说明文档
模型概述
这是一个 xlm-roberta 模型,经过微调后可恢复47种语言的标点符号、大小写正确转换(首字母大写)以及检测句子边界(句号)。
使用方法
如果您只是想试用该模型本页面的组件(widget)就足够了。若要离线使用模型,以下代码片段展示了如何使用该模型,包括使用包装器(这是我编写的,可从 PyPI 获取)和手动使用(使用本仓库中的 ONNX 和 SentencePiece 模型)。
通过 punctuators 包使用
<details>
<summary>点击查看使用包装器的方法</summary>
使用此模型最简单的方式是安装 punctuators:
$ pip install punctuators
但这只是一个 ONNX 和 SentencePiece 模型,因此您可以随意运行它。
punctuators API 的输入是一个字符串列表(批次)。每个字符串都将被添加标点符号、进行大小写正确转换,并根据预测的句号进行分句。因此输出是一个字符串列表的列表:每个输入文本对应一个分句后的字符串列表。
若要禁用句号分割,请使用 m.infer(texts, apply_sbd=False)。输出将是一个字符串列表:每个输入文本对应一个添加了标点符号并进行了大小写正确转换的字符串。
<details open>
<summary>使用示例</summary>
from typing import List
from punctuators.models import PunctCapSegModelONNX
m: PunctCapSegModelONNX = PunctCapSegModelONNX.from_pretrained(
"1-800-BAD-CODE/xlm-roberta_punctuation_fullstop_truecase"
)
input_texts: List[str] = [
"hola mundo cómo estás estamos bajo el sol y hace mucho calor santa coloma abre los huertos urbanos a las escuelas de la ciudad",
"hello friend how's it going it's snowing outside right now in connecticut a large storm is moving in",
"未來疫苗將有望覆蓋3歲以上全年齡段美國與北約軍隊已全部撤離還有鐵路公路在內的各項基建的來源都將枯竭",
"በባለፈው ሳምንት ኢትዮጵያ ከሶማሊያ 3 ሺህ ወታደሮቿንም እንዳስወጣች የሶማሊያው ዳልሳን ሬድዮ ዘግቦ ነበር ጸጥታ ሃይሉና ህዝቡ ተቀናጅቶ በመስራቱ በመዲናዋ ላይ የታቀደው የጥፋት ሴራ ከሽፏል",
"こんにちは友人" "調子はどう" "今日は雨の日でしたか" "乾いた状態を保つために一日中室内で過ごしました",
"hallo freund wie geht's es war heute ein regnerischer tag nicht wahr ich verbrachte den tag drinnen um trocken zu bleiben",
"हैलो दोस्त ये कैसा चल रहा है आज बारिश का दिन था न मैंने सूखा रहने के लिए दिन घर के अंदर बिताया",
"كيف تجري الامور كان يومًا ممطرًا اليوم أليس كذلك قضيت اليوم في الداخل لأظل جافًا",
]
results: List[List[str]] = m.infer(
texts=input_texts, apply_sbd=True,
)
for input_text, output_texts in zip(input_texts, results):
print(f"Input: {input_text}")
print(f"Outputs:")
for text in output_texts:
print(f"\t{text}")
print()
</details>
<details open>
<summary>预期输出</summary>
Input: hola mundo cómo estás estamos bajo el sol y hace mucho calor santa coloma abre los huertos urbanos a las escuelas de la ciudad
Outputs:
Hola mundo, ¿cómo estás?
Estamos bajo el sol y hace mucho calor.
Santa Coloma abre los huertos urbanos a las escuelas de la ciudad.
Input: hello friend how's it going it's snowing outside right now in connecticut a large storm is moving in
Outputs:
Hello friend, how's it going?
It's snowing outside right now.
In Connecticut, a large storm is moving in.
Input: 未來疫苗將有望覆蓋3歲以上全年齡段美國與北約軍隊已全部撤離還有鐵路公路在內的各項基建的來源都將枯竭
Outputs:
未來,疫苗將有望覆蓋3歲以上全年齡段。
美國與北約軍隊已全部撤離。
還有,鐵路,公路在內的各項基建的來源都將枯竭。
Input: በባለፈው ሳምንት ኢትዮጵያ ከሶማሊያ 3 ሺህ ወታደሮቿንም እንዳስወጣች የሶማሊያው ዳልሳን ሬድዮ ዘግቦ ነበር ጸጥታ ሃይሉና ህዝቡ ተቀናጅቶ በመስራቱ በመዲናዋ ላይ የታቀደው የጥፋት ሴራ ከሽፏል
Outputs:
በባለፈው ሳምንት ኢትዮጵያ ከሶማሊያ 3 ሺህ ወታደሮቿንም እንዳስወጣች የሶማሊያው ዳልሳን ሬድዮ ዘግቦ ነበር።
ጸጥታ ሃይሉና ህዝቡ ተቀናጅቶ በመስራቱ በመዲናዋ ላይ የታቀደው የጥፋት ሴራ ከሽፏል።
Input: こんにちは友人調子はどう今日は雨の日でしたか乾いた状態を保つために一日中室内で過ごしました
Outputs:
こんにちは、友人、調子はどう?
