_mask: Optional[bool] = None,
+        return_overflowing_tokens: bool = False,
+        return_special_tokens_mask: bool = False,
+        return_offsets_mapping: bool = False,
+        return_length: bool = False,
+        verbose: bool = True,
+    ) -> Tuple[Dict[str, Any], List[EncodingFast]]:
+        """
+        Convert the encoding representation (from low-level HuggingFace tokenizer output) to a python Dict and a list
+        of encodings, take care of building a batch from overflowing tokens.
+
+        Overflowing tokens are converted to additional examples (like batches) so the output values of the dict are
+        lists (overflows) of lists (tokens).
+
+        Output shape: (overflows, sequence length)
+        """
+        if return_token_type_ids is None:
+            return_token_type_ids = "token_type_ids" in self.model_input_names
+        if return_attention_mask is None:
+            return_attention_mask = "attention_mask" in self.model_input_names
+
+        if return_overflowing_tokens and encoding.overflowing is not None:
+            encodings = [encoding] + encoding.overflowing
+        else:
+            encodings = [encoding]
+
+        encoding_dict = defaultdict(list)
+        added_vocab, _ = self.get_added_vocab_hacking()
+        for e in encodings:
+            # encoding_dict["input_ids"].append(e.ids)
+            # Reassign ids of tokens due to the hacking strategy
+            ids = []
+            for id, token in zip(e.ids, e.tokens):
+                if id <= self.mask_token_id:
+                    ids.append(id)
+                else:
+                    if token.strip() in added_vocab:
+                        ids.append(added_vocab[token.strip()])
+                    else:
+                        ids.append(self.unk_token_id)
+
+            encoding_dict["input_ids"].append(ids)
+
+            if return_token_type_ids:
+                encoding_dict["token_type_ids"].append(e.type_ids)
+            if return_attention_mask:
+                encoding_dict["attention_mask"].append(e.attention_mask)
+            if return_special_tokens_mask:
+                encoding_dict["special_tokens_mask"].append(e.special_tokens_mask)
+            if return_offsets_mapping:
+                encoding_dict["offset_mapping"].append(e.offsets)
+            if return_length:
+                # encoding_dict["length"].append(len(e.ids))
+                encoding_dict["length"].append(len(ids))
+
+        return encoding_dict, encodings
+
+    def build_inputs_with_special_tokens(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
+        adding special tokens. A BARTpho sequence has the following format:
+
+        - single sequence: `<s> X </s>`
+        - pair of sequences: `<s> A </s></s> B </s>`
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs to which the special tokens will be added.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of [input IDs](../glossary#input-ids) with the appropriate special tokens.
+        """
+
+        if token_ids_1 is None:
+            return [self.cls_token_id] + token_ids_0 + [self.sep_token_id]
+        cls = [self.cls_token_id]
+        sep = [self.sep_token_id]
+        return cls + token_ids_0 + sep + sep + token_ids_1 + sep
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Create a mask from the two sequences passed to be used in a sequence-pair classification task. BARTpho does not
+        make use of token type ids, therefore a list of zeros is returned.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of zeros.
+
+        """
+
+        sep = [self.sep_token_id]
+        cls = [self.cls_token_id]
+
+        if token_ids_1 is None:
+            return len(cls + token_ids_0 + sep) * [0]
+        return len(cls + token_ids_0 + sep + sep + token_ids_1 + sep) * [0]
+
+    def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        if not self.can_save_slow_tokenizer:
+            raise ValueError(
+                "Your fast tokenizer does not have the necessary information to save the vocabulary for a "
+                "slow tokenizer."
+            )
+
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory.")
+            return
+
+        out_vocab_file = os.path.join(
+            save_directory,
+            (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"],
+        )
+
+        out_monolingual_vocab_file = os.path.join(
+            save_directory,
+            (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["monolingual_vocab_file"],
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+
+        if os.path.abspath(self.monolingual_vocab_file) != os.path.abspath(out_monolingual_vocab_file):
+            copyfile(self.monolingual_vocab_file, out_monolingual_vocab_file)
+
+        return (out_vocab_file, out_monolingual_vocab_file)

tokenizer_config.json

@@ -0,0 +1,55 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "40029": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "model_max_length": 1024,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "BartphoTokenizer",
+  "unk_token": "<unk>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d185703bf08499ca9c2f9c3612d9f7ec11b7666a7cdb13773aac6c1b1e3d400
+size 5048