QuAcc/quacc/error.py

56 lines
1.5 KiB
Python
Raw Normal View History

import numpy as np
2023-05-20 20:23:17 +02:00
2023-05-20 20:23:17 +02:00
def from_name(err_name):
assert err_name in ERROR_NAMES, f"unknown error {err_name}"
callable_error = globals()[err_name]
return callable_error
2023-09-26 07:58:40 +02:00
# def f1(prev):
# # https://github.com/dice-group/gerbil/wiki/Precision,-Recall-and-F1-measure
# if prev[0] == 0 and prev[1] == 0 and prev[2] == 0:
# return 1.0
# elif prev[0] == 0 and prev[1] > 0 and prev[2] == 0:
# return 0.0
# elif prev[0] == 0 and prev[1] == 0 and prev[2] > 0:
# return float('NaN')
# else:
# recall = prev[0] / (prev[0] + prev[1])
# precision = prev[0] / (prev[0] + prev[2])
2023-09-26 07:58:40 +02:00
# return 2 * (precision * recall) / (precision + recall)
def f1(prev):
den = (2 * prev[3]) + prev[1] + prev[2]
2023-09-26 07:58:40 +02:00
if den == 0:
return 0.0
2023-06-05 21:54:22 +02:00
else:
return (2 * prev[3]) / den
def f1e(prev):
return 1 - f1(prev)
2023-09-26 07:58:40 +02:00
def acc(prev: np.ndarray) -> float:
return (prev[0] + prev[3]) / np.sum(prev)
def accd(true_prevs: np.ndarray, estim_prevs: np.ndarray) -> np.ndarray:
vacc = np.vectorize(acc, signature="(m)->()")
a_tp = vacc(true_prevs)
a_ep = vacc(estim_prevs)
return np.abs(a_tp - a_ep)
def maccd(true_prevs: np.ndarray, estim_prevs: np.ndarray) -> float:
return accd(true_prevs, estim_prevs).mean()
ACCURACY_ERROR = {maccd}
ACCURACY_ERROR_SINGLE = {accd}
ACCURACY_ERROR_NAMES = {func.__name__ for func in ACCURACY_ERROR}
ACCURACY_ERROR_SINGLE_NAMES = {func.__name__ for func in ACCURACY_ERROR_SINGLE}
ERROR_NAMES = ACCURACY_ERROR_NAMES | ACCURACY_ERROR_SINGLE_NAMES