今日は雨の日でしたか。
乾いた状態を保つために、一日中、室内で過ごしました。
Input: hallo freund wie geht's es war heute ein regnerischer tag nicht wahr ich verbrachte den tag drinnen um trocken zu bleiben
Outputs:
Hallo Freund, wie geht's?
Es war heute ein regnerischer Tag, nicht wahr?
Ich verbrachte den Tag drinnen, um trocken zu bleiben.
Input: हैलो दोस्त ये कैसा चल रहा है आज बारिश का दिन था न मैंने सूखा रहने के लिए दिन घर के अंदर बिताया
Outputs:
हैलो दोस्त, ये कैसा चल रहा है?
आज बारिश का दिन था न, मैंने सूखा रहने के लिए दिन घर के अंदर बिताया।
Input: كيف تجري الامور كان يومًا ممطرًا اليوم أليس كذلك قضيت اليوم في الداخل لأظل جافًا
Outputs:
كيف تجري الامور؟
كان يومًا ممطرًا اليوم، أليس كذلك؟
قضيت اليوم في الداخل لأظل جافًا.
</details>
</details>
手动使用方法
如果您想不使用包装器直接使用 ONNX 和 SP 模型,请参阅以下示例。
<details>
<summary>点击查看手动使用方法</summary>
from typing import List
import numpy as np
import onnxruntime as ort
from huggingface_hub import hf_hub_download
from omegaconf import OmegaConf
from sentencepiece import SentencePieceProcessor
# Download the models from HF hub. Note: to clean up, you can find these files in your HF cache directory
spe_path = hf_hub_download(repo_id="1-800-BAD-CODE/xlm-roberta_punctuation_fullstop_truecase", filename="sp.model")
onnx_path = hf_hub_download(repo_id="1-800-BAD-CODE/xlm-roberta_punctuation_fullstop_truecase", filename="model.onnx")
config_path = hf_hub_download(
repo_id="1-800-BAD-CODE/xlm-roberta_punctuation_fullstop_truecase", filename="config.yaml"
)
# Load the SP model
tokenizer: SentencePieceProcessor = SentencePieceProcessor(spe_path) # noqa
# Load the ONNX graph
ort_session: ort.InferenceSession = ort.InferenceSession(onnx_path)
# Load the model config with labels, etc.
config = OmegaConf.load(config_path)
# Potential classification labels before each subtoken
pre_labels: List[str] = config.pre_labels
# Potential classification labels after each subtoken
post_labels: List[str] = config.post_labels
# Special class that means "predict nothing"
null_token = config.get("null_token", "<NULL>")
# Special class that means "all chars in this subtoken end with a period", e.g., "am" -> "a.m."
acronym_token = config.get("acronym_token", "<ACRONYM>")
# Not used in this example, but if your sequence exceed this value, you need to fold it over multiple inputs
max_len = config.max_length
# For reference only, graph has no language-specific behavior
languages: List[str] = config.languages
# Encode some input text, adding BOS + EOS
input_text = "hola mundo cómo estás estamos bajo el sol y hace mucho calor santa coloma abre los huertos urbanos a las escuelas de la ciudad"
input_ids = [tokenizer.bos_id()] + tokenizer.EncodeAsIds(input_text) + [tokenizer.eos_id()]
