Tinsae/Finetuning
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

cleaned up code

Browse Source
This commit is contained in:
Tinsae
2026-07-08 23:53:07 +02:00
parent 31f461342e
commit 04875a62e9
23 changed files with 1362 additions and 715 deletions
Download Patch File Download Diff File Expand all files Collapse all files

507
Tune.py
View File

@@ -1,268 +1,339 @@
# Finetuning a selected model import torch
# on a selected dataset import torch.nn as nn
# using selected parameters
from torch.utils.data import DataLoader from torch.utils.data import DataLoader
from sklearn.metrics import classification_report from sklearn.metrics import classification_report
import copy
# Framework and Utility Imports
import SetUp import SetUp
#from Data import * import Util
# from datasets.Casia import *
#from IdentitySubset import IdentitySubset
from sets.Data import * from sets.Data import *
from sets.IdentitySubset import IdentitySubset from sets.IdentitySubset import IdentitySubset
# models
from architectures.Model import Model, Architecture from architectures.Model import Model, Architecture
from unlearning.CertifiedUnlearning import CertifiedUnlearning
from unlearning.LinearFiltration import LinearFiltration from unlearning.LinearFiltration import LinearFiltration
from unlearning.CertifiedRemoval import CertifiedRemoval
from unlearning.WeightFiltration import WeightFiltration from unlearning.WeightFiltration import WeightFiltration
from eval.UnlearningAttack import UnlearningAttack
from unlearning.Retrain import Retrain
import Util
# WeightFiltration, CertifiedRemoval
# numbre of classes # Global Hyperparameters
CLASS_SIZE = 20 CLASS_SIZE:int = 20
# batch BATCH_SIZE:int = 16
BATCH_SIZE = 16 SAMPLE_SIZE:int = 30
TRAINING_SAMPLE:int = 27
# size of images per class trainset + testset
# 30 works best, more than that and we dont have enough data
SAMPLE_SIZE = 30
# this is then (full_sample - test_sample)
TRAINING_SMPLE = 27
# learning rate
LR_RATE = 0.0001
EPOCHS = 10
# depends on model architecture # depends on model architecture
# ResNet, DenseNet = 224 # ResNet, DenseNet = 224
# Inception = 299 # Inception = 299
RESOLUTION = 224 RESOLUTION:int = 224
FINETUNE = False # whether to fintune or just load finetuned model from dir # specify the model architecture,
# model architecture options are # Options here are the following
# - RESNET18 '''
# - RESNET50 RESNET18 # candidate
# - DENSENET121 RESNET50
# - INCEPTION RESNET34
# - GOOGLENET INCEPTION # candidate / or googleNet
# - EFFICIENTNET DENSENET121 # candidate
# - SHUFFLENET GOOGLENET # candidate / or Inception
arch = Architecture.RESNET50 EFFICIENTNET # candidate
SHUFFLENET
# DATA PREPARATION WIDE_RESNET
# load data set and prepare '''
dataset_name = Set_Name.CELEBA ARCH = Architecture.RESNET34
set = Set_Name.CELEBA
dataset = get_set(set_name=dataset_name)
print(f"> {dataset.__class__.__name__} dataset loaded")
# select identities for experiment
#selected_identities = select_ids(
# dataset = dataset,
# sample_size = SAMPLE_SIZE,
# class_size = CLASS_SIZE
# )
# this selects the top 50 based on sample size
# that way repeated calls return the same classes
selected_identities = select_top_ids(
dataset=dataset,
class_size=CLASS_SIZE
)
print(f'> Selected {CLASS_SIZE} random identity classes from CelebA dataset.')
print(f'> A class has {TRAINING_SMPLE} train and {SAMPLE_SIZE-TRAINING_SMPLE} test samples')
# split class images to train/test indices
train_indices, test_indices = get_indices(
dataset = dataset,
identities = selected_identities,
split_at = TRAINING_SMPLE,
size= SAMPLE_SIZE
)
# helps map class id to index
id_map = {old_id: new_id for new_id, old_id in enumerate(selected_identities)}
# we remap identities because crossEntropyLoss requires in indices 0 -> (n-1)
# where n = class size.
tr_transform = train_transform(RESOLUTION)
train_data = IdentitySubset(
dataset=dataset,
indices=train_indices,
id_mapping=id_map,
transform=tr_transform)
train_loader = DataLoader(
train_data,
batch_size = BATCH_SIZE,
shuffle = True)
print(f"> Total training images: {len(train_data)}")
print(f'> Constants : Classes = {CLASS_SIZE}, Batch = {BATCH_SIZE}, epochs = {EPOCHS}')
# MODEL PREPARATION
# cuda if exists (it does here)
device = SetUp.get_device()
for i in range(0,1):#CLASS_SIZE): # Data preparation and model setup
FORGET_CLASS_IDX = i def prepare_data_and_model_environment():
# Create model using Factory """
Handles environment discovery, downloads/loads datasets, generates
train-test class splits, and configures the architecture base.
"""
device = SetUp.get_device()
dataset_name = Set_Name.CASIAFACES
if dataset_name == Set_Name.CASIAFACES:
SAMPLE_SIZE = 400
TRAINING_SAMPLE = 320
model = None dataset = get_set(set_name=dataset_name)
print(f"> {dataset.__class__.__name__} dataset loaded")
if FINETUNE: # Select target identities (deterministic top sample identities)
model = Model.create( selected_identities = select_top_ids(dataset=dataset, class_size=CLASS_SIZE)
arch = arch, print(f'> Selected {CLASS_SIZE} random identity classes from {dataset_name.name} dataset.')
device = device, print(f'> A class has {TRAINING_SAMPLE} train and {SAMPLE_SIZE - TRAINING_SAMPLE} test samples')
size = CLASS_SIZE)
# we may need to load existing model or finetune # Isolate sample index partitions
model.train( train_indices, test_indices = get_indices(
epochs = EPOCHS, dataset=dataset,
loader = train_loader, identities=selected_identities,
rate = LR_RATE) split_at=TRAINING_SAMPLE,
size=SAMPLE_SIZE
)
# save. # Remap identities to 0 -> (N-1) range required by CrossEntropyLoss
file_name = f"{arch.name.lower}_{dataset_name.name.lower()}" id_map = {old_id: new_id for new_id, old_id in enumerate(selected_identities)}
model.save(filename=arch.name.lower())
# Build internal datasets using custom transforms
tr_transform = train_transform(RESOLUTION)
train_set = IdentitySubset(
dataset=dataset,
indices=train_indices,
id_mapping=id_map,
transform=tr_transform
)
# done tuning
# EVALUATE
te_transform = test_transform(RESOLUTION) te_transform = test_transform(RESOLUTION)
# Testing test_set = IdentitySubset(
test_data = IdentitySubset( dataset=dataset,
dataset = dataset,
indices=test_indices, indices=test_indices,
id_mapping=id_map, id_mapping=id_map,
transform=te_transform) transform=te_transform
)
test_loader = DataLoader( print(f"> Total training images: {len(train_set)}")
test_data, print(f'> Constants : Classes = {CLASS_SIZE}, Batch = {BATCH_SIZE}')
batch_size=BATCH_SIZE,
shuffle=False)
print(f"Total test images for these {CLASS_SIZE} classes: {len(test_data)}") # Create the base target model instance
base_model = Model.create(arch=ARCH, device=device, size=CLASS_SIZE)
# Evaluate return {
"device": device,
"train_set": train_set,
"test_set": test_set,
"base_model": base_model
}
# Fine tunning and evaluation
def run_finetuning_or_baseline_eval(env_dict, run_training=False, lr_rate=0.0001, epochs=14):
"""
Handles model training (if flag is true) and logs the baseline fine-tuned
performance to file metrics.
"""
model = env_dict["base_model"]
train_set = env_dict["train_set"]
test_set = env_dict["test_set"]
test_loader = DataLoader(test_set, batch_size=BATCH_SIZE, shuffle=False)
train_loader = DataLoader(train_set, batch_size=BATCH_SIZE, shuffle=True)
if not run_training:
return
# Finetuning
model.train(epochs=epochs, loader=train_loader, rate=lr_rate)
model.save(filename=ARCH.name.lower())
print(f"Model saved to trained_models/{ARCH.name.lower()}.pth")
print(f"Total test images for these {CLASS_SIZE} classes: {len(test_set)}")
# Evaluate original base checkpoint performance
current_mode = "Finetuned" current_mode = "Finetuned"
if FINETUNE:
# evaluate finetuned model
#current_mode = "Finetuned" try:
accuracy, report_dict = model.evaluate( accuracy, report_dict = model.evaluate(loader=test_loader, mode=current_mode)
loader = test_loader,
mode=current_mode
)
Util._log_to_csv( Util._log_to_csv(
arch=model.__class__.__name__, arch=ARCH.name,#model.__class__.__name__,
mode = current_mode, mode=current_mode,
accuracy=accuracy, accuracy=accuracy,
report_dict=report_dict, report_dict=report_dict,
strategy="base" strategy="base"
) )
except Exception as e:
print(f">> Skipping baseline log generation. Reason: {e}")
# unlearning algorithms
#linear_filtration = LinearFiltration(target_class_index=FORGET_CLASS_IDX)
#filtration.apply(reloaded.model)
#weight_filtration = WeightFiltration(num_classes = CLASS_SIZE,target_class_idx=FORGET_CLASS_IDX) # saves evaluation metrics to log files
#weight_filtration.apply(reloaded.model) def log_metrics(evaluation_domains, reloaded, strategy_in_use):
# Process and append metrics to target reporting paths
for domain in evaluation_domains:
print(domain["label"])
accuracy, report_dict = reloaded.evaluate(loader=domain["loader"], mode=domain["mode"])
Util._log_to_csv(
arch=ARCH.name,
mode=domain["mode"],
accuracy=accuracy,
report_dict=report_dict,
strategy=strategy_in_use
)
certified_removal = CertifiedRemoval( # Unlearning and strategy eval
target_class_index=FORGET_CLASS_IDX, def run_unlearning_and_strategy_eval(env_dict, forget_class_idx, strategy, evaluate = False, suite_runner=None):
s1=2, """
s2=500, Reloads a clean model state, applies the isolated unlearning framework,
unlearn_bs=2, and runs specific target evaluation domain checks.
scale=100.0, # Drop scale to match lower s2 depth """
std=0.00001) device = env_dict["device"]
#,removal_bound=0.05, epsilon=0.5, l2_reg=15) train_set = env_dict["train_set"]
#certified_removal.apply(reloaded.model) test_set = env_dict["test_set"]
# to be unlearned
forget_train_loader, retain_train_loader = get_unlearning_loaders( # Segment specific unlearning loaders using class index boundaries
dataset=train_data, retain_train_loader , forget_train_loader= get_unlearning_loaders(
forget_class_idx=FORGET_CLASS_IDX, dataset=train_set, forget_class_idx=forget_class_idx, batch_size=BATCH_SIZE
batch_size=BATCH_SIZE )
retain_test_loader, forget_test_loader = get_unlearning_loaders(
dataset=test_set, forget_class_idx=forget_class_idx, batch_size=BATCH_SIZE
) )
# to evaluate # Instantiate a clean copy of the model to keep weights isolated
forget_test_loader, retain_test_loader = get_unlearning_loaders( reloaded = Model.create(arch=ARCH, device=device, size=CLASS_SIZE)
dataset=test_data, reloaded.load(arch=ARCH)
forget_class_idx=FORGET_CLASS_IDX,
batch_size=BATCH_SIZE # Clean un-manipulated snapshot to serve as the Parameter-Space shadow proxy reference
) shadow_model = copy.deepcopy(reloaded)
if evaluate:
reloaded.evaluate(
loader=retain_test_loader, mode="finetuned"
)
print("fine tunned model loaded into evaluation sandbox")
# Execute strategic parameter unlearning step
# we are using only training data to unlearn.
# Test data is never touched here.
unlearned = strategy.apply(reloaded.model, train_set)
strategy_in_use = strategy.__class__.__name__
if isinstance(unlearned,nn.Module):
reloaded.model = unlearned
else:
reloaded = unlearned
#strategies = [linear_filtration, weight_filtration, certified_removal]
strategies = [certified_removal] is_retrained = isinstance(strategy, Retrain)
for strategy in strategies:
# test again if is_retrained:
reloaded = Model.create( os.makedirs("trained_models", exist_ok=True)
arch=arch, reloaded.save(filename=f"class_{forget_class_idx}_retrained.pth")
device = device,
size = CLASS_SIZE
# here we add a condition conditional statement
if suite_runner is not None:
test_loader = DataLoader(test_set, batch_size=BATCH_SIZE, shuffle=False)
suite_runner.run_complete_evaluation(
framework_name=strategy_in_use,
test_loader = test_loader,
target_class=forget_class_idx,
forget_train_loader=forget_train_loader,
forget_test_loader=forget_test_loader,
unlearned_instance=reloaded,
base_shadow_instance=shadow_model,
device=device
)
# Define validation tracking steps dynamically
evaluation_domains = [
{"loader": retain_test_loader, "mode": "retain", "label": "\n--- Performance on Retained Classes"},
{"loader": forget_test_loader, "mode": "forget", "label": "\n--- Performance on Forgotten Class"},
{"loader": forget_train_loader, "mode": "forget_train", "label": "\n--- Performance on Forgotten Class (Train Set - Verifying Unlearning)"}
]
log_metrics(evaluation_domains, reloaded, strategy_in_use)
# entry
if __name__ == "__main__":
outer_loop = 10
inner_loop = CLASS_SIZE
for k in range(outer_loop):
try:
# Data Infrastructure and Architecture
runtime_environment = prepare_data_and_model_environment()
# Baseline Evaluation
# switch finetuning for tests on strategies only,
# to avoid finetunning every time we test a strategy
finetuning = False
run_finetuning_or_baseline_eval(runtime_environment, run_training = finetuning)
# scale 16400.0 for ResNet
scale = 20100
# batch 8 for resNet,
unlearning_batches = 16
# regularis
# strategies
# implementation of Certified Removal for DNNs
certified_unlearning = CertifiedUnlearning(
target_class_index=0, #arch ResNet18 GoogLeNet Inception
l2_reg=0.000002 , # 0.000002 0.00001 0.0
gamma=0.01, # 0.1 0.1 0.01
scale= scale, # 16400.0 35000.0
s1=2, # 2
s2=350, # 300
std=0.00001, # 0.00001
unlearn_bs=unlearning_batches # 8 32 8
) )
reloaded.load(arch = arch)
print("fine tunned model loaded")
# reloaded.evaluate(
# loader = test_loader
#)
if not FINETUNE: # Normalisation Filtration
reloaded.evaluate( linear_filtration = LinearFiltration(
loader = test_loader,
mode=current_mode
)
# Unlearning target_class_index=0,
# train loaders passed here num_classes=CLASS_SIZE
strategy.apply(reloaded.model, forget_train_loader, retain_train_loader) )
# Performance Analysis # WF-Net
strategy_in_use = strategy.__class__.__name__ weight_filtration = WeightFiltration(
target_class_index=0, #arch ResNet18 GoogLeNet/Inception
epochs=6, #
lr=250.0, # ResNet18 = 150 # 150 100
gamma=0.001, # 0.001
lambda_1=30, # 25 100
arch=ARCH
)
# evaluation on retain Test_set retrain = Retrain(
current_mode = "retain" target_class_index = 0,
print("\n--- Performance on Retained Classes") arch = ARCH,
accuracy, report_dict = reloaded.evaluate(loader=retain_test_loader, mode=current_mode) size = CLASS_SIZE,
lr = 0.0001,
epochs = 14
Util._log_to_csv( )
arch=reloaded.__class__.__name__,
mode = current_mode, strategies = [
accuracy=accuracy, retrain,
report_dict=report_dict, linear_filtration,
strategy=strategy_in_use weight_filtration,
) certified_unlearning,
]
suite_runner = UnlearningAttack(arch=ARCH, class_size=CLASS_SIZE)
# Unlearning Iteration
for i in range(inner_loop):
for strategy in strategies:
# update target class to be unlearned
strategy.set_target_class(i)
print(f"Unlearning class {i} with {strategy.strategy_name}")
# forget
run_unlearning_and_strategy_eval(
runtime_environment,
forget_class_idx=i,
strategy=strategy,
# if we are finetuning, no need to evaluate base model.
# or may be never when not either!
evaluate = not finetuning,
suite_runner=suite_runner
)
# evaluation on forget Test_set except KeyboardInterrupt:
print("\n--- Performance on Forgotten Class") print("\nprogram interrupted. Exit!")
current_mode = "forget" break
accuracy, report_dict = reloaded.evaluate(loader=forget_test_loader,mode=current_mode)
Util._log_to_csv(
arch=reloaded.__class__.__name__,
mode = current_mode,
accuracy=accuracy,
report_dict=report_dict,
strategy=strategy_in_use
)
# evaluation on forget Train_set
# we expect this to be equal or highr than accuracy on Forget Test_set
current_mode = "forget_train"
print("\n--- Performance on Forgotten Class (Train Set - Verifying Unlearning)")
accuracy, report_dict = reloaded.evaluate(loader=forget_train_loader, mode=current_mode)
Util._log_to_csv(
arch= reloaded.__class__.__name__,
mode = current_mode,
accuracy=accuracy,
report_dict=report_dict,
strategy=strategy_in_use
)

