如何将 SHAP 与 SpaCy 模型一起使用？

Question

我正在尝试提高 SpaCy 二进制文本分类模型的可解释性，该模型是通过使用 SHAP 解释预测来训练的。这是我迄今为止尝试过的（遵循this 教程）：

nlp = spacy.load("my_model") # load my model
explainer = shap.Explainer(nlp_predict)
shap_values = explainer(["This is an example"])

但我得到AttributeError: 'str' object has no attribute 'shape'。 nlp_predict 是我编写的一种方法，它采用文本列表并以教程中使用的格式输出每个文本的预测概率。我在这里错过了什么？

---

这是我的格式化函数：

def nlp_predict(texts):
    result = []
    for text in texts:
        prediction = nlp_fn(text) # This returns label probability but in the wrong format
        sub_result = []
        sub_result.append({'label': 'label1', 'score': prediction["label1"]})
        sub_result.append({'label': 'label2', 'score': prediction["label2"]})
        result.append(sub_result)
    return(result)

这是他们在教程中使用的预测格式（针对 2 个数据点）：

[[{'label': 'label1', 'score': 2.850986311386805e-05},
  {'label': 'label2', 'score': 0.9999715089797974}],
 [{'label': 'label1', 'score': 0.00010622951958794147},
  {'label': 'label2', 'score': 0.9998937845230103}]]

我的函数的输出与此匹配，但我仍然得到AttributeError。这是我收到的完整错误消息：。

Answer 1

问题在于 shap 只为转换器库的标记器和模型实现了方法。

SpaCy 标记器的工作方式非常不同，特别是不会返回标记 id 作为标记化的结果。

因此，完成这项工作的解决方案需要编写一个函数来包装 spacy 分词器以返回与转换器分词器相同的数据（例如 [{'input_ids': [101, 7592, ...], 'offset_mapping': [(0, 5), (6, 9), ...], ...}] 之类的东西），或者在 shap 中添加对 spacy 分词器/模型的支持。

这是一个例子，我破解了前者的解决方案。

• 围绕我的 spacy 模型进行预测/标记化的包装器：

import spacy
textcat_spacy = spacy.load("my-model")
tokenizer_spacy = spacy.tokenizer.Tokenizer(textcat_spacy.vocab)

# Run the spacy pipeline on some random text just to retrieve the classes
doc = textcat_spacy("hi")
classes = list(doc.cats.keys())

# Define a function to predict
def predict(texts):
    # convert texts to bare strings
    texts = [str(text) for text in texts]
    results = []
    for doc in textcat_spacy.pipe(texts):
        # results.append([{'label': cat, 'score': doc.cats[cat]} for cat in doc.cats])
        results.append([doc.cats[cat] for cat in classes])
    return results

# Create a function to create a transformers-like tokenizer to match shap's expectations
def tok_adapter(text, return_offsets_mapping=False):
    doc = tokenizer_spacy(text)
    out = {"input_ids": [tok.norm for tok in doc]}
    if return_offsets_mapping:
        out["offset_mapping"] = [(tok.idx, tok.idx + len(tok)) for tok in doc]
    return out

• 形状解释器配置：

import shap
# Create the Shap Explainer
# - predict is the "model" function, adapted to a transformers-like model
# - masker is the masker used by shap, which relies on a transformers-like tokenizer
# - algorithm is set to permuation, which is the one used for transformers models
# - output_names are the classes (altough it is not propagated to the permutation explainer currently, which is why plots do not have the labels)
# - max_evals is set to a high number to reduce the probability of cases where the explainer fails because there are too many tokens
explainer = shap.Explainer(predict, masker=shap.maskers.Text(tok_adapter), algorithm="permutation", output_names=classes, max_evals=1500)

• 用法：

sample = "Some text to classify"
# Process the text using SpaCy
doc = textcat_spacy(sample)
# Get the shap values
shap_values = explainer([sample])