# Create a numpy array with shape [B, T], as the graph expects as input.
# Note that we do not pass lengths to the graph; if you are using a batch, padding should be tokenizer.pad_id() and the
# graph's attention mechanisms will ignore pad_id() without requiring explicit sequence lengths.
input_ids_arr: np.array = np.array([input_ids])
# Run the graph, get outputs for all analytics
pre_preds, post_preds, cap_preds, sbd_preds = ort_session.run(None, {"input_ids": input_ids_arr})
# Squeeze off the batch dimensions and convert to lists
pre_preds = pre_preds[0].tolist()
post_preds = post_preds[0].tolist()
cap_preds = cap_preds[0].tolist()
sbd_preds = sbd_preds[0].tolist()
# Segmented sentences
output_texts: List[str] = []
# Current sentence, which is built until we hit a sentence boundary prediction
current_chars: List[str] = []
# Iterate over the outputs, ignoring the first (BOS) and final (EOS) predictions and tokens
for token_idx in range(1, len(input_ids) - 1):
token = tokenizer.IdToPiece(input_ids[token_idx])
# Simple SP decoding
if token.startswith("▁") and current_chars:
current_chars.append(" ")
# Token-level predictions
pre_label = pre_labels[pre_preds[token_idx]]
post_label = post_labels[post_preds[token_idx]]
# If we predict "pre-punct", insert it before this token
if pre_label != null_token:
current_chars.append(pre_label)
# Iterate over each char. Skip SP's space token,
char_start = 1 if token.startswith("▁") else 0
for token_char_idx, char in enumerate(token[char_start:], start=char_start):
# If this char should be capitalized, apply upper case
if cap_preds[token_idx][token_char_idx]:
char = char.upper()
# Append char
current_chars.append(char)
# if this is an acronym, add a period after every char (p.m., a.m., etc.)
if post_label == acronym_token:
current_chars.append(".")
# Maybe this subtoken ends with punctuation
if post_label != null_token and post_label != acronym_token:
current_chars.append(post_label)
# If this token is a sentence boundary, finalize the current sentence and reset
if sbd_preds[token_idx]:
output_texts.append("".join(current_chars))
current_chars.clear()
# Maybe push final sentence, if the final token was not classified as a sentence boundary
if current_chars:
output_texts.append("".join(current_chars))
# Pretty print
print(f"Input: {input_text}")
print("Outputs:")
for text in output_texts:
print(f"\t{text}")
预期输出:
Input: hola mundo cómo estás estamos bajo el sol y hace mucho calor santa coloma abre los huertos urbanos a las escuelas de la ciudad
Outputs:
Hola mundo, ¿cómo estás?
Estamos bajo el sol y hace mucho calor.
Santa Coloma abre los huertos urbanos a las escuelas de la ciudad.
</details>
模型架构
该模型实现了以下架构,允许在每种语言中进行标点符号、大小写正确转换和句号预测,而无需特定语言的行为:

<details>
<summary>点击查看架构说明</summary>
我们首先对文本进行分词,并使用 XLM-Roberta 进行编码,这是该架构的预训练部分。
然后,我们预测每个子词(subtoken)前后的标点符号。 在每个词符(token)之前进行预测可以实现西班牙语的反向问号。 在每个词符之后进行预测可以处理所有其他标点符号,包括连续脚本语言中的标点符号和缩写词。
我们使用嵌入来表示预测的标点符号词符,以告知句子边界检测头(head)将在文本中插入哪些标点符号。这允许正确的句号预测,因为某些标点符号(句号、问号等)与句子边界密切相关。
然后,我们将句号预测向右移动一个位置,以告知大小写正确转换头每个新句子的开头在哪里。这很重要,因为大小写正确转换与句子边界密切相关。
对于大小写正确转换,我们对每个子词进行 N 次预测,其中 N 是该子词中的字符数。实际上,N 是最大子词长度,多余的预测会被忽略。从本质上讲,大小写正确转换被建模为一个多标签问题。这允许对任意字符进行大写,例如:"NATO"、"MacDonald"、"mRNA" 等。
