12 KiB
12 KiB
In [1]:
import numpy as np
import json
import requests
from sklearn.linear_model import LogisticRegression
import os
# Replace <EVALUATOR_IP> and <PORT> with the correct values
evaluator_base_url = "http://154.57.164.81:30900"
# Example: evaluator_base_url = "http://127.0.0.1:5000"
In [2]:
dataset_filename = "label_flipping_dataset.npz"
try:
data = np.load(dataset_filename)
X_train = data["Xtr"]
y_train = data["ytr"]
X_test = data["Xte"]
y_test = data["yte"]
print("Data loaded successfully from single .npz file.")
print(f"X_train shape: {X_train.shape}")
print(f"y_train shape: {y_train.shape}")
print(f"X_test shape: {X_test.shape}")
print(f"y_test shape: {y_test.shape}")
data.close()
except FileNotFoundError:
print(f"Error: Dataset file '{dataset_filename}' not found.")
print("Make sure the .npz data file is in the correct directory.")
raise
except KeyError as e:
print(f"Error: Could not find expected array key '{e}' in the .npz file.")
raiseData loaded successfully from single .npz file. X_train shape: (700, 2) y_train shape: (700,) X_test shape: (300, 2) y_test shape: (300,)
In [3]:
# Implement your attack code in this stub
def flip_labels(y, poison_percentage, seed):
if not 0 <= poison_percentage <= 1:
raise ValueError("poison_percentage must be between 0 and 1.")
n_samples = len(y)
n_to_flip = int(n_samples * poison_percentage)
if n_to_flip == 0:
print("Warning: Poison percentage is 0 or too low to flip any labels.")
# Return unchanged labels and empty indices if no flips are needed
return y.copy(), np.array([], dtype=int)
# Use the defined SEED for the random number generator
rng_instance = np.random.default_rng(seed)
# Select unique indices to flip
flipped_indices = rng_instance.choice(n_samples, size=n_to_flip, replace=False)
# Create a copy to avoid modifying the original array
y_poisoned = y.copy()
# Get the original labels at the indices we are about to flip
original_labels_at_flipped = y_poisoned[flipped_indices]
# Apply the flip: if original was 0, set to 1; otherwise (if 1), set to 0
y_poisoned[flipped_indices] = np.where(original_labels_at_flipped == 0, 1, 0)
print(f"Flipping {n_to_flip} labels ({poison_percentage * 100:.1f}%).")
return y_poisoned, flipped_indices
# ------------------------------------------------------------------------
# --- The rest is templated and you should not need to change anything ---
# ------------------------------------------------------------------------
poison_rate = 0.60
random_seed = 1337
y_train_poisoned, flipped_idx = flip_labels(y_train, poison_rate, random_seed)
print(f"Shape of poisoned labels: {y_train_poisoned.shape}")
print(f"Number of labels flipped: {len(flipped_idx)}")
print(f"Original labels at flipped indices (first 5): {y_train[flipped_idx[:5]]}")
print(
f"Poisoned labels at flipped indices (first 5): {y_train_poisoned[flipped_idx[:5]]}"
)
Flipping 420 labels (60.0%). Shape of poisoned labels: (700,) Number of labels flipped: 420 Original labels at flipped indices (first 5): [1 1 0 1 1] Poisoned labels at flipped indices (first 5): [0 0 1 0 0]
In [4]:
model = LogisticRegression(random_state=random_seed)
model.fit(X_train, y_train_poisoned)
print("Model trained successfully on poisoned data.")
weights = model.coef_
intercept = model.intercept_
print(f"Extracted weights (shape): {weights.shape}")
print(f"Extracted intercept (shape): {intercept.shape}")Model trained successfully on poisoned data. Extracted weights (shape): (1, 2) Extracted intercept (shape): (1,)
In [5]:
health_check_url = f"{evaluator_base_url}/health"
print(f"Checking evaluator health at: {health_check_url}")
if "<EVALUATOR_IP>" in evaluator_base_url:
print("\n--- WARNING ---")
print(
"Please update the 'evaluator_base_url' variable with the correct IP and Port before running!"