342
Tune_new.py
View File

@@ -1,342 +0,0 @@
import torch
import torch.nn as nn
from torch.utils.data import DataLoader
from sklearn.metrics import classification_report
import copy
# Framework and Utility Imports
import SetUp
import Util
from sets.Data import *
from sets.IdentitySubset import IdentitySubset
from architectures.Model import Model, Architecture
from unlearning.CertifiedUnlearning import CertifiedUnlearning
from unlearning.LinearFiltration import LinearFiltration
from unlearning.WeightFiltration import WeightFiltration
from eval.UnlearningAttack import UnlearningAttack
from unlearning.Retrain import Retrain
# Global Hyperparameters
CLASS_SIZE = 20
BATCH_SIZE = 16
SAMPLE_SIZE = 30
TRAINING_SAMPLE = 27
# depends on model architecture
# ResNet, DenseNet = 224
# Inception = 299
RESOLUTION = 224
# specify the model architecture,
# Options here are the following
'''
RESNET18 # candidate
RESNET50
RESNET34
INCEPTION # candidate / or googleNet
DENSENET121 # candidate
GOOGLENET # candidate / or Inception
EFFICIENTNET # candidate
SHUFFLENET
WIDE_RESNET
'''
ARCH = Architecture.RESNET34
# Data preparation and model setup
def prepare_data_and_model_environment():
"""
Handles environment discovery, downloads/loads datasets, generates
train-test class splits, and configures the architecture base.
"""
device = SetUp.get_device()
dataset_name = Set_Name.CASIAFACES
if dataset_name == Set_Name.CASIAFACES:
SAMPLE_SIZE = 400
TRAINING_SAMPLE = 320
dataset = get_set(set_name=dataset_name)
print(f"> {dataset.__class__.__name__} dataset loaded")
# Select target identities (deterministic top sample identities)
selected_identities = select_top_ids(dataset=dataset, class_size=CLASS_SIZE)
print(f'> Selected {CLASS_SIZE} random identity classes from {dataset_name.name} dataset.')
print(f'> A class has {TRAINING_SAMPLE} train and {SAMPLE_SIZE - TRAINING_SAMPLE} test samples')
# Isolate sample index partitions
train_indices, test_indices = get_indices(
dataset=dataset,
identities=selected_identities,
split_at=TRAINING_SAMPLE,
size=SAMPLE_SIZE
)
# Remap identities to 0 -> (N-1) range required by CrossEntropyLoss
id_map = {old_id: new_id for new_id, old_id in enumerate(selected_identities)}
# Build internal datasets using custom transforms
tr_transform = train_transform(RESOLUTION)
train_data = IdentitySubset(
dataset=dataset,
indices=train_indices,
id_mapping=id_map,
transform=tr_transform
)
te_transform = test_transform(RESOLUTION)
test_data = IdentitySubset(
dataset=dataset,
indices=test_indices,
id_mapping=id_map,
transform=te_transform
)
print(f"> Total training images: {len(train_data)}")
print(f'> Constants : Classes = {CLASS_SIZE}, Batch = {BATCH_SIZE}')
# Create the base target model instance
base_model = Model.create(arch=ARCH, device=device, size=CLASS_SIZE)
return {
"device": device,
"train_data": train_data,
"test_data": test_data,
"base_model": base_model
}
# Fine tunning and evaluation
def run_finetuning_or_baseline_eval(env_dict, run_training=False, lr_rate=0.0001, epochs=14):
"""
Handles model training (if flag is true) and logs the baseline fine-tuned
performance to file metrics.
"""
model = env_dict["base_model"]
train_data = env_dict["train_data"]
test_data = env_dict["test_data"]
test_loader = DataLoader(test_data, batch_size=BATCH_SIZE, shuffle=False)
train_loader = DataLoader(train_data, batch_size=BATCH_SIZE, shuffle=True)
if not run_training:
return
# Finetuning
model.train(epochs=epochs, loader=train_loader, rate=lr_rate)
model.save(filename=ARCH.name.lower())
print(f"Model saved to trained_models/{ARCH.name.lower()}.pth")
print(f"Total test images for these {CLASS_SIZE} classes: {len(test_data)}")
# Evaluate original base checkpoint performance
current_mode = "Finetuned"
# evaluate finetuned model
try:
accuracy, report_dict = model.evaluate(loader=test_loader, mode=current_mode)
Util._log_to_csv(
arch=ARCH.name,#model.__class__.__name__,
mode=current_mode,
accuracy=accuracy,
report_dict=report_dict,
strategy="base"
)
except Exception as e:
print(f">> Skipping baseline log generation. Reason: {e}")
# saves evaluation metrics to log files
def log_metrics(evaluation_domains, reloaded, strategy_in_use):
# Process and append metrics to target reporting paths
for domain in evaluation_domains:
print(domain["label"])
accuracy, report_dict = reloaded.evaluate(loader=domain["loader"], mode=domain["mode"])
Util._log_to_csv(
arch=ARCH.name,
mode=domain["mode"],
accuracy=accuracy,
report_dict=report_dict,
strategy=strategy_in_use
)
# Unlearning and strategy eval
def run_unlearning_and_strategy_eval(env_dict, forget_class_idx, strategy, evaluate = False, suite_runner=None):
"""
Reloads a clean model state, applies the isolated unlearning framework,
and runs specific target evaluation domain checks.
"""
device = env_dict["device"]
train_data = env_dict["train_data"]
test_data = env_dict["test_data"]
# Segment specific unlearning loaders using class index boundaries
retain_train_loader , forget_train_loader= get_unlearning_loaders(
dataset=train_data, forget_class_idx=forget_class_idx, batch_size=BATCH_SIZE
)
retain_test_loader, forget_test_loader = get_unlearning_loaders(
dataset=test_data, forget_class_idx=forget_class_idx, batch_size=BATCH_SIZE
)
# Instantiate a clean copy of the model to keep weights isolated
reloaded = Model.create(arch=ARCH, device=device, size=CLASS_SIZE)
reloaded.load(arch=ARCH)
# Clean un-manipulated snapshot to serve as the Parameter-Space shadow proxy reference
shadow_model = copy.deepcopy(reloaded)
if evaluate:
reloaded.evaluate(
loader=retain_test_loader, mode="finetuned"
)
print("fine tunned model loaded into evaluation sandbox")
# Execute strategic parameter unlearning step
# we are using only training data to unlearn.
# Test data is never touched here.
unlearned = strategy.apply(reloaded.model, train_data)
strategy_in_use = strategy.__class__.__name__
if isinstance(unlearned,nn.Module):
reloaded.model = unlearned
else:
reloaded = unlearned
is_retrained = isinstance(strategy, Retrain)
if is_retrained:
os.makedirs("trained_models", exist_ok=True)
reloaded.save(filename=f"class_{forget_class_idx}_retrained.pth")
# here we add a condition conditional statement
if suite_runner is not None:
suite_runner.run_complete_evaluation(
framework_name=strategy_in_use,
target_class=forget_class_idx,
forget_loader=forget_train_loader,
retain_test_loader=forget_test_loader,
unlearned_instance=reloaded,
base_shadow_instance=shadow_model,
device=device
)
# Define validation tracking steps dynamically
evaluation_domains = [
{"loader": retain_test_loader, "mode": "retain", "label": "\n--- Performance on Retained Classes"},
{"loader": forget_test_loader, "mode": "forget", "label": "\n--- Performance on Forgotten Class"},
{"loader": forget_train_loader, "mode": "forget_train", "label": "\n--- Performance on Forgotten Class (Train Set - Verifying Unlearning)"}
]
log_metrics(evaluation_domains, reloaded, strategy_in_use)
# entry
if __name__ == "__main__":
outer_loop = 1
inner_loop = CLASS_SIZE
for k in range(outer_loop):
try:
# Data Infrastructure and Architecture
runtime_environment = prepare_data_and_model_environment()
# Baseline Evaluation
# switch finetuning for tests on strategies only,
# to avoid finetunning every time we test a strategy
finetuning = False
run_finetuning_or_baseline_eval(runtime_environment, run_training = finetuning)
# scale 16400.0 for ResNet
scale = 20100
# batch 8 for resNet,
unlearning_batches = 16
# regularis
# strategies
# implementation of Certified Removal for DNNs
certified_unlearning = CertifiedUnlearning(
target_class_index=0, #arch ResNet18 GoogLeNet Inception
l2_reg=0.000002 , # 0.000002 0.00001 0.0
gamma=0.01, # 0.1 0.1 0.01
scale= scale, # 16400.0 35000.0
s1=2, # 2
s2=350, # 300
std=0.00001, # 0.00001
unlearn_bs=unlearning_batches # 8 32 8
)
# Normalisation Filtration
linear_filtration = LinearFiltration(
target_class_index=0,
num_classes=CLASS_SIZE
)
# WF-Net
weight_filtration = WeightFiltration(
target_class_index=0, #arch ResNet18 GoogLeNet/Inception
epochs=6, #
lr=250.0, # ResNet18 = 150 # 150 100
gamma=0.001, # 0.001
lambda_1=30, # 25 100
arch=ARCH
)
retrain = Retrain(
target_class_index = 0,
arch = ARCH,
size = CLASS_SIZE,
lr = 0.0001,
epochs = 14
)
strategies = [
retrain,
linear_filtration,
weight_filtration,
certified_unlearning,
]
suite_runner = UnlearningAttack(arch=ARCH, class_size=CLASS_SIZE)
# Unlearning Iteration
for i in range(inner_loop):
for strategy in strategies:
# update target class to be unlearned
strategy.set_target_class(i)
print(f"Unlearning class {i} with {strategy.strategy_name}")
# forget
run_unlearning_and_strategy_eval(
runtime_environment,
forget_class_idx=i,
strategy=strategy,
# if we are finetuning, no need to evaluate base model.
# or may be never when not either!
evaluate = not finetuning,
suite_runner=suite_runner
)
# just a single class run before running all remaining classes.
#print(">> Single check run complete. Verification successful!")
#break
#dist_attacker.run_adversarial_evaluation()
#dist_attacker.run_incremental_evaluation(current_class_step=i)
#if suite_runner is not None:
#suite_runner.shutdown_hook()
except KeyboardInterrupt:
print("\nprogram interrupted. Exit!")
break

80
eval/UnlearningAttack.py
View File

@@ -39,6 +39,7 @@ class UnlearningAttack:
# JS Distance is the square root of JS Divergence # JS Distance is the square root of JS Divergence
return np.mean(jensenshannon(np.array(probs1), np.array(probs2), axis=1)) return np.mean(jensenshannon(np.array(probs1), np.array(probs2), axis=1))
def calculate_a_dist(self, latent1, latent2): def calculate_a_dist(self, latent1, latent2):
@@ -79,10 +80,29 @@ class UnlearningAttack:
np.array(all_losses).reshape(-1, 1) np.array(all_losses).reshape(-1, 1)
]) ])
def run_parameter_space_mia(self, unlearned_model, shadow_model, forget_loader, retain_test_loader, device, index): def run_parameter_space_mia(
self,
unlearned_model,
shadow_model,
forget_train_loader,
forget_test_loader,
device,
index
):
X_shadow_mem = self._extract_attack_features(shadow_model, forget_loader, device, index) X_shadow_mem = self._extract_attack_features(
X_shadow_non = self._extract_attack_features(shadow_model, retain_test_loader, device, index) shadow_model,
forget_train_loader,
device,
index
)
X_shadow_non = self._extract_attack_features(
shadow_model,
forget_test_loader,
device,
index
)
# Train MIA Classifier # Train MIA Classifier
min_train = min(len(X_shadow_mem), len(X_shadow_non)) min_train = min(len(X_shadow_mem), len(X_shadow_non))
@@ -93,8 +113,8 @@ class UnlearningAttack:
attack_classifier.fit(X_train, y_train) attack_classifier.fit(X_train, y_train)
# Evaluate MIA # Evaluate MIA
X_eval_mem = self._extract_attack_features(unlearned_model, forget_loader, device, index) X_eval_mem = self._extract_attack_features(unlearned_model, forget_train_loader, device, index)
X_eval_non = self._extract_attack_features(unlearned_model, retain_test_loader, device, index) X_eval_non = self._extract_attack_features(unlearned_model, forget_test_loader, device, index)
min_test = min(len(X_eval_mem), len(X_eval_non)) min_test = min(len(X_eval_mem), len(X_eval_non))
X_test = np.vstack([X_eval_mem[:min_test], X_eval_non[:min_test]]) X_test = np.vstack([X_eval_mem[:min_test], X_eval_non[:min_test]])
@@ -138,7 +158,15 @@ class UnlearningAttack:
# evaluate similarity of outputs # evaluate similarity of outputs
return accuracy_score(label_test, adversary.predict(data_test)) return accuracy_score(label_test, adversary.predict(data_test))
def run_logit_space_lookalike_mia(self, filtered_model, naive_retrained, forget_loader, device, target_class): def run_logit_space_lookalike_mia(
self,
filtered_model,
naive_retrained,
test_loader,
device,
target_class
):
filtered_model.eval() filtered_model.eval()
naive_retrained.eval() naive_retrained.eval()
@@ -146,7 +174,7 @@ class UnlearningAttack:
naive_logits = [] naive_logits = []
with torch.no_grad(): with torch.no_grad():
for data, _ in forget_loader: for data, _ in test_loader:
data = data.to(device) data = data.to(device)
if filtered_model.__class__.__name__ == "WF_Module": if filtered_model.__class__.__name__ == "WF_Module":
@@ -169,8 +197,20 @@ class UnlearningAttack:
# so that the metric is between 0 and 1. # so that the metric is between 0 and 1.
return 2.0 * np.abs(lookalike_accuracy - 0.5) return 2.0 * np.abs(lookalike_accuracy - 0.5)
def run_complete_evaluation(self, framework_name, target_class, forget_loader, retain_test_loader, unlearned_instance, base_shadow_instance, device): def run_complete_evaluation(
"""Orchestrates specific pipeline routing cleanly using cached constructor parameters.""" self,
framework_name,
target_class,
forget_train_loader,
forget_test_loader,
test_loader,
unlearned_instance,
base_shadow_instance,
device
):
# load from disk if saved model available
target_dir = os.path.join("reports", framework_name) target_dir = os.path.join("reports", framework_name)
os.makedirs(target_dir, exist_ok=True) os.makedirs(target_dir, exist_ok=True)
current_log_file = os.path.join(target_dir, "attack_values.csv") current_log_file = os.path.join(target_dir, "attack_values.csv")
@@ -179,12 +219,12 @@ class UnlearningAttack:
with open(current_log_file, "w") as f: with open(current_log_file, "w") as f:
f.write("target_class, parameter_mia_accuracy, latent_distance_tell, lookalike_accuracy, A-Dist, JS-Dist\n") f.write("target_class, parameter_mia_accuracy, latent_distance_tell, lookalike_accuracy, A-Dist, JS-Dist\n")
# 1. Parameter-Space MIA and Latent Distance # Parameter-Space MIA
parameter_mia_acc = self.run_parameter_space_mia( parameter_mia_acc = self.run_parameter_space_mia(
unlearned_model=unlearned_instance.model, unlearned_model=unlearned_instance.model,
shadow_model=base_shadow_instance.model, shadow_model=base_shadow_instance.model,
forget_loader=forget_loader, forget_train_loader=forget_train_loader,
retain_test_loader=retain_test_loader, forget_test_loader=forget_test_loader,
device=device, device=device,
index=target_class index=target_class
) )
@@ -206,17 +246,23 @@ class UnlearningAttack:
lookalike_acc = self.run_logit_space_lookalike_mia( lookalike_acc = self.run_logit_space_lookalike_mia(
filtered_model=unlearned_instance.model, filtered_model=unlearned_instance.model,
naive_retrained=reference_model_torch, naive_retrained=reference_model_torch,
forget_loader=forget_loader, test_loader=test_loader,
device=device, device=device,
target_class=target_class target_class=target_class
) )
# Calculate JS-Dist # Calculate JS-Dist
js_dist = self.calculate_js_dist(unlearned_instance.model, reference_model_torch, forget_loader, device, target_class) js_dist = self.calculate_js_dist(
unlearned_instance.model,
reference_model_torch,
forget_train_loader,
device,
target_class
)
# Extract features # Extract features
unlearned_features = self._extract_attack_features(unlearned_instance.model, forget_loader, device, target_class) unlearned_features = self._extract_attack_features(unlearned_instance.model, forget_train_loader, device, target_class)
retrained_features = self._extract_attack_features(reference_model_torch, forget_loader, device, target_class) retrained_features = self._extract_attack_features(reference_model_torch, forget_train_loader, device, target_class)
# Calculate A-Dist using these features # Calculate A-Dist using these features
a_dist = self.calculate_a_dist(unlearned_features, retrained_features) a_dist = self.calculate_a_dist(unlearned_features, retrained_features)
@@ -225,7 +271,7 @@ class UnlearningAttack:
print(f"[{framework_name}] Class {target_class} | Parameter MIA: {parameter_mia_acc:.4f} Lookalike: {lookalike_acc:.4f}" ) print(f"[{framework_name}] Class {target_class} | Parameter MIA: {parameter_mia_acc:.4f} Lookalike: {lookalike_acc:.4f}" )
with open(current_log_file, "a") as f: with open(current_log_file, "a") as f:
f.write(f"{target_class},{parameter_mia_acc:.6f},{0.00000},{lookalike_acc:.6f}, {a_dist:.6f}, {js_dist:.6f}\n") f.write(f"{target_class},{parameter_mia_acc:.6f},{lookalike_acc:.6f}, {a_dist:.6f}, {js_dist:.6f}\n")
return { return {
"parameter_mia_accuracy": parameter_mia_acc, "parameter_mia_accuracy": parameter_mia_acc,