将所有这些预测应用到输入文本,我们可以在任何语言中为文本添加标点符号、进行大小写正确转换和分句。
</details>
分词器
<details>
<summary>点击查看 XLM-Roberta 分词器是如何被修复的</summary>
与 FairSeq 使用的复杂包装器不同(该包装器被奇怪地移植到 HuggingFace,但从未被修复),xlm-roberta 的 SentencePiece 模型经过调整后可以正确编码文本。根据 HF 的注释,
# Original fairseq vocab and spm vocab must be "aligned":
# Vocab | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9
# -------- | ------- | ------- | ------ | ------- | --- | --- | --- | ----- | ----- | ----
# fairseq | '<s>' | '<pad>' | '</s>' | '<unk>' | ',' | '.' | '▁' | 's' | '▁de' | '-'
# spm | '<unk>' | '<s>' | '</s>' | ',' | '.' | '▁' | 's' | '▁de' | '-' | '▁a'
SP 模型使用以下代码进行了修复(SentencePiece 专家们,如果这里有问题请告诉我):
from sentencepiece import SentencePieceProcessor
from sentencepiece.sentencepiece_model_pb2 import ModelProto
m = ModelProto()
m.ParseFromString(open("/path/to/xlmroberta/sentencepiece.bpe.model", "rb").read())
pieces = list(m.pieces)
pieces = (
[
ModelProto.SentencePiece(piece="<s>", type=ModelProto.SentencePiece.Type.CONTROL),
ModelProto.SentencePiece(piece="<pad>", type=ModelProto.SentencePiece.Type.CONTROL),
ModelProto.SentencePiece(piece="</s>", type=ModelProto.SentencePiece.Type.CONTROL),
ModelProto.SentencePiece(piece="<unk>", type=ModelProto.SentencePiece.Type.UNKNOWN),
]
+ pieces[3:]
+ [ModelProto.SentencePiece(piece="<mask>", type=ModelProto.SentencePiece.Type.USER_DEFINED)]
)
del m.pieces[:]
m.pieces.extend(pieces)
with open("/path/to/new/sp.model", "wb") as f:
f.write(m.SerializeToString())
现在我们可以只使用 SP 模型而无需包装器。
</details>
后标点符号词符
该模型预测每个子词后的以下标点符号词符:
| 词符 | 描述 | 相关语言 |
|---|---|---|
| <NULL> | 无标点 | 全部 |
| <ACRONYM> | 该子词中的每个字符后跟句号 | 主要为英语,部分欧洲语言 |
| . | 拉丁句号 | 多种语言 |
| , | 拉丁逗号 | 多种语言 |
| ? | 拉丁问号 | 多种语言 |
| ? | 全角问号 | 中文、日文 |
| , | 全角逗号 | 中文、日文 |
| 。 | 全角句号 | 中文、日文 |
| 、 | 顿号 | 中文、日文 |
| ・ | 中点 | 日文 |
| । | 逗点 | 印地语、孟加拉语、奥里亚语 |
| ؟ | 阿拉伯问号 | 阿拉伯语 |
| ; | 希腊问号 | 希腊语 |
| ። | 埃塞俄比亚句号 | 阿姆哈拉语 |
| ፣ | 埃塞俄比亚逗号 | 阿姆哈拉语 |
| ፧ | 埃塞俄比亚问号 | 阿姆哈拉语 |
前标点符号词符
该模型预测每个子词前的以下标点符号词符:
| 词符 | 描述 | 相关语言 |
|---|---|---|
| <NULL> | 无标点 | 全部 |
| ¿ | 西班牙语反向问号 | 西班牙语 |
训练细节
该模型在 NeMo 框架中训练,使用 A100 大约 7 小时。
您可以在 tensorboard.dev 上查看 tensorboard 日志。
该模型使用 WMT 的 News Crawl 数据进行训练。 每种语言使用 100 万行文本,除了少数低资源语言可能使用较少。 语言的选择基于 News Crawl 语料库是否包含足够的高质量数据(由作者判断)。
局限性
该模型基于新闻数据进行训练,在对话或非正式数据上可能表现不佳。
该模型不太可能达到生产质量。 每种语言仅使用"100万"行进行训练,开发集可能由于网络抓取新闻数据的特性而存在噪声。
该模型过度预测西班牙语问号,尤其是反向问号 ¿(见下方指标)。由于 ¿ 是一个稀有词符,特别是在47语言模型的背景下,西班牙语疑问句被过度采样,即从未使用的额外训练数据中选择了更多这些句子。然而,这似乎"过度纠正"了问题,导致预测了过多的西班牙语问号。
该模型可能还过度预测逗号。
如果您发现此处未提及的任何一般性局限性,请告诉我,以便在下一次微调中解决所有局限性。
评估
在这些指标中,请记住:
-
数据有噪声
-
句子边界和大小写正确转换是基于预测的标点符号进行条件判断的,这是最困难的任务,有时会不正确。 当使用参考标点符号进行条件判断时,大小写正确转换和 SBD(句子边界检测)实际上在大多数语言中达到近 100%。
-
标点符号可能是主观的。例如:
Hola mundo, ¿cómo estás?或者
Hola mundo. ¿Cómo estás?当句子更长更实际,这些歧义会大量存在,并影响所有三个指标。
测试数据与示例生成
每个测试示例使用以下程序生成:
- 连接 11 个随机句子(测试集中每个句子对应 1 + 10 个句子)
- 将连接的句子转为小写
- 移除所有标点符号
目标是在我们小写字母和移除标点符号时生成。
数据是 News Crawl 的保留部分,已去重。每种语言使用 3000 行数据,生成 3000 个唯一的 11 句子示例。
我们生成 3000 个示例,其中示例 i 以句子 i 开头,后跟从 3000 句子测试集中随机选择的 10 个句子。
为了测量大小写正确转换和句子边界检测,使用参考标点符号词符进行条件判断(见上方架构图)。如果我们改用预测的标点符号,那么不正确的标点符号将导致大小写正确转换和 SBD 目标不正确对齐,这些指标将被人为降低。
选定的语言评估报告
目前,下方显示了几种选定语言的指标。鉴于收集和整理 47 种语言指标所需的工作量,我会最终添加更多。
展开以下任意选项卡可查看该语言的指标。