)
print("-------------")
else:
try:
response = requests.get(health_check_url, timeout=10)
response.raise_for_status()
health_status = response.json()
print("\n--- Health Check Response ---")
print(f"Status: {health_status.get('status', 'N/A')}")
print(f"Message: {health_status.get('message', 'No message received.')}")
if health_status.get("status") != "healthy":
print(
"\nWarning: Evaluator service reported an unhealthy status. It might still be starting up or encountered an issue (like loading data)."
)
except requests.exceptions.ConnectionError as e:
print(f"\nConnection Error: Could not connect to {health_check_url}.")
print("Please check:")
print(" 1. The evaluator URL (IP address and port) is correct.")
print(" 2. The evaluator Docker container is running.")
print(
" 3. There are no network issues (firewalls, etc.) blocking the connection."
)
except requests.exceptions.Timeout:
print(f"\nTimeout Error: The request to {health_check_url} timed out.")
print(
"The server might be taking too long to respond or there could be network issues."
)
except requests.exceptions.RequestException as e:
print(f"\nError during health check request: {e}")
print("Check the URL format and ensure the server is running.")
except json.JSONDecodeError:
print("\nError: Could not decode JSON response from health check.")
print("The server might have sent an invalid response.")
print(
f"Raw response status: {response.status_code}, Raw response text: {response.text}"
)
except Exception as e:
print(f"\nAn unexpected error occurred during health check: {e}")Checking evaluator health at: http://154.57.164.81:30900/health --- Health Check Response --- Status: healthy Message: Evaluator API running.
In [6]:
evaluator_url = f"{evaluator_base_url}/evaluate"
payload = {"weights": weights.tolist(), "intercept": intercept.tolist()}
print(f"Attempting submission to: {evaluator_url}")
if "<EVALUATOR_IP>" in evaluator_base_url:
print("\n--- WARNING ---")
print(
"Please update the 'evaluator_base_url' variable with the correct IP address and Port before running this cell!"
)
print("-------------")
else:
print(f"Payload: {json.dumps(payload)}")
try:
response = requests.post(evaluator_url, json=payload, timeout=30)
response.raise_for_status()
result = response.json()
print("\n--- Evaluator Response ---")
if result.get("success"):
print("Attack Successful!")
print(f"Accuracy evaluated by server: {result.get('accuracy'):.4f}")
print(f"Flag: {result.get('flag')}")
else:
print("Evaluation Failed.")
accuracy_val = result.get("accuracy")
accuracy_str = f"{accuracy_val:.4f}" if accuracy_val is not None else "N/A"
print(f"Accuracy evaluated by server: {accuracy_str}")
print(f"Message: {result.get('message')}")
print(
"Hints: Did you poison exactly 60% of the data? Did you use the seed 1337 for flipping labels?"
)
except requests.exceptions.ConnectionError as e:
print(
f"\nConnection Error: Could not connect to the evaluator API at {evaluator_url}."
)
print("Please check:")
print(" 1. The evaluator URL (IP address and port) is correct.")
print(" 2. The evaluator Docker instance is spawned.")
print(
" 3. There are no network issues (firewalls, etc.) blocking the connection."
)
except requests.exceptions.Timeout:
print(f"\nTimeout Error: The request to {evaluator_url} timed out.")
print("The server might be slow, or there could be network issues.")
except requests.exceptions.RequestException as e:
print(f"\nError connecting to evaluator API: {e}")
print("Please check the evaluator URL and ensure the instance is spawned.")
except json.JSONDecodeError:
print("\nError decoding JSON response from the evaluator.")
print("The server might have sent an invalid response.")
print(
f"Raw response status: {response.status_code}, Raw response text: {response.text}"
)
except Exception as e:
print(f"\nAn unexpected error occurred: {e}")Attempting submission to: http://154.57.164.81:30900/evaluate
Payload: {"weights": [[-0.12683149742377367, 0.02548467051700444]], "intercept": [0.29202685494234054]}
--- Evaluator Response ---
Attack Successful!
Accuracy evaluated by server: 0.0033
Flag: HTB{l4b3l_fl1pp1ng_pwnz_d3f4ult}
In [ ]: