Logo: Signin
logo = """ #################################################################################################### ############################ # ##################################################################### #### ## ########### ## ## ########### ######## ###### #### ## ######### ########## ###### ### #### ######## ######## ########################## ############# ############ ## #### ########### ###### ########### ############## ############ ################### ### ############# ##### ########### ############# ############ ################### ### ############## #### ############# ############# ############# ################### ### ############## ### ############# ############ ############## ################# ### ############## ### ############### ############ ############### ############### ### ############## #### ############### ############ ################# ############# ### ############# #### ############### ########### #################### ########## ### ############ ##### ############### ########### ###################### ######### ### #### ###### ############## ########### ######################## ####### ### #### ######## ############# ########### ########################## ###### ### ######### ######### ############# ############ ########################## ###### ### ########## ######## ########### ############# ########################## ###### ### ########### ######## ######### ############# ############# ######### ###### ### ############ ######### ###### ############ ############### #### ####### ### ############## ########### ############ ######### ######### #### ############################### ############################################## ############ #################################################################################################### """
Logo: 2024
有长度限制，考虑对logo进行压缩。
首先把第一行和最后一行换成#*100
中间的内容通过相同字符长度进行压缩。正好是‘#’和’ ’间隔。
这里用39以后的字符防止出现单引号/双引号。
from RestrictedPython import compile_restricted, safe_builtins from RestrictedPython.Eval import default_guarded_getitem from RestrictedPython.Guards import full_write_guard ROIS_LOGO = """ #################################################################################################### ############################ # ##################################################################### #### ## ########### ## ## ########### ######## ###### #### ## ######### ########## ###### ### #### ######## ######## ########################## ############# ############ ## #### ########### ###### ########### ############## ############ ################### ### ############# ##### ########### ############# ############ ################### ### ############## #### ############# ############# ############# ################### ### ############## ### ############# ############ ############## ################# ### ############## ### ############### ############ ############### ############### ### ############## #### ############### ############ ################# ############# ### ############# #### ############### ########### #################### ########## ### ############ ##### ############### ########### ###################### ######### ### #### ###### ############## ########### ######################## ####### ### #### ######## ############# ########### ########################## ###### ### ######### ######### ############# ############ ########################## ###### ### ########## ######## ########### ############# ########################## ###### ### ########### ######## ######### ############# ############# ######### ###### ### ############ ######### ###### ############ ############### #### ####### ### ############## ########### ############ ######### ######### #### ############################### ############################################## ############ #################################################################################################### """ logo = "".join(ROIS_LOGO.strip().split("\n")[1:-1]) from string import printable arr = [] lastchr = logo[0] count = 1 for i in range(1,len(logo)): if logo[i] == lastchr: count += 1 else: arr.append(count) lastchr = logo[i] count = 1 arr.append(count) payload = bytearray([i+39 for i in arr]).decode(); print(len(payload)) loc = {} cmdline = f"""t=100 c=f"{payload[::-1]}" i=229 s="#"*t while i: i=i-1 s=s+"# "[i&1]*(ord(c[i])-39) s=s+'#'*t r='' while s: r=r+s[:t]+'\\n';s=s[t:] logo=r""" exec(compile_restricted(cmdline,"<inline>","exec"),{ "__builtins__": safe_builtins, "_getitem_": default_guarded_getitem, "_write_": full_write_guard, },loc) assert ROIS_LOGO.strip() == loc["logo"].strip() assert len(cmdline) < len(ROIS_LOGO) * .2024 print("[+] Payload Generated: ",cmdline) print("[+] Length:",len(cmdline))