54
reports/CertifiedUnlearning/RESNET34/forget.csv
View File

@@ -376,3 +376,57 @@ accuracy,macro_precision,macro_recall,macro_f1,weighted_precision,weighted_recal
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.1750,1.0000,0.1750,0.2979,1.0000,0.1750,0.2979
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.1250,1.0000,0.1250,0.2222,1.0000,0.1250,0.2222
0.1000,1.0000,0.1000,0.1818,1.0000,0.1000,0.1818
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0125,1.0000,0.0125,0.0247,1.0000,0.0125,0.0247
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.5750,1.0000,0.5750,0.7302,1.0000,0.5750,0.7302
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.6000,1.0000,0.6000,0.7500,1.0000,0.6000,0.7500
0.1125,1.0000,0.1125,0.2022,1.0000,0.1125,0.2022
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.2375,1.0000,0.2375,0.3838,1.0000,0.2375,0.3838
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.4500,1.0000,0.4500,0.6207,1.0000,0.4500,0.6207
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.3250,1.0000,0.3250,0.4906,1.0000,0.3250,0.4906
0.0625,1.0000,0.0625,0.1176,1.0000,0.0625,0.1176
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.4375,1.0000,0.4375,0.6087,1.0000,0.4375,0.6087
0.7125,1.0000,0.7125,0.8321,1.0000,0.7125,0.8321
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.3000,1.0000,0.3000,0.4615,1.0000,0.3000,0.4615
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0250,1.0000,0.0250,0.0488,1.0000,0.0250,0.0488
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
1 accuracy macro_precision macro_recall macro_f1 weighted_precision weighted_recall weighted_f1
376 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
377 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
378 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
379 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
380 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
381 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
382 0.1750 1.0000 0.1750 0.2979 1.0000 0.1750 0.2979
383 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
384 0.1250 1.0000 0.1250 0.2222 1.0000 0.1250 0.2222
385 0.1000 1.0000 0.1000 0.1818 1.0000 0.1000 0.1818
386 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
387 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
388 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
389 0.0125 1.0000 0.0125 0.0247 1.0000 0.0125 0.0247
390 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
391 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
392 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
393 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
394 0.5750 1.0000 0.5750 0.7302 1.0000 0.5750 0.7302
395 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
396 0.6000 1.0000 0.6000 0.7500 1.0000 0.6000 0.7500
397 0.1125 1.0000 0.1125 0.2022 1.0000 0.1125 0.2022
398 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
399 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
400 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
401 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
402 0.2375 1.0000 0.2375 0.3838 1.0000 0.2375 0.3838
403 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
404 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
405 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
406 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
407 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
408 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
409 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
410 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
411 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
412 0.4500 1.0000 0.4500 0.6207 1.0000 0.4500 0.6207
413 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
414 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
415 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
416 0.3250 1.0000 0.3250 0.4906 1.0000 0.3250 0.4906
417 0.0625 1.0000 0.0625 0.1176 1.0000 0.0625 0.1176
418 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
419 0.4375 1.0000 0.4375 0.6087 1.0000 0.4375 0.6087
420 0.7125 1.0000 0.7125 0.8321 1.0000 0.7125 0.8321
421 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
422 0.3000 1.0000 0.3000 0.4615 1.0000 0.3000 0.4615
423 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
424 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
425 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
426 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
427 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
428 0.0250 1.0000 0.0250 0.0488 1.0000 0.0250 0.0488
429 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
430 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
431 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
432 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000

54
reports/CertifiedUnlearning/RESNET34/forget_train.csv
View File

@@ -376,3 +376,57 @@ accuracy,macro_precision,macro_recall,macro_f1,weighted_precision,weighted_recal
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.1594,1.0000,0.1594,0.2749,1.0000,0.1594,0.2749
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.1094,1.0000,0.1094,0.1972,1.0000,0.1094,0.1972
0.1781,1.0000,0.1781,0.3024,1.0000,0.1781,0.3024
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0125,1.0000,0.0125,0.0247,1.0000,0.0125,0.0247
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.7344,1.0000,0.7344,0.8468,1.0000,0.7344,0.8468
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.7844,1.0000,0.7844,0.8792,1.0000,0.7844,0.8792
0.1125,1.0000,0.1125,0.2022,1.0000,0.1125,0.2022
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.1812,1.0000,0.1812,0.3069,1.0000,0.1812,0.3069
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.6594,1.0000,0.6594,0.7947,1.0000,0.6594,0.7947
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.3812,1.0000,0.3812,0.5520,1.0000,0.3812,0.5520
0.1062,1.0000,0.1062,0.1921,1.0000,0.1062,0.1921
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.4969,1.0000,0.4969,0.6639,1.0000,0.4969,0.6639
0.9156,1.0000,0.9156,0.9560,1.0000,0.9156,0.9560
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.3375,1.0000,0.3375,0.5047,1.0000,0.3375,0.5047
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0094,1.0000,0.0094,0.0186,1.0000,0.0094,0.0186
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
1 accuracy macro_precision macro_recall macro_f1 weighted_precision weighted_recall weighted_f1
376 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
377 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
378 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
379 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
380 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
381 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
382 0.1594 1.0000 0.1594 0.2749 1.0000 0.1594 0.2749
383 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
384 0.1094 1.0000 0.1094 0.1972 1.0000 0.1094 0.1972
385 0.1781 1.0000 0.1781 0.3024 1.0000 0.1781 0.3024
386 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
387 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
388 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
389 0.0125 1.0000 0.0125 0.0247 1.0000 0.0125 0.0247
390 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
391 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
392 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
393 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
394 0.7344 1.0000 0.7344 0.8468 1.0000 0.7344 0.8468
395 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
396 0.7844 1.0000 0.7844 0.8792 1.0000 0.7844 0.8792
397 0.1125 1.0000 0.1125 0.2022 1.0000 0.1125 0.2022
398 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
399 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
400 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
401 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
402 0.1812 1.0000 0.1812 0.3069 1.0000 0.1812 0.3069
403 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
404 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
405 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
406 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
407 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
408 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
409 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
410 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
411 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
412 0.6594 1.0000 0.6594 0.7947 1.0000 0.6594 0.7947
413 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
414 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
415 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
416 0.3812 1.0000 0.3812 0.5520 1.0000 0.3812 0.5520
417 0.1062 1.0000 0.1062 0.1921 1.0000 0.1062 0.1921
418 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
419 0.4969 1.0000 0.4969 0.6639 1.0000 0.4969 0.6639
420 0.9156 1.0000 0.9156 0.9560 1.0000 0.9156 0.9560
421 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
422 0.3375 1.0000 0.3375 0.5047 1.0000 0.3375 0.5047
423 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
424 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
425 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
426 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
427 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
428 0.0094 1.0000 0.0094 0.0186 1.0000 0.0094 0.0186
429 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
430 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
431 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
432 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000

54
reports/CertifiedUnlearning/RESNET34/retain.csv
View File

@@ -376,3 +376,57 @@ accuracy,macro_precision,macro_recall,macro_f1,weighted_precision,weighted_recal
0.9053,0.9214,0.9053,0.9081,0.9214,0.9053,0.9081 0.9053,0.9214,0.9053,0.9081,0.9214,0.9053,0.9081
0.0533,0.0343,0.0533,0.0077,0.0343,0.0533,0.0077 0.0533,0.0343,0.0533,0.0077,0.0343,0.0533,0.0077
0.7171,0.8599,0.7171,0.7316,0.8599,0.7171,0.7316 0.7171,0.8599,0.7171,0.7316,0.8599,0.7171,0.7316
0.9230,0.9359,0.9230,0.9258,0.9359,0.9230,0.9258
0.8842,0.9131,0.8842,0.8898,0.9131,0.8842,0.8898
0.5572,0.8343,0.5572,0.5496,0.8343,0.5572,0.5496
0.9395,0.9433,0.9395,0.9400,0.9433,0.9395,0.9400
0.4888,0.8411,0.4888,0.4827,0.8411,0.4888,0.4827
0.9408,0.9432,0.9408,0.9412,0.9432,0.9408,0.9412
0.9329,0.9414,0.9329,0.9341,0.9414,0.9329,0.9341
0.9362,0.9418,0.9362,0.9369,0.9418,0.9362,0.9369
0.9461,0.9486,0.9461,0.9462,0.9486,0.9461,0.9462
0.0546,0.0554,0.0546,0.0091,0.0554,0.0546,0.0091
0.9480,0.9517,0.9480,0.9486,0.9517,0.9480,0.9486
0.9276,0.9364,0.9276,0.9288,0.9364,0.9276,0.9288
0.9230,0.9317,0.9230,0.9234,0.9317,0.9230,0.9234
0.8934,0.9175,0.8934,0.8983,0.9175,0.8934,0.8983
0.6401,0.8488,0.6401,0.6484,0.8488,0.6401,0.6484
0.9474,0.9493,0.9474,0.9476,0.9493,0.9474,0.9476
0.1704,0.2866,0.1704,0.0875,0.2866,0.1704,0.0875
0.9480,0.9498,0.9480,0.9483,0.9498,0.9480,0.9483
0.9428,0.9470,0.9428,0.9431,0.9470,0.9428,0.9431
0.8059,0.8877,0.8059,0.8163,0.8877,0.8059,0.8163
0.9513,0.9531,0.9513,0.9515,0.9531,0.9513,0.9515
0.2480,0.4844,0.2480,0.2043,0.4844,0.2480,0.2043
0.8934,0.9220,0.8934,0.8995,0.9220,0.8934,0.8995
0.9447,0.9494,0.9447,0.9455,0.9494,0.9447,0.9455
0.9355,0.9410,0.9355,0.9364,0.9410,0.9355,0.9364
0.8836,0.9200,0.8836,0.8914,0.9200,0.8836,0.8914
0.1046,0.1952,0.1046,0.0543,0.1952,0.1046,0.0543
0.8441,0.8947,0.8441,0.8513,0.8947,0.8441,0.8513
0.9125,0.9234,0.9125,0.9135,0.9234,0.9125,0.9135
0.4151,0.7134,0.4151,0.3699,0.7134,0.4151,0.3699
0.9454,0.9500,0.9454,0.9463,0.9500,0.9454,0.9463
0.9375,0.9438,0.9375,0.9386,0.9438,0.9375,0.9386
0.7559,0.8854,0.7559,0.7774,0.8854,0.7559,0.7774
0.9421,0.9467,0.9421,0.9427,0.9467,0.9421,0.9427
0.8013,0.8818,0.8013,0.8134,0.8818,0.8013,0.8134
0.8132,0.9213,0.8132,0.8447,0.9213,0.8132,0.8447
0.5289,0.8430,0.5289,0.5306,0.8430,0.5289,0.5306
0.9474,0.9493,0.9474,0.9477,0.9493,0.9474,0.9477
0.9368,0.9410,0.9368,0.9371,0.9410,0.9368,0.9371
0.9086,0.9250,0.9086,0.9116,0.9250,0.9086,0.9116
0.9500,0.9517,0.9500,0.9503,0.9517,0.9500,0.9503
0.9513,0.9533,0.9513,0.9516,0.9533,0.9513,0.9516
0.9237,0.9333,0.9237,0.9255,0.9333,0.9237,0.9255
0.9447,0.9472,0.9447,0.9451,0.9472,0.9447,0.9451
0.8316,0.9040,0.8316,0.8458,0.9040,0.8316,0.8458
0.8941,0.9206,0.8941,0.8978,0.9206,0.8941,0.8978
0.8974,0.9165,0.8974,0.8989,0.9165,0.8974,0.8989
0.5586,0.8841,0.5586,0.5802,0.8841,0.5586,0.5802
0.8382,0.9133,0.8382,0.8565,0.9133,0.8382,0.8565
0.9428,0.9483,0.9428,0.9437,0.9483,0.9428,0.9437
0.9086,0.9200,0.9086,0.9094,0.9200,0.9086,0.9094
0.7046,0.8702,0.7046,0.7201,0.8702,0.7046,0.7201
0.0789,0.0145,0.0789,0.0229,0.0145,0.0789,0.0229
0.8559,0.9075,0.8559,0.8674,0.9075,0.8559,0.8674
1 accuracy macro_precision macro_recall macro_f1 weighted_precision weighted_recall weighted_f1
376 0.9053 0.9214 0.9053 0.9081 0.9214 0.9053 0.9081
377 0.0533 0.0343 0.0533 0.0077 0.0343 0.0533 0.0077
378 0.7171 0.8599 0.7171 0.7316 0.8599 0.7171 0.7316
379 0.9230 0.9359 0.9230 0.9258 0.9359 0.9230 0.9258
380 0.8842 0.9131 0.8842 0.8898 0.9131 0.8842 0.8898
381 0.5572 0.8343 0.5572 0.5496 0.8343 0.5572 0.5496
382 0.9395 0.9433 0.9395 0.9400 0.9433 0.9395 0.9400
383 0.4888 0.8411 0.4888 0.4827 0.8411 0.4888 0.4827
384 0.9408 0.9432 0.9408 0.9412 0.9432 0.9408 0.9412
385 0.9329 0.9414 0.9329 0.9341 0.9414 0.9329 0.9341
386 0.9362 0.9418 0.9362 0.9369 0.9418 0.9362 0.9369
387 0.9461 0.9486 0.9461 0.9462 0.9486 0.9461 0.9462
388 0.0546 0.0554 0.0546 0.0091 0.0554 0.0546 0.0091
389 0.9480 0.9517 0.9480 0.9486 0.9517 0.9480 0.9486
390 0.9276 0.9364 0.9276 0.9288 0.9364 0.9276 0.9288
391 0.9230 0.9317 0.9230 0.9234 0.9317 0.9230 0.9234
392 0.8934 0.9175 0.8934 0.8983 0.9175 0.8934 0.8983
393 0.6401 0.8488 0.6401 0.6484 0.8488 0.6401 0.6484
394 0.9474 0.9493 0.9474 0.9476 0.9493 0.9474 0.9476
395 0.1704 0.2866 0.1704 0.0875 0.2866 0.1704 0.0875
396 0.9480 0.9498 0.9480 0.9483 0.9498 0.9480 0.9483
397 0.9428 0.9470 0.9428 0.9431 0.9470 0.9428 0.9431
398 0.8059 0.8877 0.8059 0.8163 0.8877 0.8059 0.8163
399 0.9513 0.9531 0.9513 0.9515 0.9531 0.9513 0.9515
400 0.2480 0.4844 0.2480 0.2043 0.4844 0.2480 0.2043
401 0.8934 0.9220 0.8934 0.8995 0.9220 0.8934 0.8995
402 0.9447 0.9494 0.9447 0.9455 0.9494 0.9447 0.9455
403 0.9355 0.9410 0.9355 0.9364 0.9410 0.9355 0.9364
404 0.8836 0.9200 0.8836 0.8914 0.9200 0.8836 0.8914
405 0.1046 0.1952 0.1046 0.0543 0.1952 0.1046 0.0543
406 0.8441 0.8947 0.8441 0.8513 0.8947 0.8441 0.8513
407 0.9125 0.9234 0.9125 0.9135 0.9234 0.9125 0.9135
408 0.4151 0.7134 0.4151 0.3699 0.7134 0.4151 0.3699
409 0.9454 0.9500 0.9454 0.9463 0.9500 0.9454 0.9463
410 0.9375 0.9438 0.9375 0.9386 0.9438 0.9375 0.9386
411 0.7559 0.8854 0.7559 0.7774 0.8854 0.7559 0.7774
412 0.9421 0.9467 0.9421 0.9427 0.9467 0.9421 0.9427
413 0.8013 0.8818 0.8013 0.8134 0.8818 0.8013 0.8134
414 0.8132 0.9213 0.8132 0.8447 0.9213 0.8132 0.8447
415 0.5289 0.8430 0.5289 0.5306 0.8430 0.5289 0.5306
416 0.9474 0.9493 0.9474 0.9477 0.9493 0.9474 0.9477
417 0.9368 0.9410 0.9368 0.9371 0.9410 0.9368 0.9371
418 0.9086 0.9250 0.9086 0.9116 0.9250 0.9086 0.9116
419 0.9500 0.9517 0.9500 0.9503 0.9517 0.9500 0.9503
420 0.9513 0.9533 0.9513 0.9516 0.9533 0.9513 0.9516
421 0.9237 0.9333 0.9237 0.9255 0.9333 0.9237 0.9255
422 0.9447 0.9472 0.9447 0.9451 0.9472 0.9447 0.9451
423 0.8316 0.9040 0.8316 0.8458 0.9040 0.8316 0.8458
424 0.8941 0.9206 0.8941 0.8978 0.9206 0.8941 0.8978
425 0.8974 0.9165 0.8974 0.8989 0.9165 0.8974 0.8989
426 0.5586 0.8841 0.5586 0.5802 0.8841 0.5586 0.5802
427 0.8382 0.9133 0.8382 0.8565 0.9133 0.8382 0.8565
428 0.9428 0.9483 0.9428 0.9437 0.9483 0.9428 0.9437
429 0.9086 0.9200 0.9086 0.9094 0.9200 0.9086 0.9094
430 0.7046 0.8702 0.7046 0.7201 0.8702 0.7046 0.7201
431 0.0789 0.0145 0.0789 0.0229 0.0145 0.0789 0.0229
432 0.8559 0.9075 0.8559 0.8674 0.9075 0.8559 0.8674

