{"cells":[{"cell_type":"markdown","metadata":{},"source":["# NAME SURNAME"]},{"cell_type":"markdown","metadata":{},"source":["## Packages"]},{"cell_type":"code","execution_count":1,"metadata":{"_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","execution":{"iopub.execute_input":"2025-01-07T10:33:46.313989Z","iopub.status.busy":"2025-01-07T10:33:46.313500Z","iopub.status.idle":"2025-01-07T10:33:47.366475Z","shell.execute_reply":"2025-01-07T10:33:47.365189Z","shell.execute_reply.started":"2025-01-07T10:33:46.313904Z"},"trusted":true},"outputs":[],"source":["import networkx as nx # For graphs\n","import pickle # For data parsing\n","import matplotlib.pyplot as plt\n","from networkx.algorithms.approximation import greedy_tsp # For approx TSP\n","import numpy as np\n","import torch\n","import torch.nn as nn\n","import torch.nn.functional as F\n","from tqdm.cli import tqdm\n","import pandas as pd\n","import random\n","\n","torch.set_float32_matmul_precision('medium')\n","torch.use_deterministic_algorithms(False)"]},{"cell_type":"code","execution_count":2,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["Mon Jan 20 20:57:32 2025       \n","+-----------------------------------------------------------------------------------------+\n","| NVIDIA-SMI 560.35.03              Driver Version: 560.35.03      CUDA Version: 12.6     |\n","|-----------------------------------------+------------------------+----------------------+\n","| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |\n","| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |\n","|                                         |                        |               MIG M. |\n","|=========================================+========================+======================|\n","|   0  NVIDIA GeForce RTX 4090        On  |   00000000:23:00.0 Off |                  Off |\n","|  0%   41C    P3             48W /  450W |       1MiB /  24564MiB |      0%      Default |\n","|                                         |                        |                  N/A |\n","+-----------------------------------------+------------------------+----------------------+\n","                                                                                         \n","+-----------------------------------------------------------------------------------------+\n","| Processes:                                                                              |\n","|  GPU   GI   CI        PID   Type   Process name                              GPU Memory |\n","|        ID   ID                                                               Usage      |\n","|=========================================================================================|\n","|  No running processes found                                                             |\n","+-----------------------------------------------------------------------------------------+\n"]}],"source":["!nvidia-smi"]},{"cell_type":"markdown","metadata":{},"source":["## Helper functions"]},{"cell_type":"code","execution_count":3,"metadata":{"trusted":true},"outputs":[],"source":["\n","def tour_length(G, tour):\n","    \"\"\"\n","    Compute the length of a tour. A tour is a list having elments 0 and -1 equal\n","    \"\"\"\n","    n = len(tour) - 1\n","    assert tour[0] == tour[-1], \"Not valid tour\"\n","    estimated = 0\n","    for i in range(n):\n","        estimated += G[tour[i]][tour[i + 1]]['weight']\n","    return estimated\n","\n","def greedy_algorithm(G):\n","    \"\"\"\n","    Run the value of the greedy approximation algorithm