sec-image
import os import shutil from PIL import Image import numpy as np for i in range(10): if os.path.exists(f'flag{i}'): shutil.rmtree(f'flag{i}') os.mkdir(f'flag{i}') p=np.array(Image.open(f'flag{i}.png')) Image.fromarray(p[::2,::2]).save(f'flag{i}/0.png') Image.fromarray(p[::2,1::2]).save(f'flag{i}/1.png') Image.fromarray(p[1::2,::2]).save(f'flag{i}/2.png') Image.fromarray(p[1::2,1::2]).save(f'flag{i}/3.png')
ezlogin
import torch import torch.nn as nn import torch.nn.functional as F import torchvision import torchvision.transforms as transforms import os from sklearn.manifold import TSNE from torch.utils.data import DataLoader import matplotlib.pyplot as plt import numpy as np import imageio as iio import warnings import functools import signal import hashlib import time import torchattacks from PIL import Image from torchvision import transforms warnings.filterwarnings("ignore") device = torch.device("cuda") num_epochs = 50 batch_size = 512 learning_rate = 1e-4 # train_data = dataset = torchvision.datasets.EMNIST(root='data/', # download=True, # transform=transforms.ToTensor(), # train=True, # split='balanced') # test_data = dataset = torchvision.datasets. EMNIST(root='data/', # download=True, # train=False, # transform=transforms.ToTensor(),split='balanced') # train_loader = DataLoader(train_data, batch_size=batch_size, shuffle=True) # test_loader = DataLoader(test_data, 1, shuffle=False) class CNN(nn.Module): def __init__(self): super(CNN, self).__init__() self.conv1 = nn.Conv2d(1, 2, 3, padding=1) self.conv2 = nn.Conv2d(2, 8, 3, padding=1) self.conv3 = nn.Conv2d(8, 32, 3, padding=1) self.pool = nn.MaxPool2d(2, 2) self.fc1 = nn.Linear(32 * 3 * 3, 1024) self.fc2 = nn.Linear(1024, 512) self.fc3 = nn.Linear(512, 256) self.fc4 = nn.Linear(256, 128) self.fc5 = nn.Linear(128, 47) def forward(self, x): x = self.pool(self.conv1(x)) x = self.pool(self.conv2(x)) x = self.pool(self.conv3(x)) x = x.view(-1, 32 * 3 * 3) x = F.relu(self.fc1(x)) x = F.relu(self.fc2(x)) x = F.relu(self.fc3(x)) x = F.relu(self.fc4(x)) x = self.fc5(x) return x if __name__ == "a": try: loaded = False try: model = CNN().to(device) model.load_state_dict(torch.load('model.pt.state')) loaded = True except: model = CNN().to(device) try: feature = torch.tensor([[-6.19499969e+01, -1.56200895e+01, -3.52624054e+01, -1.34233132e-01, -6.48261490e+01, -1.47979248e+02, -5.15059547e+01, -1.14444227e+01, 4.33434563e+01, -3.69645386e+01, 2.00579977e+00, 4.74611549e+01, -6.33986130e+01, -1.57887411e+01, -2.87570419e+01, -5.35021248e+01, -1.73028266e+00, -3.61370316e+01, -7.58331375e+01, -7.46535110e+01, -7.24118347e+01, -4.76773834e+01, 6.51892662e+00, -5.07196846e+01, -1.03041328e+02, 4.72574463e+01, 9.03826065e+01, 5.30947495e+01, -5.03226738e+01, -1.50200531e+02, -3.46447792e+01, -4.23207245e+01, 6.44030609e+01, -5.05351334e+01, -4.11206970e+01, -2.18300457e+01, 2.70750694e+01, -1.00022865e+02, 3.77698517e+01, -3.60703392e+01, -6.88536682e+01, 1.16945248e+01, -4.62400284e+01, -4.79546585e+01, 6.10636101e+01, -1.12650543e+02, -1.34837357e+02,]], dtype=torch.float32).numpy() print("input image bytes(28 * 28):") # readres = readNumpy(28 * 28, np.uint8) readres=np.array([[154,254, 0,102,232,137, 76,112, 68, 52,137, 60,108,239, 46, 72,170, 226,248,107,115,194, 55,195,223,150, 48,237], [162, 81,129, 26, 48,211, 61,196, 14,142,181, 20, 32,150,222,144, 43, 120, 63, 1,118, 99, 54, 0, 29,173,174,104], [ 0, 37, 98,113,151,219,115, 68, 75, 41, 11, 97,177,103,160,157, 71, 136,156,124,142,142,152, 3, 20, 15, 59,150], [ 35,156, 66, 77,111, 12, 0,154,112,187,138,222,127,171, 85, 98,125, 105, 50, 69,112, 96,143, 69, 70,165, 26,154], [ 90,168,120,114, 95, 4, 61, 80,215,213,162, 61,219, 0, 92, 99, 42, 142,122,185,154,195,172,164,145,111, 11, 53], [119, 54, 89,211,131, 28,242,210,242, 69, 36,249, 86,165,218,203, 97, 232, 52,136, 66,110,136,106,190, 1,176, 89], [219, 55,170, 34,206,174,115,252,226, 55, 94,158, 37, 62,100,153, 0, 179,254,254, 33,249,204,244,205,186, 78,197], [254, 54,186,238,214,128,147, 39,168,112,138,143,162, 75,120, 42,199, 235,116, 18,184,251, 25,132,252,229,202,251], [ 23,237,199,116,125, 41,105,205, 55,111,165, 13,197,254, 7, 37,144, 235,183, 0,188, 96, 8, 1,193,124, 26,220], [193,156,152,195,254,125,231, 23,217,219,220, 0,183,193,222,250, 52, 187,213, 97, 96, 18, 24, 0,223, 41, 67,139], [112,254, 76,236,219,173,133,129, 86,171,150, 80,210, 74,189,190,160, 98,130,254,227, 19, 88, 81,225,187, 8,113], [212,205,221, 46, 0,210,111, 21,114,193, 48,214,188,158, 41, 27,238, 222,234,106,192, 91, 25,131,150, 75,159, 31], [ 16,150,140,183, 7, 89,151, 11, 75, 2, 60,166, 14,140, 36, 50,150, 186, 17,107,184, 61,201, 52, 31,176, 38, 20], [ 0,158,140,121, 37,155, 4,182,137,167,141, 61, 93,145, 15,156, 30, 221,109,164,167, 34,232,122, 52,167,217, 75], [243, 69,130,189,105,167,195,115, 31, 35, 58,170, 84,197, 85,225,216, 167,119, 32, 26,208, 19,123, 98,130,194, 57], [203, 50,225,242,228,153, 39,231, 75, 79,143, 16,246, 13, 5, 83,229, 59, 12,212,254, 39, 90,246,135,133,110,158], [ 63,146, 40,131,238,136,137, 45, 95,149, 11,119, 55,218,104,254,199, 101, 83,220, 20,124,243,114, 50,219, 20, 63], [ 32,103,114,188,230, 39, 0,132, 97,254,254,165,175,144, 0,207,198, 13,211,189,100,157,145,161,117, 90,120,137], [254, 60, 92, 5,202,159,167, 66, 75,165,252, 56,115, 8,233,240,180, 56, 65,179, 87, 5,107,162,151,155,222, 44], [133, 73,223,189,205, 0, 7, 59, 52, 0, 76,249,120, 0, 35, 94,161, 160, 0,167, 69,225, 0, 71,199,199,250,153], [ 38,154, 84,214, 0,162,193,238,158,159, 65,170, 23, 32, 80,186, 80, 90,163,165,175,212, 57,204,174, 81,184,183], [203, 27, 84, 65, 66,119,221,215, 60, 55, 3, 39,166, 33, 75,142, 96, 76,179, 94,181, 15,220,107,122,130, 37, 51], [ 99,247,113,161, 77,188,136, 74, 0, 81, 94, 6, 72, 51, 81, 89, 84, 55,169, 97, 42,183,209,112,229, 2,157, 49], [199,181, 46,111,115, 51,151,150, 65,103, 86,167, 79,130,205, 47,110, 73, 95,177,221, 19,110, 86,142, 63, 85,140], [174,189,224,133,251, 68, 0, 0, 0, 0,113, 24,145, 25,153,124,163, 106, 93,250, 60,117,253, 88, 58, 54,160,163], [ 95, 36,111,253,199,192,138,209, 54, 7, 47,174, 67,188, 77,103,254, 223, 39, 55,196, 12,127,139,143,132, 67,233], [ 64, 26,125,228,129,254,127,254,147,119,114, 85, 89, 6, 67,209,149, 254,144,236,254,205, 93, 20, 27,120,159,149], [143,109, 32,250,190,121, 85,254,113,253, 66, 20, 74,106,189,121,117, 215, 43,254,190, 76,239,193,100,131,147, 32]]) if isinstance(readres, None.__class__): print("Error!") os._exit(0) img_new_tensor = torch.tensor(readres.reshape(28, 28), dtype=torch.float32) / 255.0 img_new_tensor = img_new_tensor.unsqueeze(0) model.eval() feature_new = model(img_new_tensor.to(device)).detach().cpu().numpy() dist = np.sqrt(np.sum((feature - feature_new)**2)) print("distance:", dist) if dist < 6: print("flag{123213123232131231232131}") except: __import__("traceback").print_exc() print("exit!") os._exit(0) except: print("exit!") os._exit(0) def crack(): try: model = CNN().to(device) model.load_state_dict(torch.load('model.pt.state')) loaded = True except: model = CNN().to(device) model.eval() attack=torchattacks.FGSM(model,eps=1/255.0) attack.set_mode_targeted_by_label() dist=1e300 # img=Image.new('L',(25,25),128) # img=transforms.ToTensor()(img) feature=feature = torch.tensor([[-6.19499969e+01, -1.56200895e+01, -3.52624054e+01, -1.34233132e-01, -6.48261490e+01, -1.47979248e+02, -5.15059547e+01, -1.14444227e+01, 4.33434563e+01, -3.69645386e+01, 2.00579977e+00, 4.74611549e+01, -6.33986130e+01, -1.57887411e+01, -2.87570419e+01, -5.35021248e+01, -1.73028266e+00, -3.61370316e+01, -7.58331375e+01, -7.46535110e+01, -7.24118347e+01, -4.76773834e+01, 6.51892662e+00, -5.07196846e+01, -1.03041328e+02, 4.72574463e+01, 9.03826065e+01, 5.30947495e+01, -5.03226738e+01, -1.50200531e+02, -3.46447792e+01, -4.23207245e+01, 6.44030609e+01, -5.05351334e+01, -4.11206970e+01, -2.18300457e+01, 2.70750694e+01, -1.00022865e+02, 3.77698517e+01, -3.60703392e+01, -6.88536682e+01, 1.16945248e+01, -4.62400284e+01, -4.79546585e+01, 6.10636101e+01, -1.12650543e+02, -1.34837357e+02,]], dtype=torch.float32).to(device) tensored = torch.rand(1, 1, 28, 28).to(device) tensored=torch.tensor([[[[ 6.0288e-01, 1.1362e+00, 5.9489e-05, 4.2852e-01, 9.1239e-01, 5.7314e-01, 3.1315e-01, 4.4792e-01, 2.8514e-01, 2.0634e-01, 5.4083e-01, 2.2748e-01, 4.1847e-01, 9.3224e-01, 1.8290e-01, 2.8779e-01, 6.7119e-01, 8.8639e-01, 9.6913e-01, 4.1018e-01, 4.4693e-01, 7.5400e-01, 2.2920e-01, 7.9548e-01, 9.1583e-01, 5.9450e-01, 1.9110e-01, 9.1565e-01], [ 6.1580e-01, 3.3205e-01, 5.0228e-01, 1.0933e-01, 2.0404e-01, 8.3139e-01, 2.2867e-01, 8.0309e-01, 7.0727e-02, 6.0615e-01, 7.1967e-01, 8.6963e-02, 1.2258e-01, 5.8602e-01, 8.7928e-01, 5.6752e-01, 1.7299e-01, 4.7335e-01, 2.8381e-01, -7.0812e-02, 5.0558e-01, 3.9065e-01, 2.2163e-01, -7.6157e-02, 1.5868e-01, 7.4678e-01, 7.0252e-01, 4.0050e-01], [-7.1028e-02, 1.1317e-01, 3.6623e-01, 4.1133e-01, 5.8208e-01, 8.3339e-01, 4.8500e-01, 3.0252e-01, 3.1078e-01, 1.8317e-01, 5.6958e-02, 3.8725e-01, 7.0658e-01, 3.9187e-01, 6.2303e-01, 6.2260e-01, 2.7760e-01, 5.3419e-01, 6.4757e-01, 5.1142e-01, 5.9661e-01, 5.4655e-01, 6.1045e-01, -4.4963e-03, 1.2481e-01, 9.2860e-02, 2.6914e-01, 5.8491e-01], [ 1.4626e-01, 6.0557e-01, 2.5293e-01, 3.1025e-01, 4.7795e-01, 5.3919e-02, -6.1463e-03, 6.0654e-01, 4.2319e-01, 7.4492e-01, 5.5084e-01, 8.6132e-01, 4.9652e-01, 6.6520e-01, 3.3005e-01, 3.7835e-01, 5.1184e-01, 4.1918e-01, 1.5690e-01, 2.5480e-01, 4.5393e-01, 3.7624e-01, 5.4659e-01, 2.7696e-01, 2.4687e-01, 7.1665e-01, 1.2837e-01, 6.1163e-01], [ 3.4478e-01, 6.6759e-01, 4.8163e-01, 4.8742e-01, 3.8671e-01, 4.2637e-02, 2.8213e-01, 3.2243e-01, 8.7444e-01, 8.3748e-01, 6.3151e-01, 2.3735e-01, 8.7876e-01, -3.9210e-02, 3.9525e-01, 3.7733e-01, 1.6370e-01, 5.4120e-01, 4.7166e-01, 7.5188e-01, 6.1869e-01, 8.4919e-01, 6.7780e-01, 6.4300e-01, 5.7504e-01, 4.7048e-01, 5.7314e-02, 2.1222e-01], [ 4.7097e-01, 2.0323e-01, 3.3424e-01, 8.1913e-01, 5.3727e-01, 9.9891e-02, 1.0234e+00, 8.4038e-01, 1.1350e+00, 2.5937e-01, 1.4993e-01, 1.0317e+00, 3.5118e-01, 6.5605e-01, 8.7078e-01, 7.9255e-01, 3.8220e-01, 8.6914e-01, 2.2715e-01, 5.4343e-01, 2.7454e-01, 4.7493e-01, 5.3477e-01, 4.1217e-01, 7.1557e-01, -1.9818e-02, 6.7947e-01, 3.4895e-01], [ 8.1369e-01, 2.2739e-01, 6.2416e-01, 1.3195e-01, 8.0578e-01, 6.8169e-01, 4.5694e-01, 9.9322e-01, 9.0871e-01, 2.2177e-01, 3.5772e-01, 6.1249e-01, 1.4451e-01, 2.1517e-01, 3.8865e-01, 6.2094e-01, -1.3825e-01, 7.2339e-01, 1.0589e+00, 1.2942e+00, 9.6191e-02, 1.0701e+00, 8.1473e-01, 1.0005e+00, 8.0526e-01, 7.1779e-01, 3.1163e-01, 7.7659e-01], [ 9.6146e-01, 1.6313e-01, 6.8609e-01, 9.1936e-01, 8.2348e-01, 4.9940e-01, 5.5067e-01, 1.3772e-01, 6.5934e-01, 4.4348e-01, 5.5429e-01, 5.5286e-01, 6.4212e-01, 2.9482e-01, 4.9009e-01, 1.5313e-01, 7.6001e-01, 9.5773e-01, 4.4870e-01, 7.0587e-02, 6.3051e-01, 9.7914e-01, 1.0483e-01, 5.0774e-01, 1.0615e+00, 9.0353e-01, 7.8602e-01, 1.0277e+00], [ 1.1235e-01, 9.0992e-01, 7.2119e-01, 4.3361e-01, 4.7386e-01, 1.4313e-01, 4.0376e-01, 8.1287e-01, 2.3066e-01, 4.5863e-01, 6.8672e-01, 5.9185e-02, 7.6155e-01, 1.0895e+00, 1.0696e-02, 1.1955e-01, 5.9255e-01, 9.0098e-01, 7.3014e-01, -2.2989e-01, 7.0312e-01, 3.3254e-01, -3.0847e-02, 1.3195e-02, 7.2566e-01, 4.7231e-01, 1.1454e-01, 8.7590e-01], [ 7.7808e-01, 6.2572e-01, 6.0420e-01, 7.3975e-01, 9.9742e-01, 4.9199e-01, 8.9071e-01, 1.0273e-01, 8.1529e-01, 8.7398e-01, 8.5837e-01, -9.3574e-02, 7.4139e-01, 7.6849e-01, 8.9423e-01, 1.0978e+00, 2.1246e-01, 7.1140e-01, 7.9466e-01, 3.9209e-01, 3.7501e-01, 7.2755e-02, 9.0850e-02, 2.2957e-02, 8.6281e-01, 1.4462e-01, 2.7115e-01, 5.4973e-01], [ 4.4107e-01, 1.0529e+00, 2.9880e-01, 9.2260e-01, 8.4810e-01, 6.4860e-01, 4.9903e-01, 4.9807e-01, 3.0239e-01, 6.5316e-01, 5.8924e-01, 3.5929e-01, 8.7163e-01, 3.3495e-01, 7.8563e-01, 7.7249e-01, 6.3549e-01, 3.7659e-01, 4.8119e-01, 1.1958e+00, 8.3194e-01, 6.4911e-02, 3.4109e-01, 3.0952e-01, 8.7574e-01, 