Steganografiya quyidagi sohalarda qo'llaniladi, lekin ular bilan cheklanmaydi

test_watermark_sklearn(X, y, base_model)

# Ridge Classifier
print('\n\nRidgeClassifier\n')
base_model = RidgeClassifier(alpha=.5)
test_watermark_sklearn(X, y, base_model)

# Logistic Regression
print('\n\nLogistic Regression\n')
base_model = LogisticRegression()
test_watermark_sklearn(X, y, base_model)


import string
import random
from sklearn.ensemble import RandomForestClassifier
from sklearn.linear_model import RidgeClassifier, LogisticRegression
from sklearn.svm import SVC
import pandas as pd
from utils import test_watermark_sklearn
from sklearn.metrics import accuracy_score

from mlmodelwatermarking.verification import verify
from sklearn.model_selection import train_test_split
from warnings import simplefilter
from math import floor
import numpy as np

from mlmodelwatermarking.marklearn import Trainer
from mlmodelwatermarking import TrainingWMArgs

simplefilter(action='ignore', category=UserWarning)


def default_key(length: int):
elements = string.ascii_uppercase + string.digits
return ''.join(random.choices(elements, k=length))


if __name__ == '__main__':
# Loading classification data
df = pd.read_csv('./data/malware_detection.csv')
X = df.drop(["Label"], axis=1)
y = df['Label'].replace('malicious', 0)\
.replace('non-malicious', 1).values

# Random Forest Classifier
print('RANDOM FOREST CLASSIFIER\n')
base_model = RandomForestClassifier(max_depth=1000, random_state=42)

X_train, X_test, y_train, y_test = train_test_split(X,