on graph G\n","    \"\"\"\n","    return tour_length(G, greedy_tsp(G, weight='weight'))\n","\n","def random_tour(G, seed = 42):\n","    \"\"\"\n","    Return the value of a random tour\n","    \"\"\"\n","    np.random.seed(seed)\n","    n = G.number_of_nodes()\n","    tour = [0]\n","    for i in range(1, n):\n","        next_node = np.random.choice([j for j in range(n) if j not in tour])\n","        tour.append(next_node)\n","    tour.append(0)\n","    return tour_length(G, tour)\n","    \n","@torch.no_grad\n","def transformer_tsp(G, model, DEVICE = 'cpu'):\n","    \"\"\"\n","    Evaluate your (trained) model on G\n","    TODO: If you used some masks, add them when needed. \n","    \"\"\"\n","    # Set the model in evaluation mode\n","    model.eval()\n","\n","    # Note: number of edges is constant ed equal to n(n-1)/2\n","    n = G.number_of_nodes()\n","    E = G.number_of_edges()\n","\n","    \n","    # Get node coordinates\n","    attr = nx.get_node_attributes(G, 'pos')\n","    x = []\n","    for i in range(n):\n","        x.append(torch.tensor(attr[i], dtype=torch.float32))\n","\n","    # From list of tensors to tensor\n","    x = torch.stack(x)    \n","\n","    tour = [0]\n","    y = torch.tensor(tour, dtype=torch.long)\n","    x = x.to(DEVICE).unsqueeze(0)\n","    y = y.to(DEVICE).unsqueeze(0)\n","    \n","    out = model(x, y)\n","    \n","    while len(tour) < n:\n","        _, idx = torch.topk(out, n, dim=2)\n","        for i in range(n):\n","            if idx[0, -1, i] not in tour:\n","                tour.append(idx[0, -1, i])\n","                break\n","        y = torch.tensor(tour)\n","        y = y.to(DEVICE).unsqueeze(0)\n","        out = model(x, y)\n","    \n","    tour = [int(i) for i in tour] + [0]\n","    return tour_length(G, tour)\n","\n","\n","@torch.no_grad\n","def gap(G, model = None, model_GA = None, random_seed = 42, device = 'cpu'):\n","    \"\"\"\n","    Compute the gap between the optimal solution on graph G and all the analyzed methods\n","    \"\"\"\n","        \n","    # Optimal value (hard-coded in the graph)\n","    TSP = sum([G[i][j]['weight']*G[i][j]['tour'] for (i, j) in G.edges()]) # Optimal\n","\n","    # Gaps dictionary\n","    gaps = {'greedy' : 0, 'random' : 0, 'transformer_tsp': 0, 'transformer_tsp_acc_grad': 0}\n","    gaps['greedy'] = 100* (greedy_algorithm(G) -  TSP) / TSP\n","    gaps['random'] = 100 * (random_tour(G, random_seed) - TSP) / TSP\n","    if model is not None:\n","        gaps['transformer_tsp'] = 100 * (transformer_tsp(G, model, DEVICE=device) - TSP) / TSP\n","    else:\n","        gaps['transformer_tsp'] = float('inf')\n","        \n","    if model_GA is not None:\n","        gaps['transformer_tsp_acc_grad'] = 100 * (transformer_tsp(G, model_GA, DEVICE=device) - TSP) / TSP\n","    else:\n","        gaps['transformer_tsp_acc_grad'] = float('inf')\n","    return gaps    \n","    "]},{"cell_type":"markdown","metadata":{},"source":["## Dataset & Dataloader"]},{"cell_type":"markdown","metadata":{},"source":["(1 pt) Load the dummy dataset, get a single data item and explain its Python type."]},{"cell_type":"code","execution_count":4,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["dummy_20_DLL_ass4.pkl  train_20_DLL_ass4.pkl\n","test_20_DLL_ass4.pkl   valid_20_DLL_ass4.pkl\n"]}],"source":["!ls data\n","# Load the dummy dataset, get a single data item and explain its Python type.\n","dummy = pickle.loads(open('data/dummy_20_DLL_ass4.pkl', 'rb').read())"]},{"cell_type":"markdown","metadata":{},"source":["(5 pts) Describe the edge attributes tour and weight, as well as the node attribute pos."]