7.5492e-01, 4.5016e-02, 4.5166e-01], [ 7.9193e-01, 8.0962e-01, 8.3640e-01, 1.7199e-01, 8.5984e-03, 8.0833e-01, 4.3951e-01, 1.0346e-01, 4.5846e-01, 7.6686e-01, 1.8354e-01, 8.5022e-01, 7.3330e-01, 5.7344e-01, 1.2528e-01, 1.0215e-01, 9.3043e-01, 8.7269e-01, 9.3331e-01, 3.8962e-01, 6.9048e-01, 3.3705e-01, 1.0508e-01, 5.4698e-01, 5.9770e-01, 3.4513e-01, 6.2667e-01, 1.2362e-01], [ 6.3149e-02, 5.4332e-01, 5.2946e-01, 7.5011e-01, 2.2203e-02, 3.5547e-01, 5.7432e-01, 4.8437e-02, 2.9004e-01, -1.6281e-01, 2.3475e-01, 6.4043e-01, 4.9892e-02, 5.3533e-01, 1.2014e-01, 1.9383e-01, 5.8046e-01, 7.5284e-01, 1.1419e-01, 4.6636e-01, 7.3830e-01, 2.5486e-01, 8.2636e-01, 2.2003e-01, 1.1708e-01, 7.0863e-01, 1.4147e-01, 7.2529e-02], [-5.0734e-03, 6.2114e-01, 5.3009e-01, 4.7144e-01, 1.8350e-01, 6.1176e-01, -2.7301e-02, 6.4604e-01, 5.3030e-01, 6.4814e-01, 5.6420e-01, 2.6860e-01, 4.0115e-01, 5.5545e-01, 6.3442e-02, 6.0154e-01, 9.4468e-02, 8.3661e-01, 4.2637e-01, 6.2187e-01, 6.8843e-01, 1.5162e-01, 9.3021e-01, 4.7756e-01, 1.6871e-01, 6.7050e-01, 8.4901e-01, 2.9408e-01], [ 9.3974e-01, 2.6410e-01, 5.2163e-01, 7.3979e-01, 4.7082e-01, 6.5093e-01, 7.8358e-01, 4.6225e-01, 1.0183e-01, 1.2838e-01, 2.3783e-01, 6.5826e-01, 3.3046e-01, 7.5239e-01, 3.4042e-01, 8.5900e-01, 8.2837e-01, 6.4222e-01, 4.5357e-01, 1.2401e-01, 7.6350e-02, 7.8813e-01, 8.9284e-02, 4.9061e-01, 3.8516e-01, 5.1263e-01, 7.8971e-01, 2.5493e-01], [ 7.9546e-01, 2.0633e-01, 8.9912e-01, 1.0262e+00, 9.0110e-01, 5.9865e-01, 2.0708e-01, 8.6242e-01, 3.3956e-01, 3.3921e-01, 5.4902e-01, 2.5711e-02, 9.4743e-01, 5.2793e-02, 2.2473e-02, 3.2953e-01, 8.9930e-01, 2.3921e-01, 6.4219e-02, 8.1889e-01, 1.0258e+00, 1.4573e-01, 3.5381e-01, 9.2302e-01, 5.0899e-01, 4.7839e-01, 4.5472e-01, 6.4312e-01], [ 2.6405e-01, 5.7001e-01, 1.5210e-01, 4.7059e-01, 9.5783e-01, 5.2991e-01, 5.2675e-01, 1.5340e-01, 3.6210e-01, 6.2503e-01, 4.0578e-02, 4.7792e-01, 2.2813e-01, 7.9308e-01, 3.9085e-01, 9.9410e-01, 7.8724e-01, 4.1535e-01, 3.3599e-01, 8.6563e-01, 8.9724e-02, 4.7490e-01, 9.5913e-01, 4.2365e-01, 1.8861e-01, 8.2992e-01, 6.7295e-02, 2.5372e-01], [ 1.2108e-01, 3.7644e-01, 4.5913e-01, 7.5106e-01, 9.0900e-01, 1.8568e-01, -2.3777e-02, 5.2048e-01, 3.9792e-01, 9.8551e-01, 1.0665e+00, 6.1337e-01, 6.7489e-01, 5.1058e-01, -1.3618e-02, 7.7188e-01, 7.8193e-01, 5.6600e-02, 8.5688e-01, 7.6902e-01, 4.1159e-01, 6.3866e-01, 4.9157e-01, 6.3980e-01, 4.3434e-01, 3.4617e-01, 4.6667e-01, 5.3848e-01], [ 1.0579e+00, 2.3414e-01, 3.1658e-01, -8.6084e-02, 8.1349e-01, 6.4511e-01, 6.8777e-01, 2.5055e-01, 2.8681e-01, 6.2201e-01, 1.0176e+00, 2.0087e-01, 4.8428e-01, 5.5284e-02, 8.5649e-01, 9.3845e-01, 7.0895e-01, 2.3617e-01, 2.6559e-01, 7.3822e-01, 3.4822e-01, 2.6721e-02, 4.1820e-01, 5.5569e-01, 5.3522e-01, 5.9240e-01, 8.4829e-01, 1.8631e-01], [ 5.0738e-01, 2.6644e-01, 8.5824e-01, 7.5214e-01, 8.4483e-01, -4.5763e-02, 3.8356e-02, 2.4536e-01, 2.2508e-01, 3.1382e-03, 3.6618e-01, 1.1078e+00, 4.5653e-01, -6.4363e-04, 1.5056e-01, 3.6254e-01, 6.6774e-01, 6.5680e-01, -4.6056e-03, 6.3040e-01, 2.7773e-01, 8.9818e-01, 2.2342e-02, 3.0652e-01, 7.3080e-01, 8.3147e-01, 9.7002e-01, 6.1938e-01], [ 1.5946e-01, 6.1125e-01, 3.4396e-01, 8.9923e-01, 1.3574e-02, 6.4339e-01, 7.7911e-01, 8.9100e-01, 6.2722e-01, 5.2762e-01, 2.7600e-01, 6.6731e-01, 1.2502e-01, 1.9134e-01, 3.0154e-01, 7.4043e-01, 3.3239e-01, 3.7606e-01, 6.9304e-01, 6.4669e-01, 6.7373e-01, 8.0201e-01, 2.3439e-01, 8.1852e-01, 6.8174e-01, 3.7022e-01, 7.2829e-01, 7.3067e-01], [ 8.2974e-01, 1.1083e-01, 3.2454e-01, 2.2676e-01, 2.3869e-01, 4.9026e-01, 8.9180e-01, 8.5484e-01, 2.5162e-01, 1.7870e-01, -4.0852e-02, 1.8148e-01, 6.2775e-01, 1.3101e-01, 2.8369e-01, 5.7616e-01, 4.3123e-01, 3.0954e-01, 7.2911e-01, 3.7670e-01, 6.7651e-01, 4.7760e-02, 8.7362e-01, 4.1234e-01, 5.3951e-01, 5.3174e-01, 1.5039e-01, 2.1479e-01], [ 3.8540e-01, 1.0125e+00, 4.5503e-01, 6.5674e-01, 2.8749e-01, 7.5905e-01, 5.4326e-01, 3.2885e-01, -1.0434e-01, 3.5212e-01, 3.8181e-01, 1.8984e-02, 2.4018e-01, 2.2099e-01, 3.4744e-01, 3.6183e-01, 3.5225e-01, 2.1809e-01, 6.7755e-01, 3.7800e-01, 1.7447e-01, 7.0573e-01, 8.0136e-01, 4.7569e-01, 9.8392e-01, 1.6173e-02, 6.1636e-01, 1.9482e-01], [ 7.8370e-01, 7.1147e-01, 