<details> <summary>英语</summary>
punct_post test report:
label precision recall f1 support
<NULL> (label_id: 0) 99.25 98.43 98.84 564908
<ACRONYM> (label_id: 1) 63.14 84.67 72.33 613
. (label_id: 2) 90.97 93.91 92.42 32040
, (label_id: 3) 73.95 84.32 78.79 24271
? (label_id: 4) 79.05 81.94 80.47 1041
? (label_id: 5) 0.00 0.00 0.00 0
, (label_id: 6) 0.00 0.00 0.00 0
。 (label_id: 7) 0.00 0.00 0.00 0
、 (label_id: 8) 0.00 0.00 0.00 0
・ (label_id: 9) 0.00 0.00 0.00 0
। (label_id: 10) 0.00 0.00 0.00 0
؟ (label_id: 11) 0.00 0.00 0.00 0
، (label_id: 12) 0.00 0.00 0.00 0
; (label_id: 13) 0.00 0.00 0.00 0
። (label_id: 14) 0.00 0.00 0.00 0
፣ (label_id: 15) 0.00 0.00 0.00 0
፧ (label_id: 16) 0.00 0.00 0.00 0
-------------------
micro avg 97.60 97.60 97.60 622873
macro avg 81.27 88.65 84.57 622873
weighted avg 97.77 97.60 97.67 622873
cap test report:
label precision recall f1 support
LOWER (label_id: 0) 99.72 99.85 99.78 2134956
UPPER (label_id: 1) 96.33 93.52 94.91 91996
-------------------
micro avg 99.59 99.59 99.59 2226952
macro avg 98.03 96.68 97.34 2226952
weighted avg 99.58 99.59 99.58 2226952
seg test report:
label precision recall f1 support
NOSTOP (label_id: 0) 99.99 99.98 99.99 591540
FULLSTOP (label_id: 1) 99.61 99.89 99.75 34333
-------------------
micro avg 99.97 99.97 99.97 625873
macro avg 99.80 99.93 99.87 625873
weighted avg 99.97 99.97 99.97 625873
</details>
<details> <summary>西班牙语</summary>
punct_pre test report:
label precision recall f1 support
<NULL> (label_id: 0) 99.94 99.89 99.92 636941
¿ (label_id: 1) 56.73 71.35 63.20 1288
-------------------
micro avg 99.83 99.83 99.83 638229
macro avg 78.34 85.62 81.56 638229
weighted avg 99.85 99.83 99.84 638229
punct_post test report:
label precision recall f1 support
<NULL> (label_id: 0) 99.19 98.41 98.80 578271
<ACRONYM> (label_id: 1) 30.10 56.36 39.24 55
. (label_id: 2) 91.92 93.12 92.52 30856
, (label_id: 3) 72.98 82.44 77.42 27761
? (label_id: 4) 52.77 71.85 60.85 1286
? (label_id: 5) 0.00 0.00 0.00 0
, (label_id: 6) 0.00 0.00 0.00 0
。 (label_id: 7) 0.00 0.00 0.00 0
、 (label_id: 8) 0.00 0.00 0.00 0
・ (label_id: 9) 0.00 0.00 0.00 0
। (label_id: 10) 0.00 0.00 0.00 0
؟ (label_id: 11) 0.00 0.00 0.00 0
, (label_id: 12) 0.00 0.00 0.00 0
; (label_id: 13) 0.00 0.00 0.00 0
። (label_id: 14) 0.00 0.00 0.00 0
፣ (label_id: 15) 0.00 0.00 0.00 0
፧ (label_id: 16) 0.00 0.00 0.00 0
-------------------
micro avg 97.40 97.40 97.40 638229
macro avg 69.39 80.44 73.77 638229
weighted avg 97.60 97.40 97.48 638229
cap test report:
label precision recall f1 support
LOWER (label_id: 0) 99.82 99.86 99.84 2324724
UPPER (label_id: 1) 95.92 94.70 95.30 79266
-------------------
micro avg 99.69 99.69 99.69 2403990
macro avg 97.87 97.28 97.57 2403990
weighted avg 99.69 99.69 99.69 2403990
seg test report:
label precision recall f1 support
NOSTOP (label_id: 0) 99.99 99.96 99.98 607057
FULLSTOP (label_id: 1) 99.31 99.88 99.60 34172
-------------------
micro avg 99.96 99.96 99.96 641229
macro avg 99.65 99.92 99.79 641229
weighted avg 99.96 99.96 99.96 641229
</details>
<details> <summary>阿姆哈拉语</summary>
punct_post test report:
label precision recall f1 support
<NULL> (label_id: 0) 99.83 99.28 99.56 729664