},{"cell_type":"code","execution_count":5,"metadata":{},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 640x480 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":["# See the actual pat\n","G, path = dummy[0] \n","plt.figure()\n","pos = nx.spring_layout(G)\n","nx.draw(G, pos, with_labels=True)\n","path_edges = list(zip(path, path[1:]))\n","nx.draw_networkx_edges(G, pos, edgelist=path_edges, edge_color='r', width=2)\n","plt.show()"]},{"cell_type":"code","execution_count":6,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["path [0, 3, 14, 2, 9, 6, 19, 13, 12, 16, 7, 18, 8, 17, 5, 11, 10, 15, 1, 4, 0]\n","edge tour attributes {(0, 1): 0, (0, 2): 0, (0, 3): 1, (0, 4): 1, (0, 5): 0, (0, 6): 0, (0, 7): 0, (0, 8): 0, (0, 9): 0, (0, 10): 0, (0, 11): 0, (0, 12): 0, (0, 13): 0, (0, 14): 0, (0, 15): 0, (0, 16): 0, (0, 17): 0, (0, 18): 0, (0, 19): 0, (1, 2): 0, (1, 3): 0, (1, 4): 1, (1, 5): 0, (1, 6): 0, (1, 7): 0, (1, 8): 0, (1, 9): 0, (1, 10): 0, (1, 11): 0, (1, 12): 0, (1, 13): 0, (1, 14): 0, (1, 15): 1, (1, 16): 0, (1, 17): 0, (1, 18): 0, (1, 19): 0, (2, 3): 0, (2, 4): 0, (2, 5): 0, (2, 6): 0, (2, 7): 0, (2, 8): 0, (2, 9): 1, (2, 10): 0, (2, 11): 0, (2, 12): 0, (2, 13): 0, (2, 14): 1, (2, 15): 0, (2, 16): 0, (2, 17): 0, (2, 18): 0, (2, 19): 0, (3, 4): 0, (3, 5): 0, (3, 6): 0, (3, 7): 0, (3, 8): 0, (3, 9): 0, (3, 10): 0, (3, 11): 0, (3, 12): 0, (3, 13): 0, (3, 14): 1, (3, 15): 0, (3, 16): 0, (3, 17): 0, (3, 18): 0, (3, 19): 0, (4, 5): 0, (4, 6): 0, (4, 7): 0, (4, 8): 0, (4, 9): 0, (4, 10): 0, (4, 11): 0, (4, 12): 0, (4, 13): 0, (4, 14): 0, (4, 15): 0, (4, 16): 0, (4, 17): 0, (4, 18): 0, (4, 19): 0, (5, 6): 0, (5, 7): 0, (5, 8): 0, (5, 9): 0, (5, 10): 0, (5, 11): 1, (5, 12): 0, (5, 13): 0, (5, 14): 0, (5, 15): 0, (5, 16): 0, (5, 17): 1, (5, 18): 0, (5, 19): 0, (6, 7): 0, (6, 8): 0, (6, 9): 1, (6, 10): 0, (6, 11): 0, (6, 12): 0, (6, 13): 0, (6, 14): 0, (6, 15): 0, (6, 16): 0, (6, 17): 0, (6, 18): 0, (6, 19): 1, (7, 8): 0, (7, 9): 0, (7, 10): 0, (7, 11): 0, (7, 12): 0, (7, 13): 0, (7, 14): 0, (7, 15): 0, (7, 16): 1, (7, 17): 0, (7, 18): 1, (7, 19): 0, (8, 9): 0, (8, 10): 0, (8, 11): 0, (8, 12): 0, (8, 13): 0, (8, 14): 0, (8, 15): 0, (8, 16): 0, (8, 17): 1, (8, 18): 1, (8, 19): 0, (9, 10): 0, (9, 11): 0, (9, 12): 0, (9, 13): 0, (9, 14): 0, (9, 15): 0, (9, 16): 0, (9, 17): 0, (9, 18): 0, (9, 19): 0, (10, 11): 1, (10, 12): 0, (10, 13): 0, (10, 14): 0, (10, 15): 1, (10, 16): 0, (10, 17): 0, (10, 18): 0, (10, 19): 0, (11, 12): 0, (11, 13): 0, (11, 14): 0, (11, 15): 0, (11, 16): 0, (11, 17): 0, (11, 18): 0, (11, 19): 0, (12, 13): 1, (12, 14): 0, (12, 15): 0, (12, 16): 1, (12, 17): 0, (12, 18): 0, (12, 19): 0, (13, 14): 0, (13, 15): 0, (13, 16): 0, (13, 17): 0, (13, 18): 0, (13, 19): 1, (14, 15): 0, (14, 16): 0, (14, 17): 0, (14, 18): 0, (14, 19): 0, (15, 16): 0, (15, 17): 0, (15, 18): 0, (15, 19): 0, (16, 17): 0, (16, 18): 0, (16, 19): 0, (17, 18): 0, (17, 19): 0, (18, 19): 0}\n","edge weight attributes {(0, 1): np.float64(0.4287846201876535), (0, 2): np.float64(0.20417626260418414), (0, 3): np.float64(0.08154537102129383), (0, 4): np.float64(0.08031174728403137), (0, 5): np.float64(0.5612080164024046), (0, 6): np.float64(0.2295012043541082), (0, 7): np.float64(0.514746498076055), (0, 8): np.float64(0.5266704818162102), (0, 9): np.float64(0.15283720045858457), (0, 10): np.float64(0.3556451198878696), (0, 11): np.float64(0.24046942531001395), (0, 12): np.float64(0.3189080728571583), (0, 13): np.float64(0.49621109228471955), (0, 14): np.float64(0.1765729477183786), (0, 