2.0800e-01, 4.3978e-01, 4.9866e-01, 2.1853e-01, 6.5248e-01, 5.9993e-01, 2.9794e-01, 4.3272e-01, 3.9274e-01, 6.4736e-01, 3.1330e-01, 5.0629e-01, 8.3233e-01, 1.8387e-01, 4.4380e-01, 2.9646e-01, 3.6449e-01, 7.1266e-01, 8.5293e-01, 7.5409e-02, 3.8425e-01, 3.4207e-01, 5.6922e-01, 2.7579e-01, 3.4487e-01, 5.4443e-01], [ 7.0230e-01, 7.7108e-01, 9.2445e-01, 5.1222e-01, 1.0493e+00, 2.8528e-01, -1.4066e-01, 2.7808e-02, 2.3065e-02, -4.0014e-02, 4.9922e-01, 6.1474e-02, 5.9423e-01, 8.5006e-02, 5.9977e-01, 5.0050e-01, 6.4506e-01, 4.3328e-01, 3.5594e-01, 9.8738e-01, 2.1839e-01, 4.5442e-01, 1.0981e+00, 3.7236e-01, 2.4674e-01, 2.6815e-01, 6.2023e-01, 6.3304e-01], [ 3.7776e-01, 1.5990e-01, 4.6500e-01, 1.0565e+00, 7.6901e-01, 7.7969e-01, 5.5995e-01, 8.5825e-01, 2.0155e-01, -4.3238e-03, 1.8181e-01, 7.4459e-01, 2.8807e-01, 7.6745e-01, 3.2275e-01, 3.9529e-01, 1.0352e+00, 8.8478e-01, 1.5051e-01, 2.3291e-01, 7.9456e-01, 7.7804e-02, 5.0582e-01, 5.6546e-01, 5.7419e-01, 5.5255e-01, 2.5548e-01, 9.1667e-01], [ 2.4563e-01, 8.7151e-02, 4.8151e-01, 8.6393e-01, 4.9966e-01, 1.2342e+00, 4.3573e-01, 1.1570e+00, 5.7784e-01, 4.3662e-01, 4.3337e-01, 3.5117e-01, 3.6724e-01, 3.8910e-02, 2.7161e-01, 8.1971e-01, 5.7101e-01, 1.1323e+00, 5.6091e-01, 9.1287e-01, 1.0213e+00, 8.5086e-01, 3.5015e-01, 6.2480e-02, 9.6676e-02, 4.8563e-01, 6.4755e-01, 6.1303e-01], [ 5.7250e-01, 4.3544e-01, 1.1649e-01, 9.5708e-01, 7.3295e-01, 4.4030e-01, 3.1571e-01, 1.0779e+00, 4.1958e-01, 9.6637e-01, 2.7041e-01, 8.4789e-02, 2.9426e-01, 4.1899e-01, 7.3498e-01, 4.8051e-01, 4.6492e-01, 8.3942e-01, 1.7416e-01, 1.0366e+00, 7.4209e-01, 3.0254e-01, 9.8799e-01, 7.8156e-01, 4.0695e-01, 5.2730e-01, 5.9528e-01, 1.4215e-01]]]], device='cuda:0', requires_grad=True) tensored=torch.tensor([[[[ 6.0183e-01, 1.1331e+00, -1.3945e-05, 4.2430e-01, 9.1187e-01, 5.6688e-01, 3.1034e-01, 4.4660e-01, 2.8221e-01, 2.0750e-01, 5.4614e-01, 2.2767e-01, 4.2403e-01, 9.2924e-01, 1.8320e-01, 2.8885e-01, 6.7083e-01, 8.9040e-01, 9.6996e-01, 4.0850e-01, 4.4688e-01, 7.5400e-01, 2.2421e-01, 7.8975e-01, 9.0751e-01, 5.9550e-01, 1.9180e-01, 9.1863e-01], [ 6.1457e-01, 3.3005e-01, 4.9625e-01, 1.0556e-01, 1.9880e-01, 8.2294e-01, 2.2961e-01, 7.9427e-01, 6.8526e-02, 5.9924e-01, 7.1882e-01, 9.0209e-02, 1.2535e-01, 5.8580e-01, 8.7349e-01, 5.6845e-01, 1.7026e-01, 4.7510e-01, 2.7932e-01, -7.0963e-02, 5.0265e-01, 3.9118e-01, 2.2380e-01, -8.1462e-02, 1.4890e-01, 7.4730e-01, 7.0031e-01, 4.0647e-01], [-7.1190e-02, 1.1456e-01, 3.6527e-01, 4.0955e-01, 5.8353e-01, 8.3534e-01, 4.7966e-01, 2.9547e-01, 3.0880e-01, 1.7921e-01, 5.1481e-02, 3.7765e-01, 7.0470e-01, 3.9020e-01, 6.1937e-01, 6.1260e-01, 2.7703e-01, 5.3320e-01, 6.4587e-01, 5.0856e-01, 5.9247e-01, 5.4697e-01, 6.0998e-01, -1.3610e-03, 1.0878e-01, 9.2086e-02, 2.6374e-01, 5.8851e-01], [ 1.4642e-01, 6.0318e-01, 2.5273e-01, 3.0557e-01, 4.7717e-01, 5.4624e-02, -1.3986e-02, 6.0689e-01, 4.2614e-01, 7.4380e-01, 5.5029e-01, 8.6025e-01, 4.8990e-01, 6.6199e-01, 3.3108e-01, 3.7729e-01, 5.0358e-01, 4.1579e-01, 1.5550e-01, 2.5433e-01, 4.4257e-01, 3.7089e-01, 5.3936e-01, 2.7260e-01, 2.4756e-01, 7.1054e-01, 1.2018e-01, 6.1151e-01], [ 3.4757e-01, 6.6652e-01, 4.8007e-01, 4.8409e-01, 3.8471e-01, 4.1360e-02, 2.8241e-01, 3.1957e-01, 8.7148e-01, 8.3691e-01, 6.3163e-01, 2.4043e-01, 8.7860e-01, -3.9612e-02, 3.9494e-01, 3.7899e-01, 1.5575e-01, 5.3373e-01, 4.6879e-01, 7.3546e-01, 6.1418e-01, 8.3589e-01, 6.7793e-01, 6.4887e-01, 5.7550e-01, 4.6637e-01, 5.5677e-02, 2.1172e-01], [ 4.7226e-01, 2.0739e-01, 3.4174e-01, 8.2292e-01, 5.3806e-01, 1.0440e-01, 1.0218e+00, 8.3894e-01, 1.1337e+00, 2.5774e-01, 1.4880e-01, 1.0320e+00, 3.5197e-01, 6.5205e-01, 8.7084e-01, 7.9014e-01, 3.8284e-01, 8.6343e-01, 2.1449e-01, 5.3598e-01, 2.7256e-01, 4.6939e-01, 5.3806e-01, 4.1115e-01, 7.1858e-01, -1.9307e-02, 6.7990e-01, 3.4873e-01], [ 8.1844e-01, 2.2505e-01, 6.3084e-01, 1.2955e-01, 8.1577e-01, 6.8635e-01, 4.5960e-01, 1.0045e+00, 9.0669e-01, 2.1967e-01, 3.5902e-01, 6.1459e-01, 1.4342e-01, 2.1527e-01, 3.8593e-01, 6.1294e-01, -1.3796e-01, 7.2262e-01, 1.0623e+00, 1.3046e+00, 9.5788e-02, 1.0577e+00, 8.1152e-01, 9.9655e-01, 8.0728e-01, 7.2040e-01, 3.0898e-01, 7.7352e-01], [ 9.6729e-01, 