<ACRONYM> (label_id: 1) 0.00 0.00 0.00 0
. (label_id: 2) 0.00 0.00 0.00 0
, (label_id: 3) 0.00 0.00 0.00 0
? (label_id: 4) 0.00 0.00 0.00 0
? (label_id: 5) 0.00 0.00 0.00 0
, (label_id: 6) 0.00 0.00 0.00 0
。 (label_id: 7) 0.00 0.00 0.00 0
、 (label_id: 8) 0.00 0.00 0.00 0
・ (label_id: 9) 0.00 0.00 0.00 0
। (label_id: 10) 0.00 0.00 0.00 0
؟ (label_id: 11) 0.00 0.00 0.00 0
, (label_id: 12) 0.00 0.00 0.00 0
; (label_id: 13) 0.00 0.00 0.00 0
። (label_id: 14) 91.27 97.90 94.47 25341
፣ (label_id: 15) 61.93 82.11 70.60 5818
፧ (label_id: 16) 67.41 81.73 73.89 1177
-------------------
micro avg 99.08 99.08 99.08 762000
macro avg 80.11 90.26 84.63 762000
weighted avg 99.21 99.08 99.13 762000
cap test report:
label precision recall f1 support
LOWER (label_id: 0) 98.40 98.03 98.21 1064
UPPER (label_id: 1) 71.23 75.36 73.24 69
-------------------
micro avg 96.65 96.65 96.65 1133
macro avg 84.81 86.69 85.73 1133
weighted avg 96.74 96.65 96.69 1133
seg test report:
label precision recall f1 support
NOSTOP (label_id: 0) 99.99 99.85 99.92 743158
FULLSTOP (label_id: 1) 95.20 99.62 97.36 21842
-------------------
micro avg 99.85 99.85 99.85 765000
macro avg 97.59 99.74 98.64 765000
weighted avg 99.85 99.85 99.85 765000
</details>
<details> <summary>中文</summary>
punct_post test report:
label precision recall f1 support
<NULL> (label_id: 0) 99.53 97.31 98.41 435611
<ACRONYM> (label_id: 1) 0.00 0.00 0.00 0
. (label_id: 2) 0.00 0.00 0.00 0
, (label_id: 3) 0.00 0.00 0.00 0
? (label_id: 4) 0.00 0.00 0.00 0
? (label_id: 5) 81.85 87.31 84.49 1513
, (label_id: 6) 74.08 93.67 82.73 35921
。 (label_id: 7) 96.51 96.93 96.72 32097
、 (label_id: 8) 0.00 0.00 0.00 0
・ (label_id: 9) 0.00 0.00 0.00 0
। (label_id: 10) 0.00 0.00 0.00 0
؟ (label_id: 11) 0.00 0.00 0.00 0
, (label_id: 12) 0.00 0.00 0.00 0
; (label_id: 13) 0.00 0.00 0.00 0
። (label_id: 14) 0.00 0.00 0.00 0
፣ (label_id: 15) 0.00 0.00 0.00 0
፧ (label_id: 16) 0.00 0.00 0.00 0
-------------------
micro avg 97.00 97.00 97.00 505142
macro avg 87.99 93.81 90.59 505142
weighted avg 97.48 97.00 97.15 505142
cap test report:
label precision recall f1 support
LOWER (label_id: 0) 94.89 94.98 94.94 2951
UPPER (label_id: 1) 81.34 81.03 81.18 796
-------------------
micro avg 92.02 92.02 92.02 3747
macro avg 88.11 88.01 88.06 3747
weighted avg 92.01 92.02 92.01 3747
seg test report:
label precision recall f1 support
NOSTOP (label_id: 0) 99.99 99.97 99.98 473642
FULLSTOP (label_id: 1) 99.55 99.90 99.72 34500
-------------------
micro avg 99.96 99.96 99.96 508142
macro avg 99.77 99.93 99.85 508142
weighted avg 99.96 99.96 99.96 508142
</details>
<details> <summary>日语</summary>
punct_post test report:
label precision recall f1 support
<NULL> (label_id: 0) 99.34 95.90 97.59 406341
<ACRONYM> (label_id: 1) 0.00 0.00 0.00 0
. (label_id: 2) 0.00 0.00 0.00 0
, (label_id: 3) 0.00 0.00 0.00 0
? (label_id: 4) 0.00 0.00 0.00 0
? (label_id: 5) 70.55 73.56 72.02 1456
, (label_id: 6) 0.00 0.00 0.00 0
。 (label_id: 7) 94.38 96.95 95.65 32537
、 (label_id: 8) 54.28 87.62 67.03 18610
・ (label_id: 9) 28.18 71.64 40.45 1100
। (label_id: 10) 0.00 0.00 0.00 0
؟ (label_id: 11) 0.00 0.00 0.00 0
, (label_id: 12) 0.00 0.00 0.00 0
; (label_id: 13) 0.00 0.00 0.00 0
