🚀 RoBERTa-base 方面级情感分析模型

本模型专为词元分类任务设计，能够提取方面词，并预测与提取的方面词相关的情感极性。提取的方面词将是输入文本中表达情感的部分。

🚀 快速开始

本模型适用于词元分类任务，可提取方面词并预测其情感极性。下面为你介绍如何使用该模型进行推理。

✨ 主要特性

• 能够从输入文本中提取方面词。
• 预测提取的方面词的情感极性。

📦 安装指南

文档未提及安装步骤，可参考 transformers 库的官方安装指南进行安装。

💻 使用示例

基础用法

直接使用管道进行端到端推理：

from transformers import pipeline

ate_sent_pipeline = pipeline(task='ner', 
                  aggregation_strategy='simple',
                  model="gauneg/roberta-base-absa-ate-sentiment")

text_input = "Been here a few times and food has always been good but service really suffers when it gets crowded."
ate_sent_pipeline(text_input)

管道输出：

[{'entity_group': 'pos', #sentiment polarity
  'score': 0.8447307,
  'word': ' food', # aspect term
  'start': 26,
  'end': 30},
 {'entity_group': 'neg', #sentiment polarity
  'score': 0.81927896,
  'word': ' service', #aspect term
  'start': 56,
  'end': 63}]

高级用法

使用 Auto 类进行词元级推理：

from transformers import AutoTokenizer, AutoModelForTokenClassification
model_id = "gauneg/roberta-base-absa-ate-sentiment"
tokenizer = AutoTokenizer.from_pretrained(model_id)


# the sequence of labels used during training
labels = {"B-neu": 1, "I-neu": 2, "O": 0, "B-neg": 3, "B-con": 4, "I-pos": 5, "B-pos": 6, "I-con": 7, "I-neg": 8, "X": -100}
id2lab = {idx: lab for lab, idx in labels.items()}
lab2id = {lab: idx for lab, idx in labels.items()}

model = AutoModelForTokenClassification.from_pretrained(model_id, 
                                                        num_labels=len(labels), id2label=id2lab, label2id=lab2id)

# making one prediction at a time (should be padded/batched and truncated for efficiency)
text_input = "Been here a few times and food has always been good but service really suffers when it gets crowded."
tok_inputs = tokenizer(text_input, return_tensors="pt")


y_pred = model(**tok_inputs) # predicting the logits

# since first and the last tokens are excluded (<s> and </s>)
# they have to be removed before decoding the labels predicted against them
y_pred_fin = y_pred.logits.argmax(dim=-1)[0][1:-1] # selecting the most favoured labels for each token from the logits


decoded_pred = [id2lab[logx.item()] for logx in y_pred_fin]


## displaying the input tokens with predictions and skipping <s> and </s> tokens at the beginning and the end respectively
decoded_toks = tok_inputs['input_ids'][0][1:-1]
tok_levl_pred = list(zip(tokenizer.convert_ids_to_tokens(decoded_toks), decoded_pred))

结果存储在 tok_level_pred 变量中：

[('Be', 'O'),
 ('en', 'O'),
 ('Ġhere', 'O'),
 ('Ġa', 'O'),
 ('Ġfew', 'O'),
 ('Ġtimes', 'O'),
 ('Ġand', 'O'),
 ('Ġfood', 'B-pos'),
 ('Ġhas', 'O'),
 ('Ġalways', 'O'),
 ('Ġbeen', 'O'),
 ('Ġgood', 'O'),
 ('Ġbut', 'O'),
 ('Ġservice', 'B-neg'),
 ('Ġreally', 'O'),
 ('Ġsuffers', 'O'),
 ('Ġwhen', 'O'),
 ('Ġit', 'O'),
 ('Ġgets', 'O'),
 ('Ġcrowded', 'O'),
 ('.', 'O')]

📚 详细文档

数据集

本模型在以下数据集上进行了训练：

1. 基于方面的情感分析 SemEval 共享任务（2014，2015，2016）
2. 多方面多情感 MAMS

基准测试数据集评估

仅方面词提取（ATE）

第一次评估是针对词元提取任务，不考虑提取词元的极性。预期提取的词元是表达了情感的方面词元。（分数以 B - I - O 标签的微平均表示）

方面情感评估

本次评估考虑了词元提取任务以及提取词元的极性。预期提取的词元是表达了情感的方面词元以及情感的极性。（分数以宏平均表示）

📄 许可证

本模型采用 Apache - 2.0 许可证。