1.6908e-01, 6.9169e-01, 9.2401e-01, 8.3486e-01, 5.0406e-01, 5.6620e-01, 1.4548e-01, 6.6348e-01, 4.4199e-01, 5.5479e-01, 5.5355e-01, 6.4420e-01, 2.9702e-01, 4.8829e-01, 1.5372e-01, 7.5503e-01, 9.5302e-01, 4.4603e-01, 6.8964e-02, 6.4311e-01, 9.8132e-01, 1.0859e-01, 5.1073e-01, 1.0665e+00, 9.0777e-01, 7.8715e-01, 1.0321e+00], [ 1.1102e-01, 9.1705e-01, 7.3000e-01, 4.4419e-01, 4.7482e-01, 1.4260e-01, 4.0760e-01, 8.1824e-01, 2.3078e-01, 4.5524e-01, 6.8247e-01, 5.8071e-02, 7.6399e-01, 1.0937e+00, 1.4122e-02, 1.2663e-01, 5.8911e-01, 8.9864e-01, 7.2794e-01, -2.3183e-01, 7.2209e-01, 3.4510e-01, -1.1247e-02, 1.0216e-02, 7.3514e-01, 4.7735e-01, 1.2182e-01, 8.7938e-01], [ 7.7203e-01, 6.2602e-01, 6.0416e-01, 7.4008e-01, 1.0017e+00, 4.8498e-01, 8.9410e-01, 1.0385e-01, 8.2208e-01, 8.7071e-01, 8.5781e-01, -9.2331e-02, 7.4263e-01, 7.7026e-01, 8.9110e-01, 1.0944e+00, 2.1360e-01, 7.1103e-01, 8.0248e-01, 3.9088e-01, 3.7665e-01, 6.9454e-02, 8.6131e-02, 2.4893e-02, 8.7208e-01, 1.4597e-01, 2.6949e-01, 5.5042e-01], [ 4.4113e-01, 1.0466e+00, 2.9536e-01, 9.1837e-01, 8.5094e-01, 6.4387e-01, 5.0018e-01, 5.0399e-01, 3.0896e-01, 6.5867e-01, 5.8687e-01, 3.6140e-01, 8.7345e-01, 3.2963e-01, 7.7959e-01, 7.6818e-01, 6.2958e-01, 3.8116e-01, 4.8393e-01, 1.2172e+00, 8.3255e-01, 6.4693e-02, 3.3501e-01, 3.1001e-01, 8.7589e-01, 7.5742e-01, 4.5209e-02, 4.4983e-01], [ 7.9809e-01, 8.0656e-01, 8.3592e-01, 1.7404e-01, 1.5537e-02, 8.0507e-01, 4.3430e-01, 9.7855e-02, 4.5777e-01, 7.6715e-01, 1.8115e-01, 8.5164e-01, 7.3493e-01, 5.8034e-01, 1.2705e-01, 1.0232e-01, 9.2705e-01, 8.7448e-01, 9.3301e-01, 3.9851e-01, 6.9817e-01, 3.3664e-01, 1.0443e-01, 5.4239e-01, 5.9645e-01, 3.4051e-01, 6.2836e-01, 1.2329e-01], [ 6.3783e-02, 5.4784e-01, 5.3141e-01, 7.4040e-01, 2.3979e-02, 3.5237e-01, 5.7602e-01, 4.7115e-02, 2.8785e-01, -1.6184e-01, 2.3283e-01, 6.4920e-01, 4.7756e-02, 5.4019e-01, 1.2220e-01, 1.9224e-01, 5.7837e-01, 7.4996e-01, 1.0765e-01, 4.6649e-01, 7.3734e-01, 2.5730e-01, 8.1332e-01, 2.1549e-01, 1.1076e-01, 7.0702e-01, 1.4395e-01, 7.0201e-02], [-7.4216e-03, 6.2153e-01, 5.3045e-01, 4.7079e-01, 1.7587e-01, 6.0710e-01, -2.7263e-02, 6.4655e-01, 5.3032e-01, 6.4479e-01, 5.6637e-01, 2.6414e-01, 4.0679e-01, 5.6134e-01, 6.1996e-02, 6.0691e-01, 9.7736e-02, 8.4430e-01, 4.2616e-01, 6.2229e-01, 6.8477e-01, 1.4979e-01, 9.2357e-01, 4.7910e-01, 1.6992e-01, 6.6975e-01, 8.4877e-01, 2.9303e-01], [ 9.4539e-01, 2.6781e-01, 5.2095e-01, 7.4461e-01, 4.5387e-01, 6.4790e-01, 7.8286e-01, 4.6087e-01, 1.0516e-01, 1.2914e-01, 2.3472e-01, 6.6450e-01, 3.3881e-01, 7.6108e-01, 3.3900e-01, 8.5897e-01, 8.2290e-01, 6.4250e-01, 4.6126e-01, 1.2250e-01, 8.2403e-02, 7.9182e-01, 9.1091e-02, 4.9397e-01, 3.8492e-01, 5.1067e-01, 7.9125e-01, 2.5133e-01], [ 7.9638e-01, 2.0701e-01, 8.9998e-01, 1.0123e+00, 8.9924e-01, 6.1194e-01, 2.0508e-01, 8.7714e-01, 3.3737e-01, 3.3824e-01, 5.4846e-01, 3.4923e-02, 9.5551e-01, 5.2093e-02, 1.9192e-02, 3.2956e-01, 8.9202e-01, 2.3645e-01, 6.0415e-02, 8.2699e-01, 1.0328e+00, 1.4348e-01, 3.5257e-01, 9.2653e-01, 5.1261e-01, 4.8053e-01, 4.5409e-01, 6.3957e-01], [ 2.5808e-01, 5.6338e-01, 1.5085e-01, 4.8231e-01, 9.4989e-01, 5.3319e-01, 5.3061e-01, 1.6468e-01, 3.6901e-01, 6.3093e-01, 3.7818e-02, 4.7828e-01, 2.2296e-01, 8.1360e-01, 3.9798e-01, 9.8903e-01, 7.7904e-01, 4.1130e-01, 3.3027e-01, 8.7141e-01, 8.9874e-02, 4.7944e-01, 9.5937e-01, 4.2781e-01, 1.9126e-01, 8.3606e-01, 7.1399e-02, 2.5239e-01], [ 1.2307e-01, 3.8085e-01, 4.5274e-01, 7.5406e-01, 9.0879e-01, 1.9486e-01, -3.0218e-02, 5.3887e-01, 3.9569e-01, 1.0054e+00, 1.0902e+00, 6.2841e-01, 6.7628e-01, 5.2903e-01, -6.2745e-03, 7.8162e-01, 7.7649e-01, 5.3411e-02, 8.4681e-01, 7.6299e-01, 4.0771e-01, 6.3883e-01, 5.0394e-01, 6.4061e-01, 4.4024e-01, 3.4866e-01, 4.6830e-01, 5.3808e-01], [ 1.0570e+00, 2.3428e-01, 3.2460e-01, -7.7438e-02, 8.1746e-01, 6.5163e-01, 6.8952e-01, 2.5616e-01, 2.8494e-01, 6.2445e-01, 1.0361e+00, 2.0589e-01, 4.7993e-01, 5.7002e-02, 8.7431e-01, 9.3684e-01, 7.1176e-01, 2.3189e-01, 2.5901e-01, 7.3493e-01, 3.4834e-01, 2.9198e-02, 4.1804e-01, 5.6902e-01, 5.4621e-01, 5.9270e-01, 8.5254e-01, 1.8216e-01], [ 5.0931e-01, 2.6628e-01, 8.5866e-01, 7.5221e-01, 8.4197e-01, -5.4657e-02, 2.3840e-02, 