። (label_id: 14) 0.00 0.00 0.00 0
፣ (label_id: 15) 0.00 0.00 0.00 0
፧ (label_id: 16) 0.00 0.00 0.00 0
-------------------
micro avg 95.51 95.51 95.51 460044
macro avg 69.35 85.13 74.55 460044
weighted avg 96.91 95.51 96.00 460044
cap test report:
label precision recall f1 support
LOWER (label_id: 0) 92.33 94.03 93.18 4174
UPPER (label_id: 1) 83.51 79.46 81.43 1587
-------------------
micro avg 90.02 90.02 90.02 5761
macro avg 87.92 86.75 87.30 5761
weighted avg 89.90 90.02 89.94 5761
seg test report:
label precision recall f1 support
NOSTOP (label_id: 0) 99.99 99.92 99.96 428544
FULLSTOP (label_id: 1) 99.07 99.87 99.47 34500
-------------------
micro avg 99.92 99.92 99.92 463044
macro avg 99.53 99.90 99.71 463044
weighted avg 99.92 99.92 99.92 463044
</details>
<details> <summary>印地语</summary>
punct_post test report:
label precision recall f1 support
<NULL> (label_id: 0) 99.75 99.44 99.59 560358
<ACRONYM> (label_id: 1) 0.00 0.00 0.00 0
. (label_id: 2) 0.00 0.00 0.00 0
, (label_id: 3) 69.55 78.48 73.75 8084
? (label_id: 4) 63.30 87.07 73.31 317
? (label_id: 5) 0.00 0.00 0.00 0
, (label_id: 6) 0.00 0.00 0.00 0
。 (label_id: 7) 0.00 0.00 0.00 0
、 (label_id: 8) 0.00 0.00 0.00 0
・ (label_id: 9) 0.00 0.00 0.00 0
। (label_id: 10) 96.92 98.66 97.78 32118
؟ (label_id: 11) 0.00 0.00 0.00 0
, (label_id: 12) 0.00 0.00 0.00 0
; (label_id: 13) 0.00 0.00 0.00 0
። (label_id: 14) 0.00 0.00 0.00 0
፣ (label_id: 15) 0.00 0.00 0.00 0
፧ (label_id: 16) 0.00 0.00 0.00 0
-------------------
micro avg 99.11 99.11 99.11 600877
macro avg 82.38 90.91 86.11 600877
weighted avg 99.17 99.11 99.13 600877
cap test report:
label precision recall f1 support
LOWER (label_id: 0) 97.19 96.72 96.95 2466
UPPER (label_id: 1) 89.14 90.60 89.86 734
-------------------
micro avg 95.31 95.31 95.31 3200
macro avg 93.17 93.66 93.41 3200
weighted avg 95.34 95.31 95.33 3200
seg test report:
label precision recall f1 support
NOSTOP (label_id: 0) 100.00 99.99 99.99 569472
FULLSTOP (label_id: 1) 99.82 99.99 99.91 34405
-------------------
micro avg 99.99 99.99 99.99 603877
macro avg 99.91 99.99 99.95 603877
weighted avg 99.99 99.99 99.99 603877
</details>
<details> <summary>阿拉伯语</summary>
punct_post test report:
label precision recall f1 support
<NULL> (label_id: 0) 99.30 96.94 98.10 688043
<ACRONYM> (label_id: 1) 93.33 77.78 84.85 18
. (label_id: 2) 93.31 93.78 93.54 28175
, (label_id: 3) 0.00 0.00 0.00 0
? (label_id: 4) 0.00 0.00 0.00 0
? (label_id: 5) 0.00 0.00 0.00 0
, (label_id: 6) 0.00 0.00 0.00 0
。 (label_id: 7) 0.00 0.00 0.00 0
、 (label_id: 8) 0.00 0.00 0.00 0
・ (label_id: 9) 0.00 0.00 0.00 0
। (label_id: 10) 0.00 0.00 0.00 0
؟ (label_id: 11) 65.93 82.79 73.40 860
, (label_id: 12) 44.89 79.20 57.30 20941
; (label_id: 13) 0.00 0.00 0.00 0
። (label_id: 14) 0.00 0.00 0.00 0
፣ (label_id: 15) 0.00 0.00 0.00 0
፧ (label_id: 16) 0.00 0.00 0.00 0
-------------------
micro avg 96.29 96.29 96.29 738037
macro avg 79.35 86.10 81.44 738037
weighted avg 97.49 96.29 96.74 738037
cap test report:
label precision recall f1 support
LOWER (label_id: 0) 97.10 99.49 98.28 4137
UPPER (label_id: 1) 98.71 92.89 95.71 1729
-------------------
micro avg 97.55 97.55 97.55 5866