70
reports/CertifiedUnlearning/attack_values.csv
View File

@@ -1,33 +1,37 @@
target_class,parameter_mia_accuracy,latent_distance_tell,lookalike_accuracy target_class, parameter_mia_accuracy, lookalike_accuracy, A-Dist, JS-Dist
0,0.500000,7.219862,0.979167 0,0.500000,1.000000, 0.062500, 0.654685
1,0.500000,3.659238,0.958333 1,0.500000,0.979167, 0.208333, 0.516170
2,0.500000,6.939345,0.885417 2,0.500000,0.958333, 0.187500, 0.526930
3,0.500000,7.800516,0.989583 3,0.500000,1.000000, 0.104167, 0.473445
4,0.500000,6.631370,1.000000 4,0.500000,0.979167, 0.312500, 0.571769
5,0.500000,7.557893,0.937500 5,0.500000,1.000000, 0.072917, 0.672217
6,0.500000,8.085600,0.937500 6,0.500000,0.958333, 0.041667, 0.513104
7,0.500000,7.518537,0.947917 7,0.500000,0.989583, 0.020833, 0.576437
8,0.500000,9.040818,0.895833 8,0.500000,0.875000, 0.052083, 0.491356
9,0.500000,8.269253,0.968750 9,0.500000,0.968750, 0.083333, 0.611189
10,0.500000,5.678658,1.000000 10,0.500000,1.000000, 0.052083, 0.615587
11,0.450000,6.800340,1.000000 11,0.500000,1.000000, 0.031250, 0.545611
12,0.500000,7.708953,0.937500 12,0.500000,0.968750, 0.093750, 0.434476
13,0.500000,6.559650,1.000000 13,0.500000,1.000000, 0.145833, 0.582930
14,0.500000,8.420962,0.989583 14,0.500000,0.916667, 0.083333, 0.471939
15,0.500000,7.526255,1.000000 15,0.500000,0.989583, 0.083333, 0.541379
16,0.500000,6.906671,0.968750 16,0.500000,0.989583, 0.218750, 0.571814
17,0.500000,5.499599,0.979167 17,0.500000,0.989583, 0.062500, 0.503747
18,0.500000,8.465704,0.979167 18,0.550000,1.000000, 0.166667, 0.495459
19,0.500000,5.958429,0.979167 19,0.500000,0.916667, 0.052083, 0.551996
0,0.500000,5.903083,0.979167
1,0.500000,4.479048,0.947917 0,0.500000,0.937500, 0.020833, 0.532267
2,0.500000,4.460630,0.979167 1,0.500000,0.937500, 0.093750, 0.503973
3,0.500000,9.916990,1.000000 2,0.500000,0.989583, 0.104167, 0.570063
4,0.500000,4.245532,0.979167 0,0.500000,0.402083, 0.031250, 0.511158
5,0.500000,5.771674,1.000000 1,0.500000,0.154167, 0.312500, 0.496193
6,0.500000,6.300947,0.979167 2,0.500000,0.602083, 0.145833, 0.530453
7,0.500000,6.803962,0.979167 3,0.500000,0.852083, 0.135417, 0.491837
8,0.500000,9.174347,1.000000 4,0.500000,0.654167, 0.041667, 0.569679
9,0.500000,7.843153,0.989583 5,0.500000,0.993750, 0.187500, 0.693986
10,0.500000,5.029855,1.000000 6,0.500000,0.489583, 0.010417, 0.549827
11,0.500000,7.945799,1.000000 7,0.500000,0.333333, 0.156250, 0.539906
8,0.500000,0.668750, 0.041667, 0.477993
9,0.500000,0.962500, 0.062500, 0.622796
10,0.500000,1.000000, 0.093750, 0.712293
11,0.500000,0.939583, 0.187500, 0.511184
1 target_class parameter_mia_accuracy latent_distance_tell lookalike_accuracy A-Dist JS-Dist
2 0 0.500000 7.219862 0.979167 1.000000 0.062500 0.654685
3 1 0.500000 3.659238 0.958333 0.979167 0.208333 0.516170
4 2 0.500000 6.939345 0.885417 0.958333 0.187500 0.526930
5 3 0.500000 7.800516 0.989583 1.000000 0.104167 0.473445
6 4 0.500000 6.631370 1.000000 0.979167 0.312500 0.571769
7 5 0.500000 7.557893 0.937500 1.000000 0.072917 0.672217
8 6 0.500000 8.085600 0.937500 0.958333 0.041667 0.513104
9 7 0.500000 7.518537 0.947917 0.989583 0.020833 0.576437
10 8 0.500000 9.040818 0.895833 0.875000 0.052083 0.491356
11 9 0.500000 8.269253 0.968750 0.083333 0.611189
12 10 0.500000 5.678658 1.000000 0.052083 0.615587
13 11 0.450000 0.500000 6.800340 1.000000 0.031250 0.545611
14 12 0.500000 7.708953 0.937500 0.968750 0.093750 0.434476
15 13 0.500000 6.559650 1.000000 0.145833 0.582930
16 14 0.500000 8.420962 0.989583 0.916667 0.083333 0.471939
17 15 0.500000 7.526255 1.000000 0.989583 0.083333 0.541379
18 16 0.500000 6.906671 0.968750 0.989583 0.218750 0.571814
19 17 0.500000 5.499599 0.979167 0.989583 0.062500 0.503747
20 18 0.500000 0.550000 8.465704 0.979167 1.000000 0.166667 0.495459
21 19 0.500000 5.958429 0.979167 0.916667 0.052083 0.551996
22 0 0.500000 5.903083 0.979167 0.937500 0.020833 0.532267
23 1 0.500000 4.479048 0.947917 0.937500 0.093750 0.503973
24 2 0.500000 4.460630 0.979167 0.989583 0.104167 0.570063
25 3 0 0.500000 9.916990 1.000000 0.402083 0.031250 0.511158
26 4 1 0.500000 4.245532 0.979167 0.154167 0.312500 0.496193
27 5 2 0.500000 5.771674 1.000000 0.602083 0.145833 0.530453
28 6 3 0.500000 6.300947 0.979167 0.852083 0.135417 0.491837
29 7 4 0.500000 6.803962 0.979167 0.654167 0.041667 0.569679
30 8 5 0.500000 9.174347 1.000000 0.993750 0.187500 0.693986
31 9 6 0.500000 7.843153 0.989583 0.489583 0.010417 0.549827
32 10 7 0.500000 5.029855 1.000000 0.333333 0.156250 0.539906
33 11 8 0.500000 7.945799 1.000000 0.668750 0.041667 0.477993
34 9 0.500000 0.962500 0.062500 0.622796
35 10 0.500000 1.000000 0.093750 0.712293
36 11 0.500000 0.939583 0.187500 0.511184
37

34
reports/CertifiedUnlearning/attack_values_int.csv Normal file
View File

@@ -0,0 +1,34 @@
target_class,parameter_mia_accuracy,latent_distance_tell,lookalike_accuracy
0,0.500000,7.219862,0.979167
1,0.500000,3.659238,0.958333
2,0.500000,6.939345,0.885417
3,0.500000,7.800516,0.989583
4,0.500000,6.631370,1.000000
5,0.500000,7.557893,0.937500
6,0.500000,8.085600,0.937500
7,0.500000,7.518537,0.947917
8,0.500000,9.040818,0.895833
9,0.500000,8.269253,0.968750
10,0.500000,5.678658,1.000000
11,0.450000,6.800340,1.000000
12,0.500000,7.708953,0.937500
13,0.500000,6.559650,1.000000
14,0.500000,8.420962,0.989583
15,0.500000,7.526255,1.000000
16,0.500000,6.906671,0.968750
17,0.500000,5.499599,0.979167
18,0.500000,8.465704,0.979167
19,0.500000,5.958429,0.979167
0,0.500000,5.903083,0.979167
1,0.500000,4.479048,0.947917
2,0.500000,4.460630,0.979167
3,0.500000,9.916990,1.000000
4,0.500000,4.245532,0.979167
5,0.500000,5.771674,1.000000
6,0.500000,6.300947,0.979167
7,0.500000,6.803962,0.979167
8,0.500000,9.174347,1.000000
9,0.500000,7.843153,0.989583
10,0.500000,5.029855,1.000000
11,0.500000,7.945799,1.000000
12,0.500000,8.373040,0.979167
1 target_class parameter_mia_accuracy latent_distance_tell lookalike_accuracy
2 0 0.500000 7.219862 0.979167
3 1 0.500000 3.659238 0.958333
4 2 0.500000 6.939345 0.885417
5 3 0.500000 7.800516 0.989583
6 4 0.500000 6.631370 1.000000
7 5 0.500000 7.557893 0.937500
8 6 0.500000 8.085600 0.937500
9 7 0.500000 7.518537 0.947917
10 8 0.500000 9.040818 0.895833
11 9 0.500000 8.269253 0.968750
12 10 0.500000 5.678658 1.000000
13 11 0.450000 6.800340 1.000000
14 12 0.500000 7.708953 0.937500
15 13 0.500000 6.559650 1.000000
16 14 0.500000 8.420962 0.989583
17 15 0.500000 7.526255 1.000000
18 16 0.500000 6.906671 0.968750
19 17 0.500000 5.499599 0.979167
20 18 0.500000 8.465704 0.979167
21 19 0.500000 5.958429 0.979167
22 0 0.500000 5.903083 0.979167
23 1 0.500000 4.479048 0.947917
24 2 0.500000 4.460630 0.979167
25 3 0.500000 9.916990 1.000000
26 4 0.500000 4.245532 0.979167
27 5 0.500000 5.771674 1.000000
28 6 0.500000 6.300947 0.979167
29 7 0.500000 6.803962 0.979167
30 8 0.500000 9.174347 1.000000
31 9 0.500000 7.843153 0.989583
32 10 0.500000 5.029855 1.000000
33 11 0.500000 7.945799 1.000000
34 12 0.500000 8.373040 0.979167

61
reports/LinearFiltration/RESNET34/forget.csv
View File

@@ -450,3 +450,64 @@ accuracy,macro_precision,macro_recall,macro_f1,weighted_precision,weighted_recal
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
1 accuracy macro_precision macro_recall macro_f1 weighted_precision weighted_recall weighted_f1
450 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
451 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
452 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
453 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
454 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
455 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
456 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
457 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
458 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
459 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
460 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
461 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
462 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
463 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
464 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
465 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
466 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
467 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
468 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
469 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
470 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
471 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
472 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
473 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
474 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
475 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
476 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
477 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
478 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
479 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
480 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
481 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
482 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
483 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
484 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
485 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
486 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
487 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
488 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
489 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
490 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
491 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
492 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
493 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
494 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
495 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
496 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
497 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
498 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
499 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
500 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
501 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
502 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
503 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
504 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
505 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
506 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
507 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
508 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
509 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
510 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
511 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
512 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
513 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000

61
reports/LinearFiltration/RESNET34/forget_train.csv
View File

@@ -450,3 +450,64 @@ accuracy,macro_precision,macro_recall,macro_f1,weighted_precision,weighted_recal
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
1 accuracy macro_precision macro_recall macro_f1 weighted_precision weighted_recall weighted_f1
450 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
451 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
452 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
453 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
454 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
455 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
456 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
457 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
458 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
459 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
460 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
461 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
462 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
463 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
464 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
465 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
466 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
467 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
468 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
469 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
470 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
471 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
472 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
473 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
474 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
475 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
476 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
477 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
478 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
479 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
480 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
481 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
482 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
483 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
484 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
485 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
486 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
487 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
488 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
489 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
490 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
491 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
492 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
493 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
494 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
495 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
496 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
497 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
498 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
499 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
500 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
501 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
502 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
503 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
504 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
505 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
506 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
507 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
508 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
509 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
510 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
511 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
512 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
513 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000

