uqlm.white_box.top_logprobs.TopLogprobsScorer#
- class uqlm.white_box.top_logprobs.TopLogprobsScorer(scorers=['min_token_negentropy', 'mean_token_negentropy', 'probability_margin'])#
Bases:
LogprobsScorer- __init__(scorers=['min_token_negentropy', 'mean_token_negentropy', 'probability_margin'])#
Class for computing WhiteBox UQ scores with a single generation
Methods
__init__([scorers])Class for computing WhiteBox UQ scores with a single generation
evaluate(logprobs_results)Compute scores from top logprobs results
extract_logprobs(single_response_logprobs)Extract log probabilities from token data
extract_probs(single_response_logprobs)Extract probabilities from token data
extract_top_logprobs(single_response_logprobs)Extract top log probabilities for each token
- evaluate(logprobs_results)#
Compute scores from top logprobs results
- Return type:
Dict[str,List[float]]
- static extract_logprobs(single_response_logprobs)#
Extract log probabilities from token data
- Return type:
ndarray
- extract_probs(single_response_logprobs)#
Extract probabilities from token data
- Return type:
ndarray
- static extract_top_logprobs(single_response_logprobs)#
Extract top log probabilities for each token
- Return type:
List[ndarray]
References