15): np.float64(0.4701608359103749), (0, 16): np.float64(0.3309416339870591), (0, 17): np.float64(0.5383964713091507), (0, 18): np.float64(0.4857390315454929), (0, 19): np.float64(0.2955332394593224), (1, 2): np.float64(0.6148965694180403), (1, 3): np.float64(0.458726401538421), (1, 4): np.float64(0.35637320892343344), (1, 5): np.float64(0.6032864976885546), (1, 6): np.float64(0.6569915301709606), (1, 7): np.float64(0.9207746398930378), (1, 8): np.float64(0.746742011300412), (1, 9): np.float64(0.5810402333348532), (1, 10): np.float64(0.16704392462309525), (1, 11): np.float64(0.44648016447297073), (1, 12): np.float64(0.7422384454019971), (1, 13): np.float64(0.9087485603102635), (1, 14): np.float64(0.5479091406494869), (1, 15): np.float64(0.14178198392466582), (1, 16): np.float64(0.754354006043937), (1, 17): np.float64(0.6438862448532572), (1, 18): np.float64(0.7726546814458304), (1, 19): np.float64(0.5987646096934333), (2, 3): np.float64(0.15622130264510733), (2, 4): np.float64(0.2841510683911888), (2, 5): np.float64(0.5519190450546716), (2, 6): np.float64(0.13946205995874017), (2, 7): np.float64(0.3114220898670106), (2, 8): np.float64(0.4194690047798088), (2, 9): np.float64(0.08856321635783988), (2, 10): np.float64(0.5083743055689474), (2, 11): np.float64(0.2644009954540205), (2, 12): np.float64(0.2208301469437315), (2, 13): np.float64(0.4003621671541578), (2, 14): np.float64(0.09093880543328159), (2, 15): np.float64(0.6278926750270484), (2, 16): np.float64(0.14517438585231934), (2, 17): np.float64(0.49907142344556654), (2, 18): np.float64(0.34060003464189553), (2, 19): np.float64(0.3852352221449024), (3, 4): np.float64(0.14654472382192782), (3, 5): np.float64(0.4946628652419162), (3, 6): np.float64(0.2312084560472005), (3, 7): np.float64(0.46429696895897604), (3, 8): np.float64(0.4453880485781254), (3, 9): np.float64(0.14324743187115518), (3, 10): np.float64(0.357006300672558), (3, 11): np.float64(0.17313884874554414), (3, 12): np.float64(0.3256331892473202), (3, 13): np.float64(0.5098507108265369), (3, 14): np.float64(0.10192800337320708), (3, 15): np.float64(0.4760316956166869), (3, 16): np.float64(0.2982239590456468), (3, 17): np.float64(0.4650265877604502), (3, 18): np.float64(0.40529952122582097), (3, 19): np.float64(0.3641842557466924), (4, 5): np.float64(0.5764321626357953), (4, 6): np.float64(0.30133829128093664), (4, 7): np.float64(0.5950103618814667), (4, 8): np.float64(0.5770449832745134), (4, 9): np.float64(0.2307340037832632), (4, 10): np.float64(0.3046349275334078), (4, 11): np.float64(0.26810007977040523), (4, 12): np.float64(0.3859450475320259), (4, 13): np.float64(0.5561153892125273), (4, 14): np.float64(0.24804828262824546), (4, 15): np.float64(0.41247528818868645), (4, 16): np.float64(0.410375847856154), (4, 17): np.float64(0.5656186289094377), (4, 18): np.float64(0.5483723633197157), (4, 19): np.float64(0.30210799803116456), (5, 6): np.float64(0.6867560124051679), (5, 7): np.float64(0.6992915692243714), (5, 8): np.float64(0.27073049472112015), (5, 9): np.float64(0.6078062084716456), (5, 10): np.float64(0.43958977527759663), (5, 11): np.float64(0.32199052710954257), (5, 12): np.float64(0.7727326567888985), (5, 13): np.float64(0.9501826066747174), (5, 14): np.float64(0.46911196324102344), (5, 15): np.float64(0.4935936115373682), (5, 16): np.float64(0.6639435819391838), (5, 17): np.float64(0.08842284642269631), (5, 18): np.float64(0.3771405718250126), (5, 19): np.float64(0.8562970048150594), (6, 7): np.float64(0.3452299871310383), (6, 8): np.float64(0.5566916991250165), (6, 