61
reports/LinearFiltration/RESNET34/retain.csv
View File

@@ -450,3 +450,64 @@ accuracy,macro_precision,macro_recall,macro_f1,weighted_precision,weighted_recal
0.9533,0.9547,0.9533,0.9535,0.9547,0.9533,0.9535 0.9533,0.9547,0.9533,0.9535,0.9547,0.9533,0.9535
0.9546,0.9561,0.9546,0.9549,0.9561,0.9546,0.9549 0.9546,0.9561,0.9546,0.9549,0.9561,0.9546,0.9549
0.9513,0.9531,0.9513,0.9516,0.9531,0.9513,0.9516 0.9513,0.9531,0.9513,0.9516,0.9531,0.9513,0.9516
0.9526,0.9541,0.9526,0.9529,0.9541,0.9526,0.9529
0.9533,0.9551,0.9533,0.9536,0.9551,0.9533,0.9536
0.9526,0.9541,0.9526,0.9529,0.9541,0.9526,0.9529
0.9533,0.9551,0.9533,0.9536,0.9551,0.9533,0.9536
0.9533,0.9545,0.9533,0.9535,0.9545,0.9533,0.9535
0.9526,0.9541,0.9526,0.9529,0.9541,0.9526,0.9529
0.9533,0.9551,0.9533,0.9536,0.9551,0.9533,0.9536
0.9533,0.9545,0.9533,0.9535,0.9545,0.9533,0.9535
0.9546,0.9562,0.9546,0.9548,0.9562,0.9546,0.9548
0.9553,0.9570,0.9553,0.9555,0.9570,0.9553,0.9555
0.9566,0.9581,0.9566,0.9568,0.9581,0.9566,0.9568
0.9553,0.9568,0.9553,0.9555,0.9568,0.9553,0.9555
0.9513,0.9532,0.9513,0.9516,0.9532,0.9513,0.9516
0.9539,0.9556,0.9539,0.9542,0.9556,0.9539,0.9542
0.9520,0.9539,0.9520,0.9523,0.9539,0.9520,0.9523
0.9533,0.9547,0.9533,0.9535,0.9547,0.9533,0.9535
0.9526,0.9541,0.9526,0.9529,0.9541,0.9526,0.9529
0.9533,0.9551,0.9533,0.9536,0.9551,0.9533,0.9536
0.9533,0.9545,0.9533,0.9535,0.9545,0.9533,0.9535
0.9526,0.9541,0.9526,0.9529,0.9541,0.9526,0.9529
0.9533,0.9551,0.9533,0.9536,0.9551,0.9533,0.9536
0.9533,0.9545,0.9533,0.9535,0.9545,0.9533,0.9535
0.9546,0.9562,0.9546,0.9548,0.9562,0.9546,0.9548
0.9553,0.9570,0.9553,0.9555,0.9570,0.9553,0.9555
0.9566,0.9581,0.9566,0.9568,0.9581,0.9566,0.9568
0.9553,0.9568,0.9553,0.9555,0.9568,0.9553,0.9555
0.9513,0.9532,0.9513,0.9516,0.9532,0.9513,0.9516
0.9539,0.9556,0.9539,0.9542,0.9556,0.9539,0.9542
0.9520,0.9539,0.9520,0.9523,0.9539,0.9520,0.9523
0.9533,0.9547,0.9533,0.9535,0.9547,0.9533,0.9535
0.9546,0.9561,0.9546,0.9549,0.9561,0.9546,0.9549
0.9513,0.9531,0.9513,0.9516,0.9531,0.9513,0.9516
0.9533,0.9549,0.9533,0.9535,0.9549,0.9533,0.9535
0.9559,0.9570,0.9559,0.9561,0.9570,0.9559,0.9561
0.9513,0.9532,0.9513,0.9516,0.9532,0.9513,0.9516
0.9513,0.9531,0.9513,0.9516,0.9531,0.9513,0.9516
0.9520,0.9538,0.9520,0.9523,0.9538,0.9520,0.9523
0.9539,0.9554,0.9539,0.9542,0.9554,0.9539,0.9542
0.9526,0.9544,0.9526,0.9529,0.9544,0.9526,0.9529
0.9526,0.9541,0.9526,0.9529,0.9541,0.9526,0.9529
0.9533,0.9551,0.9533,0.9536,0.9551,0.9533,0.9536
0.9533,0.9545,0.9533,0.9535,0.9545,0.9533,0.9535
0.9546,0.9562,0.9546,0.9548,0.9562,0.9546,0.9548
0.9526,0.9541,0.9526,0.9529,0.9541,0.9526,0.9529
0.9526,0.9541,0.9526,0.9529,0.9541,0.9526,0.9529
0.9533,0.9551,0.9533,0.9536,0.9551,0.9533,0.9536
0.9533,0.9545,0.9533,0.9535,0.9545,0.9533,0.9535
0.9546,0.9562,0.9546,0.9548,0.9562,0.9546,0.9548
0.9526,0.9541,0.9526,0.9529,0.9541,0.9526,0.9529
0.9533,0.9551,0.9533,0.9536,0.9551,0.9533,0.9536
0.9533,0.9545,0.9533,0.9535,0.9545,0.9533,0.9535
0.9546,0.9562,0.9546,0.9548,0.9562,0.9546,0.9548
0.9553,0.9570,0.9553,0.9555,0.9570,0.9553,0.9555
0.9566,0.9581,0.9566,0.9568,0.9581,0.9566,0.9568
0.9553,0.9568,0.9553,0.9555,0.9568,0.9553,0.9555
0.9513,0.9532,0.9513,0.9516,0.9532,0.9513,0.9516
0.9539,0.9556,0.9539,0.9542,0.9556,0.9539,0.9542
0.9520,0.9539,0.9520,0.9523,0.9539,0.9520,0.9523
0.9533,0.9547,0.9533,0.9535,0.9547,0.9533,0.9535
0.9546,0.9561,0.9546,0.9549,0.9561,0.9546,0.9549
0.9513,0.9531,0.9513,0.9516,0.9531,0.9513,0.9516
1 accuracy macro_precision macro_recall macro_f1 weighted_precision weighted_recall weighted_f1
450 0.9533 0.9547 0.9533 0.9535 0.9547 0.9533 0.9535
451 0.9546 0.9561 0.9546 0.9549 0.9561 0.9546 0.9549
452 0.9513 0.9531 0.9513 0.9516 0.9531 0.9513 0.9516
453 0.9526 0.9541 0.9526 0.9529 0.9541 0.9526 0.9529
454 0.9533 0.9551 0.9533 0.9536 0.9551 0.9533 0.9536
455 0.9526 0.9541 0.9526 0.9529 0.9541 0.9526 0.9529
456 0.9533 0.9551 0.9533 0.9536 0.9551 0.9533 0.9536
457 0.9533 0.9545 0.9533 0.9535 0.9545 0.9533 0.9535
458 0.9526 0.9541 0.9526 0.9529 0.9541 0.9526 0.9529
459 0.9533 0.9551 0.9533 0.9536 0.9551 0.9533 0.9536
460 0.9533 0.9545 0.9533 0.9535 0.9545 0.9533 0.9535
461 0.9546 0.9562 0.9546 0.9548 0.9562 0.9546 0.9548
462 0.9553 0.9570 0.9553 0.9555 0.9570 0.9553 0.9555
463 0.9566 0.9581 0.9566 0.9568 0.9581 0.9566 0.9568
464 0.9553 0.9568 0.9553 0.9555 0.9568 0.9553 0.9555
465 0.9513 0.9532 0.9513 0.9516 0.9532 0.9513 0.9516
466 0.9539 0.9556 0.9539 0.9542 0.9556 0.9539 0.9542
467 0.9520 0.9539 0.9520 0.9523 0.9539 0.9520 0.9523
468 0.9533 0.9547 0.9533 0.9535 0.9547 0.9533 0.9535
469 0.9526 0.9541 0.9526 0.9529 0.9541 0.9526 0.9529
470 0.9533 0.9551 0.9533 0.9536 0.9551 0.9533 0.9536
471 0.9533 0.9545 0.9533 0.9535 0.9545 0.9533 0.9535
472 0.9526 0.9541 0.9526 0.9529 0.9541 0.9526 0.9529
473 0.9533 0.9551 0.9533 0.9536 0.9551 0.9533 0.9536
474 0.9533 0.9545 0.9533 0.9535 0.9545 0.9533 0.9535
475 0.9546 0.9562 0.9546 0.9548 0.9562 0.9546 0.9548
476 0.9553 0.9570 0.9553 0.9555 0.9570 0.9553 0.9555
477 0.9566 0.9581 0.9566 0.9568 0.9581 0.9566 0.9568
478 0.9553 0.9568 0.9553 0.9555 0.9568 0.9553 0.9555
479 0.9513 0.9532 0.9513 0.9516 0.9532 0.9513 0.9516
480 0.9539 0.9556 0.9539 0.9542 0.9556 0.9539 0.9542
481 0.9520 0.9539 0.9520 0.9523 0.9539 0.9520 0.9523
482 0.9533 0.9547 0.9533 0.9535 0.9547 0.9533 0.9535
483 0.9546 0.9561 0.9546 0.9549 0.9561 0.9546 0.9549
484 0.9513 0.9531 0.9513 0.9516 0.9531 0.9513 0.9516
485 0.9533 0.9549 0.9533 0.9535 0.9549 0.9533 0.9535
486 0.9559 0.9570 0.9559 0.9561 0.9570 0.9559 0.9561
487 0.9513 0.9532 0.9513 0.9516 0.9532 0.9513 0.9516
488 0.9513 0.9531 0.9513 0.9516 0.9531 0.9513 0.9516
489 0.9520 0.9538 0.9520 0.9523 0.9538 0.9520 0.9523
490 0.9539 0.9554 0.9539 0.9542 0.9554 0.9539 0.9542
491 0.9526 0.9544 0.9526 0.9529 0.9544 0.9526 0.9529
492 0.9526 0.9541 0.9526 0.9529 0.9541 0.9526 0.9529
493 0.9533 0.9551 0.9533 0.9536 0.9551 0.9533 0.9536
494 0.9533 0.9545 0.9533 0.9535 0.9545 0.9533 0.9535
495 0.9546 0.9562 0.9546 0.9548 0.9562 0.9546 0.9548
496 0.9526 0.9541 0.9526 0.9529 0.9541 0.9526 0.9529
497 0.9526 0.9541 0.9526 0.9529 0.9541 0.9526 0.9529
498 0.9533 0.9551 0.9533 0.9536 0.9551 0.9533 0.9536
499 0.9533 0.9545 0.9533 0.9535 0.9545 0.9533 0.9535
500 0.9546 0.9562 0.9546 0.9548 0.9562 0.9546 0.9548
501 0.9526 0.9541 0.9526 0.9529 0.9541 0.9526 0.9529
502 0.9533 0.9551 0.9533 0.9536 0.9551 0.9533 0.9536
503 0.9533 0.9545 0.9533 0.9535 0.9545 0.9533 0.9535
504 0.9546 0.9562 0.9546 0.9548 0.9562 0.9546 0.9548
505 0.9553 0.9570 0.9553 0.9555 0.9570 0.9553 0.9555
506 0.9566 0.9581 0.9566 0.9568 0.9581 0.9566 0.9568
507 0.9553 0.9568 0.9553 0.9555 0.9568 0.9553 0.9555
508 0.9513 0.9532 0.9513 0.9516 0.9532 0.9513 0.9516
509 0.9539 0.9556 0.9539 0.9542 0.9556 0.9539 0.9542
510 0.9520 0.9539 0.9520 0.9523 0.9539 0.9520 0.9523
511 0.9533 0.9547 0.9533 0.9535 0.9547 0.9533 0.9535
512 0.9546 0.9561 0.9546 0.9549 0.9561 0.9546 0.9549
513 0.9513 0.9531 0.9513 0.9516 0.9531 0.9513 0.9516

73
reports/LinearFiltration/attack_values.csv
View File

@@ -1,35 +1,38 @@
target_class,parameter_mia_accuracy,latent_distance_tell,lookalike_accuracy target_class, parameter_mia_accuracy, lookalike_accuracy, A-Dist, JS-Dist
0,0.500000,3.219560,1.000000 0,0.500000,1.000000, 0.062500, 0.485899
1,0.500000,3.573733,1.000000 1,0.500000,1.000000, 0.083333, 0.484178
2,0.500000,3.924550,1.000000 2,0.500000,0.989583, 0.010417, 0.525418
3,0.500000,3.515182,1.000000 3,0.500000,0.989583, 0.333333, 0.487282
4,0.500000,3.369485,1.000000 4,0.500000,1.000000, 0.020833, 0.505002
5,0.500000,4.697085,1.000000 5,0.500000,1.000000, 0.093750, 0.491621
6,0.500000,4.185269,1.000000 6,0.500000,1.000000, 0.052083, 0.482068
7,0.500000,3.473583,1.000000 7,0.500000,0.979167, 0.010417, 0.519800
8,0.500000,4.518519,1.000000 8,0.500000,0.989583, 0.020833, 0.485863
9,0.500000,4.484542,1.000000 9,0.500000,0.989583, 0.208333, 0.483455
10,0.500000,3.353875,1.000000 10,0.500000,1.000000, 0.145833, 0.454615
11,0.500000,3.815037,0.989583 11,0.500000,1.000000, 0.333333, 0.495618
12,0.500000,3.493994,1.000000 12,0.500000,1.000000, 0.072917, 0.458427
13,0.500000,4.018041,1.000000 13,0.500000,1.000000, 0.135417, 0.477052
14,0.500000,4.059689,1.000000 14,0.500000,1.000000, 0.020833, 0.472611
15,0.500000,4.097189,1.000000 15,0.500000,1.000000, 0.239583, 0.479607
16,0.500000,3.287521,1.000000 16,0.500000,1.000000, 0.135417, 0.468667
17,0.500000,3.872488,1.000000 17,0.500000,1.000000, 0.208333, 0.469592
18,0.500000,3.698952,0.989583 18,0.500000,1.000000, 0.197917, 0.454318
19,0.500000,3.844830,0.989583 19,0.500000,0.989583, 0.041667, 0.499034
0,0.500000,3.413494,1.000000 0,0.500000,1.000000, 0.072917, 0.483694
0,0.500000,3.416399,1.000000 1,0.500000,1.000000, 0.239583, 0.482249
1,0.500000,3.396268,1.000000 2,0.500000,0.989583, 0.062500, 0.528699
2,0.500000,3.648122,1.000000 3,0.500000,1.000000, 0.177083, 0.488769
3,0.500000,3.889339,0.989583 0,0.500000,1.000000, 0.208333, 0.479962
4,0.500000,3.408539,1.000000 1,0.500000,1.000000, 0.177083, 0.480766
5,0.500000,4.625342,1.000000 2,0.500000,1.000000, 0.156250, 0.530593
6,0.500000,3.925502,0.989583 3,0.500000,1.000000, 0.031250, 0.490093
7,0.500000,3.033561,0.989583 4,0.500000,1.000000, 0.062500, 0.514937
8,0.500000,3.819170,0.989583 5,0.500000,1.000000, 0.041667, 0.483999
9,0.500000,4.497495,1.000000 6,0.500000,1.000000, 0.041667, 0.480639
10,0.500000,3.701839,1.000000 7,0.500000,1.000000, 0.114583, 0.516643
11,0.500000,4.369391,0.989583 8,0.500000,1.000000, 0.322917, 0.482295
12,0.500000,3.945474,1.000000 9,0.500000,1.000000, 0.156250, 0.482486
10,0.500000,1.000000, 0.208333, 0.454807
11,0.500000,1.000000, 0.197917, 0.496418
12,0.500000,1.000000, 0.083333, 0.449275
1 target_class parameter_mia_accuracy latent_distance_tell lookalike_accuracy A-Dist JS-Dist
2 0 0.500000 3.219560 1.000000 0.062500 0.485899
3 1 0.500000 3.573733 1.000000 0.083333 0.484178
4 2 0.500000 3.924550 1.000000 0.989583 0.010417 0.525418
5 3 0.500000 3.515182 1.000000 0.989583 0.333333 0.487282
6 4 0.500000 3.369485 1.000000 0.020833 0.505002
7 5 0.500000 4.697085 1.000000 0.093750 0.491621
8 6 0.500000 4.185269 1.000000 0.052083 0.482068
9 7 0.500000 3.473583 1.000000 0.979167 0.010417 0.519800
10 8 0.500000 4.518519 1.000000 0.989583 0.020833 0.485863
11 9 0.500000 4.484542 1.000000 0.989583 0.208333 0.483455
12 10 0.500000 3.353875 1.000000 0.145833 0.454615
13 11 0.500000 3.815037 0.989583 1.000000 0.333333 0.495618
14 12 0.500000 3.493994 1.000000 0.072917 0.458427
15 13 0.500000 4.018041 1.000000 0.135417 0.477052
16 14 0.500000 4.059689 1.000000 0.020833 0.472611
17 15 0.500000 4.097189 1.000000 0.239583 0.479607
18 16 0.500000 3.287521 1.000000 0.135417 0.468667
19 17 0.500000 3.872488 1.000000 0.208333 0.469592
20 18 0.500000 3.698952 0.989583 1.000000 0.197917 0.454318
21 19 0.500000 3.844830 0.989583 0.041667 0.499034
22 0 0.500000 3.413494 1.000000 0.072917 0.483694
23 0 1 0.500000 3.416399 1.000000 0.239583 0.482249
24 1 2 0.500000 3.396268 1.000000 0.989583 0.062500 0.528699
25 2 3 0.500000 3.648122 1.000000 0.177083 0.488769
26 3 0 0.500000 3.889339 0.989583 1.000000 0.208333 0.479962
27 4 1 0.500000 3.408539 1.000000 0.177083 0.480766
28 5 2 0.500000 4.625342 1.000000 0.156250 0.530593
29 6 3 0.500000 3.925502 0.989583 1.000000 0.031250 0.490093
30 7 4 0.500000 3.033561 0.989583 1.000000 0.062500 0.514937
31 8 5 0.500000 3.819170 0.989583 1.000000 0.041667 0.483999
32 9 6 0.500000 4.497495 1.000000 0.041667 0.480639
33 10 7 0.500000 3.701839 1.000000 0.114583 0.516643
34 11 8 0.500000 4.369391 0.989583 1.000000 0.322917 0.482295
35 12 9 0.500000 3.945474 1.000000 0.156250 0.482486
36 10 0.500000 1.000000 0.208333 0.454807
37 11 0.500000 1.000000 0.197917 0.496418
38 12 0.500000 1.000000 0.083333 0.449275

35
reports/LinearFiltration/attack_values_int.csv Normal file
View File

@@ -0,0 +1,35 @@
target_class,parameter_mia_accuracy,latent_distance_tell,lookalike_accuracy
0,0.500000,3.219560,1.000000
1,0.500000,3.573733,1.000000
2,0.500000,3.924550,1.000000
3,0.500000,3.515182,1.000000
4,0.500000,3.369485,1.000000
5,0.500000,4.697085,1.000000
6,0.500000,4.185269,1.000000
7,0.500000,3.473583,1.000000
8,0.500000,4.518519,1.000000
9,0.500000,4.484542,1.000000
10,0.500000,3.353875,1.000000
11,0.500000,3.815037,0.989583
12,0.500000,3.493994,1.000000
13,0.500000,4.018041,1.000000
14,0.500000,4.059689,1.000000
15,0.500000,4.097189,1.000000
16,0.500000,3.287521,1.000000
17,0.500000,3.872488,1.000000
18,0.500000,3.698952,0.989583
19,0.500000,3.844830,0.989583
0,0.500000,3.413494,1.000000
0,0.500000,3.416399,1.000000
1,0.500000,3.396268,1.000000
2,0.500000,3.648122,1.000000
3,0.500000,3.889339,0.989583
4,0.500000,3.408539,1.000000
5,0.500000,4.625342,1.000000
6,0.500000,3.925502,0.989583
7,0.500000,3.033561,0.989583
8,0.500000,3.819170,0.989583
9,0.500000,4.497495,1.000000
10,0.500000,3.701839,1.000000
11,0.500000,4.369391,0.989583
12,0.500000,3.945474,1.000000
1 target_class parameter_mia_accuracy latent_distance_tell lookalike_accuracy
2 0 0.500000 3.219560 1.000000
3 1 0.500000 3.573733 1.000000
4 2 0.500000 3.924550 1.000000
5 3 0.500000 3.515182 1.000000
6 4 0.500000 3.369485 1.000000
7 5 0.500000 4.697085 1.000000
8 6 0.500000 4.185269 1.000000
9 7 0.500000 3.473583 1.000000
10 8 0.500000 4.518519 1.000000
11 9 0.500000 4.484542 1.000000
12 10 0.500000 3.353875 1.000000
13 11 0.500000 3.815037 0.989583
14 12 0.500000 3.493994 1.000000
15 13 0.500000 4.018041 1.000000
16 14 0.500000 4.059689 1.000000
17 15 0.500000 4.097189 1.000000
18 16 0.500000 3.287521 1.000000
19 17 0.500000 3.872488 1.000000
20 18 0.500000 3.698952 0.989583
21 19 0.500000 3.844830 0.989583
22 0 0.500000 3.413494 1.000000
23 0 0.500000 3.416399 1.000000
24 1 0.500000 3.396268 1.000000
25 2 0.500000 3.648122 1.000000
26 3 0.500000 3.889339 0.989583
27 4 0.500000 3.408539 1.000000
28 5 0.500000 4.625342 1.000000
29 6 0.500000 3.925502 0.989583
30 7 0.500000 3.033561 0.989583
31 8 0.500000 3.819170 0.989583
32 9 0.500000 4.497495 1.000000
33 10 0.500000 3.701839 1.000000
34 11 0.500000 4.369391 0.989583
35 12 0.500000 3.945474 1.000000

62
reports/Retrain/RESNET34/forget.csv
View File

@@ -80,3 +80,65 @@ accuracy,macro_precision,macro_recall,macro_f1,weighted_precision,weighted_recal
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
1 accuracy macro_precision macro_recall macro_f1 weighted_precision weighted_recall weighted_f1
80 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
81 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
82 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
83 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
84 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
85 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
86 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
87 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
88 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
89 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
90 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
91 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
92 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
93 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
94 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
95 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
96 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
97 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
98 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
99 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
100 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
102 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
103 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
104 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
105 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
106 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
107 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
108 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
109 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
110 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
111 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
112 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
113 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
114 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
115 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
116 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
117 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
118 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
119 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
120 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
121 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
122 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
123 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
124 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
125 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
126 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
127 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
128 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
129 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
130 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
131 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
132 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
133 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
134 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
135 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
136 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
137 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
138 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
139 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
140 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
141 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
142 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
143 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
144 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000

