Thiru1122's picture
feat(deepfake-detection): comprehensive cross-model evaluation, parity audits, and Phase 7 v2 reference preprocessing
151f388
Raw History Blame Contribute Delete
4.71 kB
import numpy as np
import tensorboardX
from sklearn.metrics import log_loss, accuracy_score, precision_score, average_precision_score, roc_auc_score, \
recall_score, confusion_matrix
import torch
import torch.nn.functional as F
class Test_time_agumentation(object):
def __init__(self, is_rotation=True):
self.is_rotation = is_rotation
def __rotation(self, img):
"""
clockwise rotation 90 180 270
"""
img90 = img.rot90(-1, [2, 3]) # 1 逆时针; -1 顺时针
img180 = img.rot90(-1, [2, 3]).rot90(-1, [2, 3])
img270 = img.rot90(1, [2, 3])
return [img90, img180, img270]
def __inverse_rotation(self, img90, img180, img270):
"""
anticlockwise rotation 90 180 270
"""
img90 = img90.rot90(1, [2, 3]) # 1 逆时针; -1 顺时针
img180 = img180.rot90(1, [2, 3]).rot90(1, [2, 3])
img270 = img270.rot90(-1, [2, 3])
return img90, img180, img270
def __flip(self, img):
"""
Flip vertically and horizontally
"""
return [img.flip(2), img.flip(3)]
def __inverse_flip(self, img_v, img_h):
"""
Flip vertically and horizontally
"""
return img_v.flip(2), img_h.flip(3)
def tensor_rotation(self, img):
"""
img size: [H, W]
rotation degree: [90 180 270]
:return a rotated list
"""
# assert img.shape == (1024, 1024)
return self.__rotation(img)
def tensor_inverse_rotation(self, img_list):
"""
img size: [H, W]
rotation degree: [90 180 270]
:return a rotated list
"""
# assert img.shape == (1024, 1024)
return self.__inverse_rotation(img_list[0], img_list[1], img_list[2])
def tensor_flip(self, img):
"""
img size: [H, W]
:return a flipped list
"""
# assert img.shape == (1024, 1024)
return self.__flip(img)
def tensor_inverse_flip(self, img_list):
"""
img size: [H, W]
:return a flipped list
"""
# assert img.shape == (1024, 1024)
return self.__inverse_flip(img_list[0], img_list[1])
class AverageMeter(object):
"""Computes and stores the average and current value"""
def __init__(self):
self.reset()
def reset(self):
self.val = 0
self.avg = 0
self.sum = 0
self.count = 0
def update(self, val, n=1):
self.val = val
self.sum += val * n
self.count += n
self.avg = self.sum / self.count
class Logger(object):
def __init__(self, model_name, header):
self.header = header
self.writer = tensorboardX.SummaryWriter(model_name)
def __del(self):
self.writer.close()
def log(self, phase, values):
epoch = values['epoch']
for col in self.header[1:]:
self.writer.add_scalar(phase + "/" + col, float(values[col]), int(epoch))
def calculate_metrics(outputs, targets, metric_name='acc'):
if len(targets.data.numpy().shape) > 1:
_, targets = torch.max(targets.detach(), dim=1)
if len(outputs.data.numpy().shape) > 1 and outputs.data.numpy().shape[1] == 1: # 尾部是sigmoid
outputs = torch.cat([1 - outputs, outputs], dim=1)
# print(outputs.shape, targets.shape, pred_labels.size())
if metric_name == 'acc':
pred_labels = torch.max(outputs, 1)[1]
# print(targets, pred_labels)
return accuracy_score(targets.data.numpy(), pred_labels.detach().numpy())
elif metric_name == 'auc':
pred_labels = outputs[:, 1] # 为假的概率
return roc_auc_score(targets.data.numpy(), pred_labels.detach().numpy())
def calculate_fnr(y_true, y_pred):
"""
计算False Negative Rate
参数:
y_true -- 真实的类别标签
y_pred -- 预测的类别标签
返回:
fnr -- False Negative Rate
"""
# 生成混淆矩阵
tn, fp, fn, tp = confusion_matrix(y_true, y_pred).ravel()
# 计算FNR
fnr = fn / float(fn + tp)
return fnr
def compute_confusion_matric(outputs, targets):
"""
| 预测 |
-----------------
真|正| TP | FN |
--------------
值|负| FP | TN |
-----------------
:param outputs:
:param targets:
:return: tp, fp, tn ,fn
"""
part = outputs ^ targets
pcount = np.bincount(part)
tp_list = list(outputs & targets)
fp_list = list(outputs & ~targets)
tp = tp_list.count(1)
fp = fp_list.count(1)
tn = pcount[0] - tp
fn = pcount[1] - fp
return tp, fp, tn, fn
if __name__ == '__main__':
pass