9): np.float64(0.08818713565270636), (6, 10): np.float64(0.5806051325658653), (6, 11): np.float64(0.38358592867555963), (6, 12): np.float64(0.09446322848406911), (6, 13): np.float64(0.27914978407080576), (6, 14): np.float64(0.2178340373451969), (6, 15): np.float64(0.6973941311815144), (6, 16): np.float64(0.1437315155664361), (6, 17): np.float64(0.6372490513538306), (6, 18): np.float64(0.4701840720104889), (6, 19): np.float64(0.2774232612572184), (7, 8): np.float64(0.45952904895181307), (7, 9): np.float64(0.37417400645996296), (7, 10): np.float64(0.7992567869828283), (7, 11): np.float64(0.5148673440983987), (7, 12): np.float64(0.33571626255030856), (7, 13): np.float64(0.4026101124667492), (7, 14): np.float64(0.37340467136695177), (7, 15): np.float64(0.9174698538239648), (7, 16): np.float64(0.20177465871599579), (7, 17): np.float64(0.6195648253244966), (7, 18): np.float64(0.33917612945209147), (7, 19): np.float64(0.6162194325540812), (8, 9): np.float64(0.500606236365931), (8, 10): np.float64(0.5829012393087726), (8, 11): np.float64(0.3226682580193532), (8, 12): np.float64(0.6236465117971378), (8, 13): np.float64(0.7794753363995585), (8, 14): np.float64(0.3687755686215689), (8, 15): np.float64(0.6744208504349039), (8, 16): np.float64(0.48341874284217323), (8, 17): np.float64(0.1823906358666875), (8, 18): np.float64(0.12039747445267121), (8, 19): np.float64(0.7920456502968736), (9, 10): np.float64(0.4958159010103946), (9, 11): np.float64(0.2981604547970567), (9, 12): np.float64(0.18263885255286078), (9, 13): np.float64(0.3672998819800728), (9, 14): np.float64(0.14074192147940914), (9, 15): np.float64(0.6136048694347368), (9, 16): np.float64(0.1807984883622037), (9, 17): np.float64(0.5635976733439557), (9, 18): np.float64(0.4275922503419513), (9, 19): np.float64(0.2972159394392491), (10, 11): np.float64(0.29767923348870867), (10, 12): np.float64(0.6728443418750315), (10, 13): np.float64(0.8518446381557372), (10, 14): np.float64(0.42882230173921154), (10, 15): np.float64(0.11952991870231995), (10, 16): np.float64(0.6531686692237917), (10, 17): np.float64(0.4770670662090856), (10, 18): np.float64(0.6168640222169508), (10, 19): np.float64(0.5961301720109755), (11, 12): np.float64(0.47581620562438115), (11, 13): np.float64(0.6599029483947126), (11, 14): np.float64(0.17346768130254947), (11, 15): np.float64(0.41048539737715684), (11, 16): np.float64(0.4025683785940153), (11, 17): np.float64(0.29925677077431195), (11, 18): np.float64(0.3274947351745373), (11, 19): np.float64(0.5346097241040995), (12, 13): np.float64(0.18490797189553923), (12, 14): np.float64(0.3065099771993463), (12, 15): np.float64(0.7887404243083697), (12, 16): np.float64(0.15629069658405714), (12, 17): np.float64(0.7187869230155136), (12, 18): np.float64(0.5268902682284917), (12, 19): np.float64(0.28846075399036875), (13, 14): np.float64(0.48887714561053647), (13, 15): np.float64(0.9657246855588165), (13, 16): np.float64(0.2962429306808471), (13, 17): np.float64(0.8913669046263939), (13, 18): np.float64(0.6721854655144677), (13, 19): np.float64(0.3719877587298442), (14, 15): np.float64(0.5480609545208327), (14, 16): np.float64(0.2321206572622659), (14, 17): np.float64(0.42288927202931514), (14, 18): np.float64(0.3126013860373729), (14, 19): np.float64(0.4234389913230374), (15, 16): np.float64(0.7726413301352619), (15, 17): np.float64(0.5469248197512611), (15, 18): np.float64(0.7204237265004912), (15, 19): np.float64(0.6911317266141075), (16, 17): np.float64(0.599767667597168), (16, 18): np.float64(0.3787620993736398), (16, 19): np.float64(0.41838423486864734), (17, 