62
reports/Retrain/RESNET34/forget_train.csv
View File

@@ -80,3 +80,65 @@ accuracy,macro_precision,macro_recall,macro_f1,weighted_precision,weighted_recal
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
1 accuracy macro_precision macro_recall macro_f1 weighted_precision weighted_recall weighted_f1
80 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
81 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
82 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
83 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
84 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
85 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
86 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
87 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
88 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
89 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
90 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
91 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
92 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
93 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
94 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
95 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
96 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
97 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
98 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
99 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
100 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
101 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
102 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
103 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
104 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
105 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
106 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
107 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
108 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
109 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
110 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
111 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
112 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
113 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
114 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
115 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
116 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
117 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
118 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
119 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
120 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
121 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
122 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
123 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
124 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
125 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
126 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
127 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
128 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
129 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
130 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
131 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
132 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
133 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
134 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
135 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
136 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
137 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
138 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
139 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
140 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
141 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
142 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
143 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
144 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000

62
reports/Retrain/RESNET34/retain.csv
View File

@@ -80,3 +80,65 @@ accuracy,macro_precision,macro_recall,macro_f1,weighted_precision,weighted_recal
0.9546,0.9554,0.9546,0.9546,0.9554,0.9546,0.9546 0.9546,0.9554,0.9546,0.9546,0.9554,0.9546,0.9546
0.9612,0.9620,0.9612,0.9613,0.9620,0.9612,0.9613 0.9612,0.9620,0.9612,0.9613,0.9620,0.9612,0.9613
0.9539,0.9551,0.9539,0.9540,0.9551,0.9539,0.9540 0.9539,0.9551,0.9539,0.9540,0.9551,0.9539,0.9540
0.9625,0.9638,0.9625,0.9626,0.9638,0.9625,0.9626
0.9559,0.9571,0.9559,0.9560,0.9571,0.9559,0.9560
0.9507,0.9518,0.9507,0.9507,0.9518,0.9507,0.9507
0.9559,0.9571,0.9559,0.9560,0.9571,0.9559,0.9560
0.9507,0.9518,0.9507,0.9507,0.9518,0.9507,0.9507
0.9612,0.9622,0.9612,0.9612,0.9622,0.9612,0.9612
0.9559,0.9571,0.9559,0.9560,0.9571,0.9559,0.9560
0.9507,0.9518,0.9507,0.9507,0.9518,0.9507,0.9507
0.9612,0.9622,0.9612,0.9612,0.9622,0.9612,0.9612
0.9553,0.9561,0.9553,0.9552,0.9561,0.9553,0.9552
0.9618,0.9626,0.9618,0.9618,0.9626,0.9618,0.9618
0.9566,0.9586,0.9566,0.9569,0.9586,0.9566,0.9569
0.9533,0.9548,0.9533,0.9535,0.9548,0.9533,0.9535
0.9539,0.9548,0.9539,0.9541,0.9548,0.9539,0.9541
0.9500,0.9511,0.9500,0.9500,0.9511,0.9500,0.9500
0.9566,0.9572,0.9566,0.9565,0.9572,0.9566,0.9565
0.9546,0.9554,0.9546,0.9546,0.9554,0.9546,0.9546
0.9559,0.9571,0.9559,0.9560,0.9571,0.9559,0.9560
0.9507,0.9518,0.9507,0.9507,0.9518,0.9507,0.9507
0.9612,0.9622,0.9612,0.9612,0.9622,0.9612,0.9612
0.9559,0.9571,0.9559,0.9560,0.9571,0.9559,0.9560
0.9507,0.9518,0.9507,0.9507,0.9518,0.9507,0.9507
0.9612,0.9622,0.9612,0.9612,0.9622,0.9612,0.9612
0.9553,0.9561,0.9553,0.9552,0.9561,0.9553,0.9552
0.9618,0.9626,0.9618,0.9618,0.9626,0.9618,0.9618
0.9566,0.9586,0.9566,0.9569,0.9586,0.9566,0.9569
0.9533,0.9548,0.9533,0.9535,0.9548,0.9533,0.9535
0.9539,0.9548,0.9539,0.9541,0.9548,0.9539,0.9541
0.9500,0.9511,0.9500,0.9500,0.9511,0.9500,0.9500
0.9566,0.9572,0.9566,0.9565,0.9572,0.9566,0.9565
0.9546,0.9554,0.9546,0.9546,0.9554,0.9546,0.9546
0.9612,0.9620,0.9612,0.9613,0.9620,0.9612,0.9613
0.9539,0.9551,0.9539,0.9540,0.9551,0.9539,0.9540
0.9625,0.9638,0.9625,0.9626,0.9638,0.9625,0.9626
0.9533,0.9548,0.9533,0.9535,0.9548,0.9533,0.9535
0.9520,0.9532,0.9520,0.9521,0.9532,0.9520,0.9521
0.9579,0.9586,0.9579,0.9580,0.9586,0.9579,0.9580
0.9566,0.9572,0.9566,0.9566,0.9572,0.9566,0.9566
0.9533,0.9544,0.9533,0.9535,0.9544,0.9533,0.9535
0.9559,0.9570,0.9559,0.9561,0.9570,0.9559,0.9561
0.9559,0.9571,0.9559,0.9560,0.9571,0.9559,0.9560
0.9507,0.9518,0.9507,0.9507,0.9518,0.9507,0.9507
0.9612,0.9622,0.9612,0.9612,0.9622,0.9612,0.9612
0.9553,0.9561,0.9553,0.9552,0.9561,0.9553,0.9552
0.9559,0.9571,0.9559,0.9560,0.9571,0.9559,0.9560
0.9559,0.9571,0.9559,0.9560,0.9571,0.9559,0.9560
0.9507,0.9518,0.9507,0.9507,0.9518,0.9507,0.9507
0.9612,0.9622,0.9612,0.9612,0.9622,0.9612,0.9612
0.9553,0.9561,0.9553,0.9552,0.9561,0.9553,0.9552
0.9559,0.9571,0.9559,0.9560,0.9571,0.9559,0.9560
0.9507,0.9518,0.9507,0.9507,0.9518,0.9507,0.9507
0.9612,0.9622,0.9612,0.9612,0.9622,0.9612,0.9612
0.9553,0.9561,0.9553,0.9552,0.9561,0.9553,0.9552
0.9618,0.9626,0.9618,0.9618,0.9626,0.9618,0.9618
0.9566,0.9586,0.9566,0.9569,0.9586,0.9566,0.9569
0.9533,0.9548,0.9533,0.9535,0.9548,0.9533,0.9535
0.9539,0.9548,0.9539,0.9541,0.9548,0.9539,0.9541
0.9500,0.9511,0.9500,0.9500,0.9511,0.9500,0.9500
0.9566,0.9572,0.9566,0.9565,0.9572,0.9566,0.9565
0.9546,0.9554,0.9546,0.9546,0.9554,0.9546,0.9546
0.9612,0.9620,0.9612,0.9613,0.9620,0.9612,0.9613
0.9539,0.9551,0.9539,0.9540,0.9551,0.9539,0.9540
1 accuracy macro_precision macro_recall macro_f1 weighted_precision weighted_recall weighted_f1
80 0.9546 0.9554 0.9546 0.9546 0.9554 0.9546 0.9546
81 0.9612 0.9620 0.9612 0.9613 0.9620 0.9612 0.9613
82 0.9539 0.9551 0.9539 0.9540 0.9551 0.9539 0.9540
83 0.9625 0.9638 0.9625 0.9626 0.9638 0.9625 0.9626
84 0.9559 0.9571 0.9559 0.9560 0.9571 0.9559 0.9560
85 0.9507 0.9518 0.9507 0.9507 0.9518 0.9507 0.9507
86 0.9559 0.9571 0.9559 0.9560 0.9571 0.9559 0.9560
87 0.9507 0.9518 0.9507 0.9507 0.9518 0.9507 0.9507
88 0.9612 0.9622 0.9612 0.9612 0.9622 0.9612 0.9612
89 0.9559 0.9571 0.9559 0.9560 0.9571 0.9559 0.9560
90 0.9507 0.9518 0.9507 0.9507 0.9518 0.9507 0.9507
91 0.9612 0.9622 0.9612 0.9612 0.9622 0.9612 0.9612
92 0.9553 0.9561 0.9553 0.9552 0.9561 0.9553 0.9552
93 0.9618 0.9626 0.9618 0.9618 0.9626 0.9618 0.9618
94 0.9566 0.9586 0.9566 0.9569 0.9586 0.9566 0.9569
95 0.9533 0.9548 0.9533 0.9535 0.9548 0.9533 0.9535
96 0.9539 0.9548 0.9539 0.9541 0.9548 0.9539 0.9541
97 0.9500 0.9511 0.9500 0.9500 0.9511 0.9500 0.9500
98 0.9566 0.9572 0.9566 0.9565 0.9572 0.9566 0.9565
99 0.9546 0.9554 0.9546 0.9546 0.9554 0.9546 0.9546
100 0.9559 0.9571 0.9559 0.9560 0.9571 0.9559 0.9560
101 0.9507 0.9518 0.9507 0.9507 0.9518 0.9507 0.9507
102 0.9612 0.9622 0.9612 0.9612 0.9622 0.9612 0.9612
103 0.9559 0.9571 0.9559 0.9560 0.9571 0.9559 0.9560
104 0.9507 0.9518 0.9507 0.9507 0.9518 0.9507 0.9507
105 0.9612 0.9622 0.9612 0.9612 0.9622 0.9612 0.9612
106 0.9553 0.9561 0.9553 0.9552 0.9561 0.9553 0.9552
107 0.9618 0.9626 0.9618 0.9618 0.9626 0.9618 0.9618
108 0.9566 0.9586 0.9566 0.9569 0.9586 0.9566 0.9569
109 0.9533 0.9548 0.9533 0.9535 0.9548 0.9533 0.9535
110 0.9539 0.9548 0.9539 0.9541 0.9548 0.9539 0.9541
111 0.9500 0.9511 0.9500 0.9500 0.9511 0.9500 0.9500
112 0.9566 0.9572 0.9566 0.9565 0.9572 0.9566 0.9565
113 0.9546 0.9554 0.9546 0.9546 0.9554 0.9546 0.9546
114 0.9612 0.9620 0.9612 0.9613 0.9620 0.9612 0.9613
115 0.9539 0.9551 0.9539 0.9540 0.9551 0.9539 0.9540
116 0.9625 0.9638 0.9625 0.9626 0.9638 0.9625 0.9626
117 0.9533 0.9548 0.9533 0.9535 0.9548 0.9533 0.9535
118 0.9520 0.9532 0.9520 0.9521 0.9532 0.9520 0.9521
119 0.9579 0.9586 0.9579 0.9580 0.9586 0.9579 0.9580
120 0.9566 0.9572 0.9566 0.9566 0.9572 0.9566 0.9566
121 0.9533 0.9544 0.9533 0.9535 0.9544 0.9533 0.9535
122 0.9559 0.9570 0.9559 0.9561 0.9570 0.9559 0.9561
123 0.9559 0.9571 0.9559 0.9560 0.9571 0.9559 0.9560
124 0.9507 0.9518 0.9507 0.9507 0.9518 0.9507 0.9507
125 0.9612 0.9622 0.9612 0.9612 0.9622 0.9612 0.9612
126 0.9553 0.9561 0.9553 0.9552 0.9561 0.9553 0.9552
127 0.9559 0.9571 0.9559 0.9560 0.9571 0.9559 0.9560
128 0.9559 0.9571 0.9559 0.9560 0.9571 0.9559 0.9560
129 0.9507 0.9518 0.9507 0.9507 0.9518 0.9507 0.9507
130 0.9612 0.9622 0.9612 0.9612 0.9622 0.9612 0.9612
131 0.9553 0.9561 0.9553 0.9552 0.9561 0.9553 0.9552
132 0.9559 0.9571 0.9559 0.9560 0.9571 0.9559 0.9560
133 0.9507 0.9518 0.9507 0.9507 0.9518 0.9507 0.9507
134 0.9612 0.9622 0.9612 0.9612 0.9622 0.9612 0.9612
135 0.9553 0.9561 0.9553 0.9552 0.9561 0.9553 0.9552
136 0.9618 0.9626 0.9618 0.9618 0.9626 0.9618 0.9618
137 0.9566 0.9586 0.9566 0.9569 0.9586 0.9566 0.9569
138 0.9533 0.9548 0.9533 0.9535 0.9548 0.9533 0.9535
139 0.9539 0.9548 0.9539 0.9541 0.9548 0.9539 0.9541
140 0.9500 0.9511 0.9500 0.9500 0.9511 0.9500 0.9500
141 0.9566 0.9572 0.9566 0.9565 0.9572 0.9566 0.9565
142 0.9546 0.9554 0.9546 0.9546 0.9554 0.9546 0.9546
143 0.9612 0.9620 0.9612 0.9613 0.9620 0.9612 0.9613
144 0.9539 0.9551 0.9539 0.9540 0.9551 0.9539 0.9540

76
reports/Retrain/attack_values.csv
View File

@@ -1,35 +1,41 @@
target_class,parameter_mia_accuracy,latent_distance_tell,lookalike_accuracy target_class, parameter_mia_accuracy, lookalike_accuracy, A-Dist, JS-Dist
0,0.500000,11.573492,0.000000 0,0.500000,0.000000, 0.156250, 0.000000
1,0.500000,12.793120,0.000000 1,0.500000,0.000000, 0.104167, 0.000000
2,0.500000,12.951434,0.000000 2,0.500000,0.000000, 0.114583, 0.000000
3,0.500000,10.942259,0.000000 0,0.500000,0.000000, 0.125000, 0.000000
4,0.500000,10.675704,0.000000 1,0.500000,0.000000, 0.072917, 0.000000
5,0.500000,13.046491,0.000000 2,0.500000,0.000000, 0.020833, 0.000000
6,0.500000,11.343709,0.000000 3,0.500000,0.000000, 0.104167, 0.000000
7,0.500000,10.020437,0.000000 4,0.500000,0.000000, 0.020833, 0.000000
8,0.500000,12.896643,0.000000 5,0.500000,0.000000, 0.041667, 0.000000
9,0.500000,12.220088,0.000000 6,0.500000,0.000000, 0.104167, 0.000000
10,0.500000,11.315407,0.000000 7,0.500000,0.000000, 0.281250, 0.000000
11,0.450000,10.556920,0.000000 8,0.500000,0.000000, 0.114583, 0.000000
12,0.500000,10.925443,0.000000 9,0.500000,0.000000, 0.072917, 0.000000
13,0.500000,12.007375,0.000000 10,0.500000,0.000000, 0.020833, 0.000000
14,0.500000,11.471087,0.000000 11,0.500000,0.000000, 0.031250, 0.000000
15,0.500000,12.301955,0.000000 12,0.500000,0.000000, 0.072917, 0.000000
16,0.500000,10.831913,0.000000 13,0.500000,0.000000, 0.052083, 0.000000
17,0.500000,10.942777,0.000000 14,0.500000,0.000000, 0.145833, 0.000000
18,0.500000,11.699934,0.000000 15,0.500000,0.000000, 0.052083, 0.000000
19,0.500000,12.526654,0.000000 16,0.500000,0.000000, 0.104167, 0.000000
0,0.500000,11.601441,0.000000 17,0.500000,0.000000, 0.093750, 0.000000
0,0.500000,11.639872,0.000000 18,0.531250,0.000000, 0.010417, 0.000000
1,0.500000,12.723348,0.000000 19,0.500000,0.000000, 0.031250, 0.000000
2,0.500000,13.151769,0.000000 0,0.500000,0.000000, 0.062500, 0.000000
3,0.500000,10.969514,0.000000 1,0.500000,0.000000, 0.072917, 0.000000
4,0.500000,10.674448,0.000000 2,0.500000,0.000000, 0.031250, 0.000000
5,0.500000,13.068334,0.000000 3,0.500000,0.000000, 0.083333, 0.000000
6,0.500000,11.328772,0.000000 0,0.500000,0.000000, 0.010417, 0.000000
7,0.500000,9.926997,0.000000 1,0.500000,0.000000, 0.062500, 0.000000
8,0.500000,12.859781,0.000000 2,0.500000,0.000000, 0.104167, 0.000000
9,0.500000,12.061253,0.000000 3,0.500000,0.000000, 0.072917, 0.000000
10,0.500000,11.358070,0.000000 4,0.500000,0.000000, 0.177083, 0.000000
11,0.500000,10.484346,0.000000 5,0.500000,0.000000, 0.135417, 0.000000
12,0.500000,10.926878,0.000000 6,0.500000,0.000000, 0.197917, 0.000000
7,0.500000,0.000000, 0.218750, 0.000000
8,0.500000,0.000000, 0.114583, 0.000000
9,0.500000,0.000000, 0.052083, 0.000000
10,0.500000,0.000000, 0.072917, 0.000000
11,0.500000,0.000000, 0.072917, 0.000000
12,0.500000,0.000000, 0.125000, 0.000000
1 target_class parameter_mia_accuracy latent_distance_tell lookalike_accuracy A-Dist JS-Dist
2 0 0.500000 11.573492 0.000000 0.156250 0.000000
3 1 0.500000 12.793120 0.000000 0.104167 0.000000
4 2 0.500000 12.951434 0.000000 0.114583 0.000000
5 3 0 0.500000 10.942259 0.000000 0.125000 0.000000
6 4 1 0.500000 10.675704 0.000000 0.072917 0.000000
7 5 2 0.500000 13.046491 0.000000 0.020833 0.000000
8 6 3 0.500000 11.343709 0.000000 0.104167 0.000000
9 7 4 0.500000 10.020437 0.000000 0.020833 0.000000
10 8 5 0.500000 12.896643 0.000000 0.041667 0.000000
11 9 6 0.500000 12.220088 0.000000 0.104167 0.000000
12 10 7 0.500000 11.315407 0.000000 0.281250 0.000000
13 11 8 0.450000 0.500000 10.556920 0.000000 0.114583 0.000000
14 12 9 0.500000 10.925443 0.000000 0.072917 0.000000
15 13 10 0.500000 12.007375 0.000000 0.020833 0.000000
16 14 11 0.500000 11.471087 0.000000 0.031250 0.000000
17 15 12 0.500000 12.301955 0.000000 0.072917 0.000000
18 16 13 0.500000 10.831913 0.000000 0.052083 0.000000
19 17 14 0.500000 10.942777 0.000000 0.145833 0.000000
20 18 15 0.500000 11.699934 0.000000 0.052083 0.000000
21 19 16 0.500000 12.526654 0.000000 0.104167 0.000000
22 0 17 0.500000 11.601441 0.000000 0.093750 0.000000
23 0 18 0.500000 0.531250 11.639872 0.000000 0.010417 0.000000
24 1 19 0.500000 12.723348 0.000000 0.031250 0.000000
25 2 0 0.500000 13.151769 0.000000 0.062500 0.000000
26 3 1 0.500000 10.969514 0.000000 0.072917 0.000000
27 4 2 0.500000 10.674448 0.000000 0.031250 0.000000
28 5 3 0.500000 13.068334 0.000000 0.083333 0.000000
29 6 0 0.500000 11.328772 0.000000 0.010417 0.000000
30 7 1 0.500000 9.926997 0.000000 0.062500 0.000000
31 8 2 0.500000 12.859781 0.000000 0.104167 0.000000
32 9 3 0.500000 12.061253 0.000000 0.072917 0.000000
33 10 4 0.500000 11.358070 0.000000 0.177083 0.000000
34 11 5 0.500000 10.484346 0.000000 0.135417 0.000000
35 12 6 0.500000 10.926878 0.000000 0.197917 0.000000
36 7 0.500000 0.000000 0.218750 0.000000
37 8 0.500000 0.000000 0.114583 0.000000
38 9 0.500000 0.000000 0.052083 0.000000
39 10 0.500000 0.000000 0.072917 0.000000
40 11 0.500000 0.000000 0.072917 0.000000
41 12 0.500000 0.000000 0.125000 0.000000

38
reports/Retrain/attack_values_int.csv Normal file
View File

@@ -0,0 +1,38 @@
target_class,parameter_mia_accuracy,latent_distance_tell,lookalike_accuracy
0,0.500000,11.573492,0.000000
1,0.500000,12.793120,0.000000
2,0.500000,12.951434,0.000000
3,0.500000,10.942259,0.000000
4,0.500000,10.675704,0.000000
5,0.500000,13.046491,0.000000
6,0.500000,11.343709,0.000000
7,0.500000,10.020437,0.000000
8,0.500000,12.896643,0.000000
9,0.500000,12.220088,0.000000
10,0.500000,11.315407,0.000000
11,0.450000,10.556920,0.000000
12,0.500000,10.925443,0.000000
13,0.500000,12.007375,0.000000
14,0.500000,11.471087,0.000000
15,0.500000,12.301955,0.000000
16,0.500000,10.831913,0.000000
17,0.500000,10.942777,0.000000
18,0.500000,11.699934,0.000000
19,0.500000,12.526654,0.000000
0,0.500000,11.601441,0.000000
0,0.500000,11.639872,0.000000
1,0.500000,12.723348,0.000000
2,0.500000,13.151769,0.000000
3,0.500000,10.969514,0.000000
4,0.500000,10.674448,0.000000
5,0.500000,13.068334,0.000000
6,0.500000,11.328772,0.000000
7,0.500000,9.926997,0.000000
8,0.500000,12.859781,0.000000
9,0.500000,12.061253,0.000000
10,0.500000,11.358070,0.000000
11,0.500000,10.484346,0.000000
12,0.500000,10.926878,0.000000
13,0.500000,12.002632,0.000000
0,0.500000,11.557904,0.000000
1,0.500000,12.819191,0.000000
1 target_class parameter_mia_accuracy latent_distance_tell lookalike_accuracy
2 0 0.500000 11.573492 0.000000
3 1 0.500000 12.793120 0.000000
4 2 0.500000 12.951434 0.000000
5 3 0.500000 10.942259 0.000000
6 4 0.500000 10.675704 0.000000
7 5 0.500000 13.046491 0.000000
8 6 0.500000 11.343709 0.000000
9 7 0.500000 10.020437 0.000000
10 8 0.500000 12.896643 0.000000
11 9 0.500000 12.220088 0.000000
12 10 0.500000 11.315407 0.000000
13 11 0.450000 10.556920 0.000000
14 12 0.500000 10.925443 0.000000
15 13 0.500000 12.007375 0.000000
16 14 0.500000 11.471087 0.000000
17 15 0.500000 12.301955 0.000000
18 16 0.500000 10.831913 0.000000
19 17 0.500000 10.942777 0.000000
20 18 0.500000 11.699934 0.000000
21 19 0.500000 12.526654 0.000000
22 0 0.500000 11.601441 0.000000
23 0 0.500000 11.639872 0.000000
24 1 0.500000 12.723348 0.000000
25 2 0.500000 13.151769 0.000000
26 3 0.500000 10.969514 0.000000
27 4 0.500000 10.674448 0.000000
28 5 0.500000 13.068334 0.000000
29 6 0.500000 11.328772 0.000000
30 7 0.500000 9.926997 0.000000
31 8 0.500000 12.859781 0.000000
32 9 0.500000 12.061253 0.000000
33 10 0.500000 11.358070 0.000000
34 11 0.500000 10.484346 0.000000
35 12 0.500000 10.926878 0.000000
36 13 0.500000 12.002632 0.000000
37 0 0.500000 11.557904 0.000000
38 1 0.500000 12.819191 0.000000

61
reports/WeightFiltration/RESNET34/forget.csv
View File

@@ -539,3 +539,64 @@ accuracy,macro_precision,macro_recall,macro_f1,weighted_precision,weighted_recal
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0250,1.0000,0.0250,0.0488,1.0000,0.0250,0.0488
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.2250,1.0000,0.2250,0.3673,1.0000,0.2250,0.3673
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.1000,1.0000,0.1000,0.1818,1.0000,0.1000,0.1818
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0875,1.0000,0.0875,0.1609,1.0000,0.0875,0.1609
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0250,1.0000,0.0250,0.0488,1.0000,0.0250,0.0488
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.1375,1.0000,0.1375,0.2418,1.0000,0.1375,0.2418
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0875,1.0000,0.0875,0.1609,1.0000,0.0875,0.1609
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0125,1.0000,0.0125,0.0247,1.0000,0.0125,0.0247
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.1750,1.0000,0.1750,0.2979,1.0000,0.1750,0.2979
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0250,1.0000,0.0250,0.0488,1.0000,0.0250,0.0488
0.0750,1.0000,0.0750,0.1395,1.0000,0.0750,0.1395
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0500,1.0000,0.0500,0.0952,1.0000,0.0500,0.0952
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0125,1.0000,0.0125,0.0247,1.0000,0.0125,0.0247
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0125,1.0000,0.0125,0.0247,1.0000,0.0125,0.0247
1 accuracy macro_precision macro_recall macro_f1 weighted_precision weighted_recall weighted_f1
539 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
540 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
541 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
542 0.0250 1.0000 0.0250 0.0488 1.0000 0.0250 0.0488
543 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
544 0.2250 1.0000 0.2250 0.3673 1.0000 0.2250 0.3673
545 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
546 0.1000 1.0000 0.1000 0.1818 1.0000 0.1000 0.1818
547 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
548 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
549 0.0875 1.0000 0.0875 0.1609 1.0000 0.0875 0.1609
550 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
551 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
552 0.0250 1.0000 0.0250 0.0488 1.0000 0.0250 0.0488
553 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
554 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
555 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
556 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
557 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
558 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
559 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
560 0.1375 1.0000 0.1375 0.2418 1.0000 0.1375 0.2418
561 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
562 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
563 0.0875 1.0000 0.0875 0.1609 1.0000 0.0875 0.1609
564 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
565 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
566 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
567 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
568 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
569 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
570 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
571 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
572 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
573 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
574 0.0125 1.0000 0.0125 0.0247 1.0000 0.0125 0.0247
575 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
576 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
577 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
578 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
579 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
580 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
581 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
582 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
583 0.1750 1.0000 0.1750 0.2979 1.0000 0.1750 0.2979
584 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
585 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
586 0.0250 1.0000 0.0250 0.0488 1.0000 0.0250 0.0488
587 0.0750 1.0000 0.0750 0.1395 1.0000 0.0750 0.1395
588 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
589 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
590 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
591 0.0500 1.0000 0.0500 0.0952 1.0000 0.0500 0.0952
592 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
593 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
594 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
595 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
596 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
597 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
598 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
599 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
600 0.0125 1.0000 0.0125 0.0247 1.0000 0.0125 0.0247
601 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
602 0.0125 1.0000 0.0125 0.0247 1.0000 0.0125 0.0247

61
reports/WeightFiltration/RESNET34/forget_train.csv
View File

@@ -539,3 +539,64 @@ accuracy,macro_precision,macro_recall,macro_f1,weighted_precision,weighted_recal
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0063,1.0000,0.0063,0.0124,1.0000,0.0063,0.0124 0.0063,1.0000,0.0063,0.0124,1.0000,0.0063,0.0124
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000 0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0312,1.0000,0.0312,0.0606,1.0000,0.0312,0.0606
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.2594,1.0000,0.2594,0.4119,1.0000,0.2594,0.4119
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.1250,1.0000,0.1250,0.2222,1.0000,0.1250,0.2222
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0781,1.0000,0.0781,0.1449,1.0000,0.0781,0.1449
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0063,1.0000,0.0063,0.0124,1.0000,0.0063,0.0124
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0187,1.0000,0.0187,0.0368,1.0000,0.0187,0.0368
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.1313,1.0000,0.1313,0.2320,1.0000,0.1313,0.2320
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0750,1.0000,0.0750,0.1395,1.0000,0.0750,0.1395
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0031,1.0000,0.0031,0.0062,1.0000,0.0031,0.0062
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0563,1.0000,0.0563,0.1065,1.0000,0.0563,0.1065
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.2062,1.0000,0.2062,0.3420,1.0000,0.2062,0.3420
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0344,1.0000,0.0344,0.0665,1.0000,0.0344,0.0665
0.0469,1.0000,0.0469,0.0896,1.0000,0.0469,0.0896
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0594,1.0000,0.0594,0.1121,1.0000,0.0594,0.1121
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0187,1.0000,0.0187,0.0368,1.0000,0.0187,0.0368
0.0000,0.0000,0.0000,0.0000,0.0000,0.0000,0.0000
0.0094,1.0000,0.0094,0.0186,1.0000,0.0094,0.0186
1 accuracy macro_precision macro_recall macro_f1 weighted_precision weighted_recall weighted_f1
539 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
540 0.0063 1.0000 0.0063 0.0124 1.0000 0.0063 0.0124
541 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
542 0.0312 1.0000 0.0312 0.0606 1.0000 0.0312 0.0606
543 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
544 0.2594 1.0000 0.2594 0.4119 1.0000 0.2594 0.4119
545 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
546 0.1250 1.0000 0.1250 0.2222 1.0000 0.1250 0.2222
547 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
548 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
549 0.0781 1.0000 0.0781 0.1449 1.0000 0.0781 0.1449
550 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
551 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
552 0.0063 1.0000 0.0063 0.0124 1.0000 0.0063 0.0124
553 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
554 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
555 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
556 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
557 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
558 0.0187 1.0000 0.0187 0.0368 1.0000 0.0187 0.0368
559 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
560 0.1313 1.0000 0.1313 0.2320 1.0000 0.1313 0.2320
561 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
562 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
563 0.0750 1.0000 0.0750 0.1395 1.0000 0.0750 0.1395
564 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
565 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
566 0.0031 1.0000 0.0031 0.0062 1.0000 0.0031 0.0062
567 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
568 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
569 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
570 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
571 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
572 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
573 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
574 0.0563 1.0000 0.0563 0.1065 1.0000 0.0563 0.1065
575 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
576 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
577 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
578 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
579 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
580 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
581 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
582 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
583 0.2062 1.0000 0.2062 0.3420 1.0000 0.2062 0.3420
584 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
585 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
586 0.0344 1.0000 0.0344 0.0665 1.0000 0.0344 0.0665
587 0.0469 1.0000 0.0469 0.0896 1.0000 0.0469 0.0896
588 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
589 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
590 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
591 0.0594 1.0000 0.0594 0.1121 1.0000 0.0594 0.1121
592 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
593 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
594 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
595 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
596 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
597 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
598 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
599 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
600 0.0187 1.0000 0.0187 0.0368 1.0000 0.0187 0.0368
601 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
602 0.0094 1.0000 0.0094 0.0186 1.0000 0.0094 0.0186