2.4812e-01, 2.1561e-01, -9.8964e-03, 3.4617e-01, 1.1136e+00, 4.6051e-01, -4.9158e-03, 1.4681e-01, 3.6429e-01, 6.5646e-01, 6.4155e-01, 2.6962e-03, 6.3122e-01, 2.7927e-01, 9.0141e-01, 1.5441e-02, 3.0391e-01, 7.3944e-01, 8.2543e-01, 9.7200e-01, 6.1424e-01], [ 1.6109e-01, 6.1200e-01, 3.3933e-01, 8.9053e-01, 1.2685e-02, 6.3852e-01, 7.7605e-01, 8.9602e-01, 6.2727e-01, 5.3226e-01, 2.6726e-01, 6.6007e-01, 1.1367e-01, 1.7282e-01, 3.0375e-01, 7.3988e-01, 3.2879e-01, 3.7334e-01, 6.7422e-01, 6.4509e-01, 6.8196e-01, 8.1858e-01, 2.3444e-01, 8.1845e-01, 6.8228e-01, 3.6365e-01, 7.3153e-01, 7.2874e-01], [ 8.2190e-01, 1.1075e-01, 3.2542e-01, 2.3088e-01, 2.3919e-01, 4.7973e-01, 8.8109e-01, 8.5061e-01, 2.3921e-01, 1.8530e-01, -4.5518e-02, 1.7482e-01, 6.2579e-01, 1.3013e-01, 2.8357e-01, 5.7316e-01, 4.1975e-01, 3.1367e-01, 7.2057e-01, 3.7517e-01, 6.7830e-01, 5.4520e-02, 8.7112e-01, 4.1475e-01, 5.3532e-01, 5.3351e-01, 1.5299e-01, 2.1696e-01], [ 3.8619e-01, 1.0056e+00, 4.5137e-01, 6.5311e-01, 2.9081e-01, 7.6705e-01, 5.2778e-01, 3.1267e-01, -1.2548e-01, 3.2731e-01, 3.7135e-01, 2.2293e-02, 2.5009e-01, 2.1048e-01, 3.4622e-01, 3.5983e-01, 3.5104e-01, 2.1220e-01, 6.7556e-01, 3.7475e-01, 1.6431e-01, 7.0422e-01, 8.0087e-01, 4.7100e-01, 9.7766e-01, 1.8277e-02, 6.1969e-01, 1.9711e-01], [ 7.8155e-01, 7.1101e-01, 2.0455e-01, 4.4153e-01, 4.8890e-01, 2.1791e-01, 6.4668e-01, 5.9280e-01, 2.9211e-01, 4.1581e-01, 3.8185e-01, 6.4024e-01, 3.1393e-01, 5.0632e-01, 8.2283e-01, 1.8128e-01, 4.4095e-01, 2.9088e-01, 3.6859e-01, 7.1484e-01, 8.6203e-01, 7.5778e-02, 3.8975e-01, 3.4041e-01, 5.6930e-01, 2.7090e-01, 3.4177e-01, 5.4593e-01], [ 6.9933e-01, 7.6190e-01, 9.1418e-01, 5.1421e-01, 1.0412e+00, 2.7945e-01, -1.5315e-01, 3.3666e-02, 2.4592e-02, -4.8662e-02, 4.8521e-01, 6.3433e-02, 5.8517e-01, 8.8535e-02, 5.8993e-01, 4.9659e-01, 6.3918e-01, 4.2742e-01, 3.5838e-01, 9.8733e-01, 2.2435e-01, 4.5459e-01, 1.0969e+00, 3.6596e-01, 2.4139e-01, 2.5908e-01, 6.2124e-01, 6.3509e-01], [ 3.7773e-01, 1.5590e-01, 4.6437e-01, 1.0532e+00, 7.7038e-01, 7.7253e-01, 5.5581e-01, 8.5207e-01, 1.9952e-01, 4.3578e-04, 1.8705e-01, 7.2782e-01, 2.8181e-01, 7.5687e-01, 3.1991e-01, 3.8863e-01, 1.0320e+00, 8.8327e-01, 1.5132e-01, 2.2798e-01, 7.8556e-01, 7.4501e-02, 5.0239e-01, 5.6149e-01, 5.7039e-01, 5.4560e-01, 2.5559e-01, 9.1737e-01], [ 2.4790e-01, 9.1259e-02, 4.8480e-01, 8.7040e-01, 5.0994e-01, 1.2352e+00, 4.3721e-01, 1.1540e+00, 5.7661e-01, 4.4002e-01, 4.3529e-01, 3.4350e-01, 3.6314e-01, 3.2342e-02, 2.7325e-01, 8.1995e-01, 5.7767e-01, 1.1382e+00, 5.6145e-01, 9.2122e-01, 1.0223e+00, 8.3465e-01, 3.4931e-01, 6.5662e-02, 9.8764e-02, 4.7991e-01, 6.3983e-01, 6.0949e-01], [ 5.7333e-01, 4.3587e-01, 1.1823e-01, 9.6195e-01, 7.3666e-01, 4.4254e-01, 3.1766e-01, 1.0803e+00, 4.1708e-01, 9.6695e-01, 2.6936e-01, 8.2981e-02, 2.9549e-01, 4.1854e-01, 7.3636e-01, 4.8149e-01, 4.6612e-01, 8.4265e-01, 1.7358e-01, 1.0392e+00, 7.4474e-01, 2.9865e-01, 9.7939e-01, 7.7768e-01, 4.0470e-01, 5.2455e-01, 5.9163e-01, 1.4044e-01]]]], device='cuda:0', requires_grad=True) # tensored = torch.rand(1, 1, 28, 28).to(device) # tensored=torch.clamp(tensored, min=0,max=1) # img_save=transforms.ToPILImage()(tensored[0]) # img_save.save('out.png') while True: tensored.requires_grad_() embedding = model(tensored)[0] dist = torch.sqrt(torch.sum((feature - embedding)**2)) if dist <= 6: tensor1 = tensored * 255 array = tensor1.detach().cpu().numpy() array = array.astype(np.uint8) array1=torch.tensor(array).to("cuda:0") img_new_tensor = torch.tensor(array1.reshape(28, 28), dtype=torch.float32) / 255.0 img_new_tensor = img_new_tensor.unsqueeze(0) feature_new = model(img_new_tensor.to(device)).detach().cpu().numpy() ddd = np.sqrt(np.sum((feature.cpu().numpy() - feature_new)**2)) if ddd<6: print(array) exit(0) dist.backward() tensored = tensored.detach() - tensored.grad**3 *0.0000001 tensored=torch.clamp(tensored, min=0,max=1) print(dist) crack()
s1ayth3sp1re
反编译搜索3000找到
两坨数组拿出来异或即可。
arr1 = [164, 158, 95, 107, 4, 215, 108, 115, 5, 8, 25, 57, 41, 236, 231, 17, 85] arr2 = [246,221,11,45,127,148,45,36,70,73,78,8,98,141,140,112,40] for i in range(len(arr2)): print(chr(arr1[i] ^ arr2[i]),end="")