promptbench.utils.dataprocess¶
- class promptbench.utils.dataprocess.InputProcess¶
Bases:
objectA utility class for processing input data for language models.
This class provides static methods to format input data based on given prompt templates and input data dictionaries.
Methods:¶
- basic_format(prompt_template, input_data_dict)
Combines a prompt template and input data to create a formatted model input.
- static basic_format(prompt_template, input_data_dict)¶
Combine the prompt and input to create an input for the model.
Parameters: - prompt_template (str): The template for the prompt with placeholders. - input_data_dict (dict): Dictionary containing data to fill in the template.
Returns: - str: The combined model input.
- class promptbench.utils.dataprocess.OutputProcess¶
Bases:
objectA utility class for processing raw predictions from language models.
This class provides static methods for various ways to process and clean up raw prediction text.
Methods:¶
- general(raw_pred, proj_func=None)
Performs general processing on the raw prediction text.
- cls(raw_pred, proj_func=None)
Processes the raw prediction text for classification tasks.
- pattern_split(raw_pred, pattern, proj_func=None)
Splits the raw prediction text based on a pattern.
- pattern_re(raw_pred, pattern, proj_func=None)
Uses regular expressions to process the raw prediction text.
- static cls(raw_pred, proj_func=None)¶
Processes the prediction by taking the last word after basic processing.
Parameters: - raw_pred (str): The raw prediction text.
Returns: - str: The last word from the processed prediction text.
- static general(raw_pred, proj_func=None)¶
General processing for predictions using the base prediction process.
Parameters: - raw_pred (str): The raw prediction text.
Returns: - str: The processed prediction text.
- static pattern_re(raw_pred, pattern, proj_func=None)¶
Processes the prediction using regular expressions to extract a specific pattern.
Parameters: - raw_pred (str): The raw prediction text. - pattern (str): The regular expression pattern to search for.
Returns: - str: The matched pattern from the prediction text, or the original text if no match.
- static pattern_split(raw_pred, pattern, proj_func=None)¶
Processes the prediction by splitting it based on a provided pattern and taking the last part.
Parameters: - raw_pred (str): The raw prediction text. - pattern (str): The pattern to split the prediction text on.
Returns: - str: The last part of the prediction text after splitting.