guillory21 imported as baseline

This commit is contained in:
Lorenzo Volpi 2023-09-17 21:47:34 +02:00
parent d6b1f6e796
commit f537ecb5e4
4 changed files with 42 additions and 22 deletions

BIN
.coverage

Binary file not shown.

Binary file not shown.

4
guillory21_doc/doc.py Normal file
View File

@ -0,0 +1,4 @@
import numpy as np
def get_doc(probs1, probs2):
return np.mean(probs2) - np.mean(probs1)

View File

@ -1,16 +1,13 @@
from ast import get_docstring
from statistics import mean
from typing import Dict
from sklearn.base import BaseEstimator
from sklearn.model_selection import cross_validate
from quapy.data import LabelledCollection
from garg22_ATC.ATC_helper import (
find_ATC_threshold,
get_ATC_acc,
get_entropy,
get_max_conf,
)
import garg22_ATC.ATC_helper as atc
import numpy as np
from jiang18_trustscore.trustscore import TrustScore
import jiang18_trustscore.trustscore as trustscore
import guillory21_doc.doc as doc
def kfcv(c_model: BaseEstimator, validation: LabelledCollection) -> Dict:
@ -19,7 +16,7 @@ def kfcv(c_model: BaseEstimator, validation: LabelledCollection) -> Dict:
return {"f1_score": mean(scores["test_f1_macro"])}
def ATC_MC(
def atc_mc(
c_model: BaseEstimator,
validation: LabelledCollection,
test: LabelledCollection,
@ -34,21 +31,23 @@ def ATC_MC(
test_probs = c_model_predict(test.X)
## score function, e.g., negative entropy or argmax confidence
val_scores = get_max_conf(val_probs)
val_scores = atc.get_max_conf(val_probs)
#pred_idxv1 #calib_probsv1/probsv1
val_preds = np.argmax(val_probs, axis=-1)
test_scores = get_max_conf(test_probs)
_, ATC_thres = find_ATC_threshold(val_scores, val_labels == val_preds)
ATC_accuracy = get_ATC_acc(ATC_thres, test_scores)
#pred_probs_new #probs_new
test_scores = atc.get_max_conf(test_probs)
#pred_probsv1 #labelsv1 #pred_idxv1
_, atc_thres = atc.find_ATC_threshold(val_scores, val_labels == val_preds)
#calib_thres_balance #pred_probs_new
atc_accuracy = atc.get_ATC_acc(atc_thres, test_scores)
return {
"true_acc": 100 * np.mean(np.argmax(test_probs, axis=-1) == test.y),
"pred_acc": ATC_accuracy,
"pred_acc": atc_accuracy,
}
def ATC_NE(
def atc_ne(
c_model: BaseEstimator,
validation: LabelledCollection,
test: LabelledCollection,
@ -63,17 +62,17 @@ def ATC_NE(
test_probs = c_model_predict(test.X)
## score function, e.g., negative entropy or argmax confidence
val_scores = get_entropy(val_probs)
val_scores = atc.get_entropy(val_probs)
val_preds = np.argmax(val_probs, axis=-1)
test_scores = get_entropy(test_probs)
test_scores = atc.get_entropy(test_probs)
_, ATC_thres = find_ATC_threshold(val_scores, val_labels == val_preds)
ATC_accuracy = get_ATC_acc(ATC_thres, test_scores)
_, atc_thres = atc.find_ATC_threshold(val_scores, val_labels == val_preds)
atc_accuracy = atc.get_ATC_acc(atc_thres, test_scores)
return {
"true_acc": 100 * np.mean(np.argmax(test_probs, axis=-1) == test.y),
"pred_acc": ATC_accuracy,
"pred_acc": atc_accuracy,
}
@ -87,8 +86,25 @@ def trust_score(
test_pred = c_model_predict(test.X)
trust_model = TrustScore()
trust_model = trustscore.TrustScore()
trust_model.fit(validation.X, validation.y)
return trust_model.get_score(test.X, test_pred)
def doc_feat(
c_model: BaseEstimator,
validation: LabelledCollection,
test: LabelledCollection,
predict_method="predict_proba",
):
c_model_predict = getattr(c_model, predict_method)
val_probs, val_labels = c_model_predict(validation.X), validation.y
test_probs = c_model_predict(test.X)
val_scores = np.max(val_probs, axis=-1)
test_scores = np.max(test_probs, axis=-1)
val_preds = np.argmax(val_probs, axis=-1)
v1acc = np.mean(val_preds == val_labels)*100
return v1acc + doc.get_doc(val_scores, test_scores)