使用 .items() 遍历字典时不可散列的类型列表错误

Question

I am trying to iterate through a dictionary, where the keys are regular expressions, and the values are categories. My goal is to go through a column, iterate through each of the different regular expressions, and assign a new column flagged for that category. The problem I am running into is that .items() creates a list, which is an unhashable object.

new_columns = ['counts']
category_columns = [dictionary]

for problem_count, problem in zip(new_columns, category_columns):
    for category, reg_expr in problem.items():
        found = notes.str.contains(reg_expr, na=False) # Errors out here 
        note_data[category] = np.where(found, True, False)
    note_data[problem_count] = note_data[problem_keys()].sum(axis=1)

What is the best work around for this?

Answer 1

If values are lists, then I would Join them with | for a regex:

found = notes.str.contains('|'.join(reg_expr), na=False)