macro avg 97.90 96.19 96.99 5866
weighted avg 97.57 97.55 97.52 5866
seg test report:
label precision recall f1 support
NOSTOP (label_id: 0) 99.99 99.97 99.98 710456
FULLSTOP (label_id: 1) 99.39 99.85 99.62 30581
-------------------
micro avg 99.97 99.97 99.97 741037
macro avg 99.69 99.91 99.80 741037
weighted avg 99.97 99.97 99.97 741037
</details>
额外内容
缩写词、首字母缩写和双大写词
本节简要展示模型在遇到以下情况时的行为:
- 缩写词:"NATO"
- 伪缩写词:用 "NHTG" 代替 "NATO"
- 可能是缩写词或专有名词的歧义词:"Tuny"
- 双大写词:"McDavid"
- 首字母缩写词:"p.m."
<details open>
<summary>缩写词等输入</summary>
from typing import List
from punctuators.models import PunctCapSegModelONNX
m: PunctCapSegModelONNX = PunctCapSegModelONNX.from_pretrained(
"1-800-BAD-CODE/xlm-roberta_punctuation_fullstop_truecase"
)
input_texts = [
"the us is a nato member as a nato member the country enjoys security guarantees notably article 5",
"the us is a nhtg member as a nhtg member the country enjoys security guarantees notably article 5",
"the us is a tuny member as a tuny member the country enjoys security guarantees notably article 5",
"connor andrew mcdavid is a canadian professional ice hockey centre and captain of the edmonton oilers of the national hockey league the oilers selected him first overall in the 2015 nhl entry draft mcdavid spent his childhood playing ice hockey against older children",
"please rsvp for the party asap preferably before 8 pm today",
]
results: List[List[str]] = m.infer(
texts=input_texts, apply_sbd=True,
)
for input_text, output_texts in zip(input_texts, results):
print(f"Input: {input_text}")
print(f"Outputs:")
for text in output_texts:
print(f"\t{text}")
print()
</details>
<details open>
<summary>预期输出</summary>
Input: the us is a nato member as a nato member the country enjoys security guarantees notably article 5
Outputs:
The U.S. is a NATO member.
As a NATO member, the country enjoys security guarantees, notably Article 5.
Input: the us is a nhtg member as a nhtg member the country enjoys security guarantees notably article 5
Outputs:
The U.S. is a NHTG member.
As a NHTG member, the country enjoys security guarantees, notably Article 5.
Input: the us is a tuny member as a tuny member the country enjoys security guarantees notably article 5
Outputs:
The U.S. is a Tuny member.
As a Tuny member, the country enjoys security guarantees, notably Article 5.
Input: connor andrew mcdavid is a canadian professional ice hockey centre and captain of the edmonton oilers of the national hockey league the oilers selected him first overall in the 2015 nhl entry draft mcdavid spent his childhood playing ice hockey against older children
Outputs:
Connor Andrew McDavid is a Canadian professional ice hockey centre and captain of the Edmonton Oilers of the National Hockey League.
The Oilers selected him first overall in the 2015 NHL entry draft.
McDavid spent his childhood playing ice hockey against older children.
Input: please rsvp for the party asap preferably before 8 pm tonight
Outputs:
Please RSVP for the party ASAP, preferably before 8 p.m. tonight.
</details>
1-800-BAD-CODE/xlm-roberta_punctuation_fullstop_truecase
作者 1-800-BAD-CODE
创建时间: 2023-05-07 22:33:05+00:00
更新时间: 2023-07-15 20:42:28+00:00
在 Hugging Face 上查看