61
reports/WeightFiltration/RESNET34/retain.csv
View File

@@ -539,3 +539,64 @@ accuracy,macro_precision,macro_recall,macro_f1,weighted_precision,weighted_recal
0.9520,0.9534,0.9520,0.9522,0.9534,0.9520,0.9522 0.9520,0.9534,0.9520,0.9522,0.9534,0.9520,0.9522
0.9539,0.9554,0.9539,0.9542,0.9554,0.9539,0.9542 0.9539,0.9554,0.9539,0.9542,0.9554,0.9539,0.9542
0.9539,0.9559,0.9539,0.9543,0.9559,0.9539,0.9543 0.9539,0.9559,0.9539,0.9543,0.9559,0.9539,0.9543
0.9500,0.9518,0.9500,0.9503,0.9518,0.9500,0.9503
0.9520,0.9542,0.9520,0.9524,0.9542,0.9520,0.9524
0.9520,0.9541,0.9520,0.9523,0.9541,0.9520,0.9523
0.9520,0.9542,0.9520,0.9524,0.9542,0.9520,0.9524
0.9520,0.9540,0.9520,0.9523,0.9540,0.9520,0.9523
0.9507,0.9515,0.9507,0.9507,0.9515,0.9507,0.9507
0.9526,0.9548,0.9526,0.9530,0.9548,0.9526,0.9530
0.9520,0.9542,0.9520,0.9524,0.9542,0.9520,0.9524
0.9520,0.9528,0.9520,0.9520,0.9528,0.9520,0.9520
0.9539,0.9557,0.9539,0.9542,0.9557,0.9539,0.9542
0.9579,0.9597,0.9579,0.9582,0.9597,0.9579,0.9582
0.9566,0.9582,0.9566,0.9568,0.9582,0.9566,0.9568
0.9533,0.9551,0.9533,0.9536,0.9551,0.9533,0.9536
0.9500,0.9528,0.9500,0.9505,0.9528,0.9500,0.9505
0.9546,0.9558,0.9546,0.9547,0.9558,0.9546,0.9547
0.9520,0.9534,0.9520,0.9522,0.9534,0.9520,0.9522
0.9546,0.9562,0.9546,0.9549,0.9562,0.9546,0.9549
0.9520,0.9543,0.9520,0.9524,0.9543,0.9520,0.9524
0.9520,0.9540,0.9520,0.9523,0.9540,0.9520,0.9523
0.9520,0.9529,0.9520,0.9520,0.9529,0.9520,0.9520
0.9513,0.9536,0.9513,0.9517,0.9536,0.9513,0.9517
0.9520,0.9538,0.9520,0.9523,0.9538,0.9520,0.9523
0.9513,0.9523,0.9513,0.9513,0.9523,0.9513,0.9513
0.9533,0.9550,0.9533,0.9536,0.9550,0.9533,0.9536
0.9586,0.9605,0.9586,0.9589,0.9605,0.9586,0.9589
0.9566,0.9583,0.9566,0.9568,0.9583,0.9566,0.9568
0.9533,0.9551,0.9533,0.9536,0.9551,0.9533,0.9536
0.9500,0.9528,0.9500,0.9505,0.9528,0.9500,0.9505
0.9559,0.9570,0.9559,0.9560,0.9570,0.9559,0.9560
0.9526,0.9541,0.9526,0.9529,0.9541,0.9526,0.9529
0.9533,0.9549,0.9533,0.9535,0.9549,0.9533,0.9535
0.9539,0.9559,0.9539,0.9543,0.9559,0.9539,0.9543
0.9500,0.9518,0.9500,0.9503,0.9518,0.9500,0.9503
0.9533,0.9547,0.9533,0.9535,0.9547,0.9533,0.9535
0.9546,0.9558,0.9546,0.9547,0.9558,0.9546,0.9547
0.9520,0.9539,0.9520,0.9522,0.9539,0.9520,0.9522
0.9480,0.9506,0.9480,0.9485,0.9506,0.9480,0.9485
0.9520,0.9541,0.9520,0.9522,0.9541,0.9520,0.9522
0.9513,0.9526,0.9513,0.9514,0.9526,0.9513,0.9514
0.9533,0.9546,0.9533,0.9535,0.9546,0.9533,0.9535
0.9526,0.9548,0.9526,0.9530,0.9548,0.9526,0.9530
0.9520,0.9542,0.9520,0.9524,0.9542,0.9520,0.9524
0.9513,0.9523,0.9513,0.9514,0.9523,0.9513,0.9514
0.9533,0.9553,0.9533,0.9536,0.9553,0.9533,0.9536
0.9513,0.9536,0.9513,0.9517,0.9536,0.9513,0.9517
0.9520,0.9540,0.9520,0.9523,0.9540,0.9520,0.9523
0.9513,0.9523,0.9513,0.9513,0.9523,0.9513,0.9513
0.9533,0.9549,0.9533,0.9535,0.9549,0.9533,0.9535
0.9513,0.9536,0.9513,0.9517,0.9536,0.9513,0.9517
0.9520,0.9540,0.9520,0.9523,0.9540,0.9520,0.9523
0.9520,0.9530,0.9520,0.9520,0.9530,0.9520,0.9520
0.9539,0.9557,0.9539,0.9542,0.9557,0.9539,0.9542
0.9579,0.9597,0.9579,0.9582,0.9597,0.9579,0.9582
0.9566,0.9583,0.9566,0.9568,0.9583,0.9566,0.9568
0.9533,0.9551,0.9533,0.9536,0.9551,0.9533,0.9536
0.9487,0.9521,0.9487,0.9494,0.9521,0.9487,0.9494
0.9546,0.9558,0.9546,0.9547,0.9558,0.9546,0.9547
0.9526,0.9539,0.9526,0.9529,0.9539,0.9526,0.9529
0.9539,0.9556,0.9539,0.9542,0.9556,0.9539,0.9542
0.9539,0.9559,0.9539,0.9543,0.9559,0.9539,0.9543
0.9500,0.9518,0.9500,0.9503,0.9518,0.9500,0.9503
1 accuracy macro_precision macro_recall macro_f1 weighted_precision weighted_recall weighted_f1
539 0.9520 0.9534 0.9520 0.9522 0.9534 0.9520 0.9522
540 0.9539 0.9554 0.9539 0.9542 0.9554 0.9539 0.9542
541 0.9539 0.9559 0.9539 0.9543 0.9559 0.9539 0.9543
542 0.9500 0.9518 0.9500 0.9503 0.9518 0.9500 0.9503
543 0.9520 0.9542 0.9520 0.9524 0.9542 0.9520 0.9524
544 0.9520 0.9541 0.9520 0.9523 0.9541 0.9520 0.9523
545 0.9520 0.9542 0.9520 0.9524 0.9542 0.9520 0.9524
546 0.9520 0.9540 0.9520 0.9523 0.9540 0.9520 0.9523
547 0.9507 0.9515 0.9507 0.9507 0.9515 0.9507 0.9507
548 0.9526 0.9548 0.9526 0.9530 0.9548 0.9526 0.9530
549 0.9520 0.9542 0.9520 0.9524 0.9542 0.9520 0.9524
550 0.9520 0.9528 0.9520 0.9520 0.9528 0.9520 0.9520
551 0.9539 0.9557 0.9539 0.9542 0.9557 0.9539 0.9542
552 0.9579 0.9597 0.9579 0.9582 0.9597 0.9579 0.9582
553 0.9566 0.9582 0.9566 0.9568 0.9582 0.9566 0.9568
554 0.9533 0.9551 0.9533 0.9536 0.9551 0.9533 0.9536
555 0.9500 0.9528 0.9500 0.9505 0.9528 0.9500 0.9505
556 0.9546 0.9558 0.9546 0.9547 0.9558 0.9546 0.9547
557 0.9520 0.9534 0.9520 0.9522 0.9534 0.9520 0.9522
558 0.9546 0.9562 0.9546 0.9549 0.9562 0.9546 0.9549
559 0.9520 0.9543 0.9520 0.9524 0.9543 0.9520 0.9524
560 0.9520 0.9540 0.9520 0.9523 0.9540 0.9520 0.9523
561 0.9520 0.9529 0.9520 0.9520 0.9529 0.9520 0.9520
562 0.9513 0.9536 0.9513 0.9517 0.9536 0.9513 0.9517
563 0.9520 0.9538 0.9520 0.9523 0.9538 0.9520 0.9523
564 0.9513 0.9523 0.9513 0.9513 0.9523 0.9513 0.9513
565 0.9533 0.9550 0.9533 0.9536 0.9550 0.9533 0.9536
566 0.9586 0.9605 0.9586 0.9589 0.9605 0.9586 0.9589
567 0.9566 0.9583 0.9566 0.9568 0.9583 0.9566 0.9568
568 0.9533 0.9551 0.9533 0.9536 0.9551 0.9533 0.9536
569 0.9500 0.9528 0.9500 0.9505 0.9528 0.9500 0.9505
570 0.9559 0.9570 0.9559 0.9560 0.9570 0.9559 0.9560
571 0.9526 0.9541 0.9526 0.9529 0.9541 0.9526 0.9529
572 0.9533 0.9549 0.9533 0.9535 0.9549 0.9533 0.9535
573 0.9539 0.9559 0.9539 0.9543 0.9559 0.9539 0.9543
574 0.9500 0.9518 0.9500 0.9503 0.9518 0.9500 0.9503
575 0.9533 0.9547 0.9533 0.9535 0.9547 0.9533 0.9535
576 0.9546 0.9558 0.9546 0.9547 0.9558 0.9546 0.9547
577 0.9520 0.9539 0.9520 0.9522 0.9539 0.9520 0.9522
578 0.9480 0.9506 0.9480 0.9485 0.9506 0.9480 0.9485
579 0.9520 0.9541 0.9520 0.9522 0.9541 0.9520 0.9522
580 0.9513 0.9526 0.9513 0.9514 0.9526 0.9513 0.9514
581 0.9533 0.9546 0.9533 0.9535 0.9546 0.9533 0.9535
582 0.9526 0.9548 0.9526 0.9530 0.9548 0.9526 0.9530
583 0.9520 0.9542 0.9520 0.9524 0.9542 0.9520 0.9524
584 0.9513 0.9523 0.9513 0.9514 0.9523 0.9513 0.9514
585 0.9533 0.9553 0.9533 0.9536 0.9553 0.9533 0.9536
586 0.9513 0.9536 0.9513 0.9517 0.9536 0.9513 0.9517
587 0.9520 0.9540 0.9520 0.9523 0.9540 0.9520 0.9523
588 0.9513 0.9523 0.9513 0.9513 0.9523 0.9513 0.9513
589 0.9533 0.9549 0.9533 0.9535 0.9549 0.9533 0.9535
590 0.9513 0.9536 0.9513 0.9517 0.9536 0.9513 0.9517
591 0.9520 0.9540 0.9520 0.9523 0.9540 0.9520 0.9523
592 0.9520 0.9530 0.9520 0.9520 0.9530 0.9520 0.9520
593 0.9539 0.9557 0.9539 0.9542 0.9557 0.9539 0.9542
594 0.9579 0.9597 0.9579 0.9582 0.9597 0.9579 0.9582
595 0.9566 0.9583 0.9566 0.9568 0.9583 0.9566 0.9568
596 0.9533 0.9551 0.9533 0.9536 0.9551 0.9533 0.9536
597 0.9487 0.9521 0.9487 0.9494 0.9521 0.9487 0.9494
598 0.9546 0.9558 0.9546 0.9547 0.9558 0.9546 0.9547
599 0.9526 0.9539 0.9526 0.9529 0.9539 0.9526 0.9529
600 0.9539 0.9556 0.9539 0.9542 0.9556 0.9539 0.9542
601 0.9539 0.9559 0.9539 0.9543 0.9559 0.9539 0.9543
602 0.9500 0.9518 0.9500 0.9503 0.9518 0.9500 0.9503

73
reports/WeightFiltration/attack_values.csv
View File

@@ -1,35 +1,38 @@
target_class,parameter_mia_accuracy,latent_distance_tell,lookalike_accuracy target_class, parameter_mia_accuracy, lookalike_accuracy, A-Dist, JS-Dist
0,0.500000,1.180800,0.958333 0,0.500000,0.979167, 0.239583, 0.537877
1,0.500000,1.279257,0.968750 1,0.500000,0.958333, 0.125000, 0.534234
2,0.500000,1.717911,0.937500 2,0.500000,0.906250, 0.062500, 0.566967
3,0.500000,1.354225,0.989583 3,0.500000,0.968750, 0.104167, 0.561083
4,0.500000,1.308562,0.968750 4,0.500000,0.968750, 0.197917, 0.569236
5,0.500000,2.087495,0.947917 5,0.500000,0.989583, 0.010417, 0.562998
6,0.500000,1.793863,0.927083 6,0.500000,0.979167, 0.145833, 0.534611
7,0.500000,1.213634,0.937500 7,0.500000,0.989583, 0.114583, 0.606172
8,0.500000,1.705230,0.979167 8,0.500000,0.927083, 0.010417, 0.531469
9,0.500000,1.713113,0.979167 9,0.500000,0.854167, 0.229167, 0.538189
10,0.500000,1.358761,0.968750 10,0.500000,0.906250, 0.197917, 0.496725
11,0.500000,1.485312,1.000000 11,0.500000,0.906250, 0.031250, 0.552913
12,0.500000,1.393625,0.947917 12,0.500000,0.906250, 0.010417, 0.515257
13,0.500000,1.677361,0.947917 13,0.500000,0.906250, 0.145833, 0.530794
14,0.500000,1.734759,0.906250 14,0.500000,0.947917, 0.062500, 0.546053
15,0.500000,1.715959,0.968750 15,0.500000,0.989583, 0.062500, 0.538760
16,0.500000,1.180073,0.947917 16,0.500000,0.947917, 0.041667, 0.548912
17,0.500000,1.597037,0.947917 17,0.500000,0.927083, 0.114583, 0.536984
18,0.500000,1.491600,0.947917 18,0.500000,0.947917, 0.072917, 0.529271
19,0.500000,1.622301,0.968750 19,0.500000,0.958333, 0.062500, 0.552560
0,0.500000,1.562955,0.968750 0,0.500000,0.968750, 0.208333, 0.536183
0,0.500000,1.526132,0.979167 1,0.500000,0.968750, 0.125000, 0.538881
1,0.500000,1.323757,0.968750 2,0.500000,0.958333, 0.031250, 0.567163
2,0.500000,1.493928,0.958333 3,0.500000,0.979167, 0.114583, 0.562516
3,0.500000,1.643880,0.989583 0,0.500000,0.952083, 0.000000, 0.537566
4,0.500000,1.479018,0.989583 1,0.500000,0.912500, 0.083333, 0.535581
5,0.500000,2.176565,0.958333 2,0.500000,0.958333, 0.010417, 0.564744
6,0.500000,1.620670,0.958333 3,0.500000,0.954167, 0.187500, 0.562515
7,0.500000,0.977823,0.937500 4,0.500000,0.947917, 0.166667, 0.572175
8,0.500000,1.455012,0.968750 5,0.500000,0.952083, 0.000000, 0.561280
9,0.500000,2.145126,0.989583 6,0.500000,0.950000, 0.125000, 0.525977
10,0.500000,1.905214,0.895833 7,0.500000,0.956250, 0.031250, 0.608897
11,0.500000,2.171935,0.958333 8,0.500000,0.960417, 0.020833, 0.531221
12,0.500000,1.761187,0.937500 9,0.500000,0.945833, 0.197917, 0.534556
10,0.500000,0.937500, 0.135417, 0.490381
11,0.500000,0.925000, 0.239583, 0.548656
12,0.500000,0.954167, 0.104167, 0.515086
1 target_class parameter_mia_accuracy latent_distance_tell lookalike_accuracy A-Dist JS-Dist
2 0 0.500000 1.180800 0.958333 0.979167 0.239583 0.537877
3 1 0.500000 1.279257 0.968750 0.958333 0.125000 0.534234
4 2 0.500000 1.717911 0.937500 0.906250 0.062500 0.566967
5 3 0.500000 1.354225 0.989583 0.968750 0.104167 0.561083
6 4 0.500000 1.308562 0.968750 0.197917 0.569236
7 5 0.500000 2.087495 0.947917 0.989583 0.010417 0.562998
8 6 0.500000 1.793863 0.927083 0.979167 0.145833 0.534611
9 7 0.500000 1.213634 0.937500 0.989583 0.114583 0.606172
10 8 0.500000 1.705230 0.979167 0.927083 0.010417 0.531469
11 9 0.500000 1.713113 0.979167 0.854167 0.229167 0.538189
12 10 0.500000 1.358761 0.968750 0.906250 0.197917 0.496725
13 11 0.500000 1.485312 1.000000 0.906250 0.031250 0.552913
14 12 0.500000 1.393625 0.947917 0.906250 0.010417 0.515257
15 13 0.500000 1.677361 0.947917 0.906250 0.145833 0.530794
16 14 0.500000 1.734759 0.906250 0.947917 0.062500 0.546053
17 15 0.500000 1.715959 0.968750 0.989583 0.062500 0.538760
18 16 0.500000 1.180073 0.947917 0.041667 0.548912
19 17 0.500000 1.597037 0.947917 0.927083 0.114583 0.536984
20 18 0.500000 1.491600 0.947917 0.072917 0.529271
21 19 0.500000 1.622301 0.968750 0.958333 0.062500 0.552560
22 0 0.500000 1.562955 0.968750 0.208333 0.536183
23 0 1 0.500000 1.526132 0.979167 0.968750 0.125000 0.538881
24 1 2 0.500000 1.323757 0.968750 0.958333 0.031250 0.567163
25 2 3 0.500000 1.493928 0.958333 0.979167 0.114583 0.562516
26 3 0 0.500000 1.643880 0.989583 0.952083 0.000000 0.537566
27 4 1 0.500000 1.479018 0.989583 0.912500 0.083333 0.535581
28 5 2 0.500000 2.176565 0.958333 0.010417 0.564744
29 6 3 0.500000 1.620670 0.958333 0.954167 0.187500 0.562515
30 7 4 0.500000 0.977823 0.937500 0.947917 0.166667 0.572175
31 8 5 0.500000 1.455012 0.968750 0.952083 0.000000 0.561280
32 9 6 0.500000 2.145126 0.989583 0.950000 0.125000 0.525977
33 10 7 0.500000 1.905214 0.895833 0.956250 0.031250 0.608897
34 11 8 0.500000 2.171935 0.958333 0.960417 0.020833 0.531221
35 12 9 0.500000 1.761187 0.937500 0.945833 0.197917 0.534556
36 10 0.500000 0.937500 0.135417 0.490381
37 11 0.500000 0.925000 0.239583 0.548656
38 12 0.500000 0.954167 0.104167 0.515086

35
reports/WeightFiltration/attack_values_int.csv Normal file
View File

@@ -0,0 +1,35 @@
target_class,parameter_mia_accuracy,latent_distance_tell,lookalike_accuracy
0,0.500000,1.180800,0.958333
1,0.500000,1.279257,0.968750
2,0.500000,1.717911,0.937500
3,0.500000,1.354225,0.989583
4,0.500000,1.308562,0.968750
5,0.500000,2.087495,0.947917
6,0.500000,1.793863,0.927083
7,0.500000,1.213634,0.937500
8,0.500000,1.705230,0.979167
9,0.500000,1.713113,0.979167
10,0.500000,1.358761,0.968750
11,0.500000,1.485312,1.000000
12,0.500000,1.393625,0.947917
13,0.500000,1.677361,0.947917
14,0.500000,1.734759,0.906250
15,0.500000,1.715959,0.968750
16,0.500000,1.180073,0.947917
17,0.500000,1.597037,0.947917
18,0.500000,1.491600,0.947917
19,0.500000,1.622301,0.968750
0,0.500000,1.562955,0.968750
0,0.500000,1.526132,0.979167
1,0.500000,1.323757,0.968750
2,0.500000,1.493928,0.958333
3,0.500000,1.643880,0.989583
4,0.500000,1.479018,0.989583
5,0.500000,2.176565,0.958333
6,0.500000,1.620670,0.958333
7,0.500000,0.977823,0.937500
8,0.500000,1.455012,0.968750
9,0.500000,2.145126,0.989583
10,0.500000,1.905214,0.895833
11,0.500000,2.171935,0.958333
12,0.500000,1.761187,0.937500
1 target_class parameter_mia_accuracy latent_distance_tell lookalike_accuracy
2 0 0.500000 1.180800 0.958333
3 1 0.500000 1.279257 0.968750
4 2 0.500000 1.717911 0.937500
5 3 0.500000 1.354225 0.989583
6 4 0.500000 1.308562 0.968750
7 5 0.500000 2.087495 0.947917
8 6 0.500000 1.793863 0.927083
9 7 0.500000 1.213634 0.937500
10 8 0.500000 1.705230 0.979167
11 9 0.500000 1.713113 0.979167
12 10 0.500000 1.358761 0.968750
13 11 0.500000 1.485312 1.000000
14 12 0.500000 1.393625 0.947917
15 13 0.500000 1.677361 0.947917
16 14 0.500000 1.734759 0.906250
17 15 0.500000 1.715959 0.968750
18 16 0.500000 1.180073 0.947917
19 17 0.500000 1.597037 0.947917
20 18 0.500000 1.491600 0.947917
21 19 0.500000 1.622301 0.968750
22 0 0.500000 1.562955 0.968750
23 0 0.500000 1.526132 0.979167
24 1 0.500000 1.323757 0.968750
25 2 0.500000 1.493928 0.958333
26 3 0.500000 1.643880 0.989583
27 4 0.500000 1.479018 0.989583
28 5 0.500000 2.176565 0.958333
29 6 0.500000 1.620670 0.958333
30 7 0.500000 0.977823 0.937500
31 8 0.500000 1.455012 0.968750
32 9 0.500000 2.145126 0.989583
33 10 0.500000 1.905214 0.895833
34 11 0.500000 2.171935 0.958333
35 12 0.500000 1.761187 0.937500