18): np.float64(0.2912935061526579), (17, 19): np.float64(0.8291779696877807), (18, 19): np.float64(0.7246549985661137)}\n","node pos attributes {0: (0.6049077053425551, 0.5748590937018008), 1: (0.38474987528197846, 0.9428085200806016), 2: (0.6102491981278754, 0.3707527129445174), 3: (0.5497610140601452, 0.514788385568776), 4: (0.5941533303116413, 0.6544475361385552), 5: (0.06187381797691738, 0.433195284467101), 6: (0.7475717305758963, 0.3950876312718402), 7: (0.6548530739834322, 0.06254140180263457), 8: (0.2210796367473482, 0.2142238067774731), 9: (0.6696714621150585, 0.4364218673039507), 10: (0.3206284506117195, 0.7885615893113229), 11: (0.37760295594024584, 0.4963855605324464), 12: (0.8300002368321365, 0.3489482457969727), 13: (0.9983722301898076, 0.2725163812162502), 14: (0.5306200040273271, 0.4146737532387711), 15: (0.2540580284004623, 0.8878378722372976), 16: (0.7024553839343133, 0.2586205516051998), 17: (0.11129635562701035, 0.35987398782201374), 18: (0.33376928628552816, 0.1718375502602505), 19: (0.8952647398936215, 0.6299289412776933)}\n"]}],"source":["print('path', path)\n","print(f\"edge tour attributes {nx.get_edge_attributes(G, 'tour')}\")\n","print(f\"edge weight attributes {nx.get_edge_attributes(G, 'weight')}\")\n","print(f\"node pos attributes {nx.get_node_attributes(G, 'pos')}\")"]},{"cell_type":"markdown","metadata":{},"source":["(10 pts) Implement a dataset class. Focus on the getitem method to\n","return:\n","- X: A tensor of node coordinates with size 20 × 2.\n","- y: A tour starting from 0 and ending with 0."]},{"cell_type":"code","execution_count":7,"metadata":{},"outputs":[{"data":{"text/plain":["[(<networkx.classes.graph.Graph at 0x7f565e68d450>,\n","  [0, 3, 14, 2, 9, 6, 19, 13, 12, 16, 7, 18, 8, 17, 5, 11, 10, 15, 1, 4, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f565bbabcd0>,\n","  [0, 13, 4, 14, 18, 9, 16, 5, 19, 7, 6, 8, 2, 10, 12, 17, 11, 1, 15, 3, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67507890>,\n","  [0, 3, 19, 4, 10, 14, 1, 2, 6, 18, 13, 16, 12, 9, 5, 17, 15, 11, 7, 8, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6751c2d0>,\n","  [0, 3, 10, 16, 19, 11, 15, 12, 7, 14, 4, 17, 2, 1, 18, 9, 13, 5, 6, 8, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6751fdd0>,\n","  [0, 17, 19, 4, 16, 18, 3, 10, 7, 11, 12, 1, 8, 13, 6, 9, 14, 2, 5, 15, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67533910>,\n","  [0, 1, 4, 3, 13, 12, 6, 17, 5, 7, 11, 15, 19, 18, 14, 2, 10, 16, 9, 8, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6753f5d0>,\n","  [0, 1, 2, 11, 9, 12, 5, 3, 7, 19, 13, 15, 4, 8, 10, 14, 17, 18, 16, 6, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6754f110>,\n","  [0, 2, 14, 1, 6, 12, 11, 19, 5, 3, 17, 13, 8, 18, 16, 7, 9, 15, 4, 10, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67561e10>,\n","  [0, 14, 16, 2, 5, 19, 18, 11, 3, 12, 9, 8, 10, 13, 7, 15, 1, 17, 4, 6, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67578fd0>,\n","  [0, 17, 13, 10, 12, 14, 5, 7, 4, 8, 9, 3, 6, 2, 19, 15, 11, 1, 16, 18, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6758c190>,\n","  [0, 15, 7, 10, 17, 8, 14, 13, 3, 12, 18, 1, 19, 6, 4, 9, 2, 11, 16, 5, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6759b190>,\n","  [0, 1, 5, 12, 19, 9, 8, 14, 16, 13, 4, 2, 17, 11, 15, 6, 18, 7, 3, 10, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b675ae350>,\n","  [0, 6, 12, 11, 9, 16, 17, 7, 5, 13, 1, 4, 15, 8, 18, 19, 2, 3, 10, 14, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b675c13d0>,\n","  [0, 10, 16, 17, 14, 8, 15, 3, 18, 6, 13, 19, 9, 5, 2, 7, 12, 4, 1, 11, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b675d4410>,\n","  [0, 2, 14, 12, 1, 4, 18, 7, 11, 17, 13, 8, 15, 9, 5, 10, 6, 16, 19, 3, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b675db490>,\n","  [0, 13, 16, 15, 8, 3, 9, 5, 14, 17, 2, 12, 10, 1, 7, 6, 11, 4, 18, 19, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b675f2510>,\n","  [0, 2, 1, 18, 4, 19, 11, 10, 9, 7, 8, 16, 15, 14, 5, 13, 6, 12, 3, 17, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67405550>,\n","  [0, 13, 9, 12, 14, 7, 5, 2, 10, 11, 4, 17, 19, 6, 18, 15, 16, 8, 1, 3, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6741c610>,\n","  [0, 15, 19, 7, 14, 16, 3, 6, 11, 2, 10, 18, 12, 9, 8, 17, 13, 1, 5, 4, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67427610>,\n","  [0, 18, 17, 12, 6, 16, 10, 1, 5, 11, 9, 7, 19, 8, 14, 13, 4, 15, 3, 2, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6743e6d0>,\n","  [0, 17, 12, 11, 7, 15, 19, 5, 8, 13, 18, 14, 2, 6, 10, 16, 9, 1, 4, 3, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67451750>,\n","  [0, 17, 10, 14, 4, 19, 8, 5, 6, 1, 16, 18, 3, 12, 11, 2, 7, 15, 13, 9, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67464790>,\n","  [0, 7, 2, 19, 8, 10, 5, 17, 13, 16, 6, 18, 12, 14, 15, 9, 4, 1, 11, 3, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67473810>,\n","  [0, 11, 19, 16, 12, 17, 3, 1, 4, 7, 18, 9, 14, 10, 13, 5, 2, 6, 8, 15, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b674828d0>,\n","  [0, 6, 5, 2, 11, 17, 7, 15, 3, 4, 14, 1, 16, 18, 13, 12, 10, 8, 19, 9, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67499910>,\n","  [0, 18, 9, 11, 14, 5, 16, 4, 7, 2, 6, 10, 19, 3, 8, 17, 15, 1, 13, 12, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b674a89d0>,\n","  [0, 8, 6, 4, 7, 17, 1, 11, 5, 9, 10, 19, 14, 15, 2, 13, 3, 16, 12, 18, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b674b7a10>,\n","  [0, 19, 10, 11, 2, 15, 12, 8, 4, 17, 7, 16, 6, 3, 14, 1, 9, 5, 13, 18, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b674caa50>,\n","  [0, 6, 13, 14, 9, 8, 5, 7, 19, 1, 2, 12, 17, 18, 15, 3, 16, 11, 4, 10, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b674ddb10>,\n","  [0, 17, 9, 5, 10, 18, 1, 11, 4, 15, 14, 13, 6, 8, 2, 19, 7, 16, 12, 3, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b674f0b50>,\n","  [0, 8, 11, 7, 9, 13, 5, 18, 2, 6, 1, 15, 14, 12, 3, 10, 19, 17, 16, 4, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67303bd0>,\n","  [0, 9, 18, 19, 13, 1, 8, 14, 4, 10, 7, 2, 6, 17, 11, 3, 15, 5, 16, 12, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67316c50>,\n","  [0, 10, 8, 2, 4, 19, 6, 14, 16, 7, 5, 11, 1, 12, 13, 17, 15, 9, 18, 3, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67329c90>,\n","  [0, 15, 17, 5, 18, 16, 19, 6, 14, 12, 10, 11, 1, 8, 3, 7, 4, 13, 9, 2, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6733cd50>,\n","  [0, 5, 18, 3, 17, 13, 9, 4, 19, 1, 10, 16, 15, 6, 12, 8, 11, 7, 2, 14, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67347d90>,\n","  [0, 6, 15, 12, 14, 19, 10, 18, 13, 11, 2, 5, 3, 16, 1, 8, 7, 4, 17, 9, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6735add0>,\n","  [0, 3, 16, 14, 18, 13, 19, 2, 7, 1, 5, 4, 6, 15, 10, 8, 11, 17, 9, 12, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6736de90>,\n","  [0, 10, 4, 13, 15, 9, 18, 5, 8, 2, 17, 16, 19, 14, 12, 3, 7, 6, 11, 1, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67380f10>,\n","  [0, 15, 16, 1, 5, 10, 18, 17, 7, 3, 6, 19, 4, 12, 14, 8, 2, 9, 13, 11, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6738ff10>,\n","  [0, 16, 19, 3, 17, 11, 6, 5, 13, 18, 12, 4, 7, 1, 9, 8, 14, 15, 2, 10, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b673a2fd0>,\n","  [0, 5, 7, 17, 3, 18, 8, 4, 15, 14, 19, 1, 12, 10, 2, 9, 16, 6, 13, 11, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b673b6010>,\n","  [0, 1, 14, 4, 8, 18, 13, 19, 10, 9, 11, 7, 2, 17, 16, 15, 3, 5, 12, 6, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b673cd0d0>,\n","  [0, 1, 19, 8, 17, 4, 13, 2, 6, 10, 3, 11, 18, 12, 5, 7, 16, 14, 9, 15, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b673dc150>,\n","  [0, 4, 2, 16, 19, 15, 13, 3, 7, 12, 9, 18, 1, 8, 17, 11, 5, 6, 10, 14, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b673e7150>,\n","  [0, 12, 3, 9, 16, 10, 18, 6, 11, 1, 4, 19, 2, 8, 14, 5, 15, 13, 7, 17, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67202210>,\n","  [0, 9, 19, 8, 3, 5, 13, 17, 14, 6, 2, 11, 18, 12, 10, 15, 1, 7, 4, 16, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67215290>,\n","  [0, 6, 5, 1, 15, 12, 10, 11, 9, 8, 14, 7, 3, 17, 13, 18, 19, 16, 2, 4, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b672242d0>,\n","  [0, 19, 13, 9, 4, 15, 18, 7, 17, 6, 1, 16, 5, 11, 8, 3, 12, 14, 10, 2, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67233350>,\n","  [0, 10, 9, 6, 14, 4, 15, 17, 11, 5, 13, 19, 18, 2, 1, 12, 7, 8, 16, 3, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b672463d0>,\n","  [0, 3, 5, 13, 16, 19, 18, 17, 2, 4, 7, 14, 9, 15, 8, 1, 10, 12, 6, 11, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67259410>,\n","  [0, 11, 10, 14, 2, 16, 17, 7, 12, 5, 15, 4, 8, 9, 1, 19, 6, 3, 13, 18, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b672704d0>,\n","  [0, 19, 2, 11, 8, 17, 12, 5, 3, 10, 6, 15, 14, 13, 4, 9, 18, 1, 7, 16, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6727b4d0>,\n","  [0, 9, 5, 7, 6, 19, 18, 2, 11, 12, 4, 10, 3, 17, 15, 16, 14, 1, 13, 8, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67292590>,\n","  [0, 14, 4, 18, 9, 5, 11, 13, 8, 19, 3, 7, 1, 17, 12, 2, 10, 6, 16, 15, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b672a1610>,\n","  [0, 12, 5, 4, 17, 7, 9, 18, 16, 11, 1, 10, 6, 19, 14, 2, 3, 13, 15, 8, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b672b8650>,\n","  [0, 15, 12, 18, 10, 1, 17, 11, 6, 9, 3, 14, 16, 8, 4, 7, 19, 13, 2, 5, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b672c36d0>,\n","  [0, 18, 14, 16, 17, 19, 6, 11, 15, 10, 4, 13, 9, 2, 12, 1, 5, 3, 7, 8, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b672d2750>,\n","  [0, 12, 3, 2, 9, 18, 13, 1, 17, 15, 10, 6, 5, 16, 11, 4, 8, 14, 19, 7, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b672e5790>,\n","  [0, 8, 2, 14, 17, 9, 18, 3, 15, 12, 6, 16, 13, 19, 5, 7, 4, 10, 11, 1, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b672f8850>,\n","  [0, 17, 13, 19, 2, 5, 12, 4, 11, 3, 16, 10, 8, 9, 1, 6, 18, 15, 14, 7, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6710b890>,\n","  [0, 18, 11, 2, 16, 19, 7, 9, 8, 13, 17, 6, 15, 12, 14, 5, 1, 3, 10, 4, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6711e8d0>,\n","  [0, 17, 14, 16, 8, 11, 3, 10, 9, 12, 13, 19, 18, 4, 15, 5, 7, 6, 1, 2, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67135990>,\n","  [0, 2, 1, 19, 4, 14, 9, 3, 10, 7, 6, 18, 13, 17, 11, 15, 5, 16, 12, 8, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b671489d0>,\n","  [0, 17, 10, 11, 2, 4, 9, 15, 1, 19, 5, 13, 16, 8, 7, 6, 18, 14, 3, 12, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67157a50>,\n","  [0, 5, 13, 6, 19, 9, 2, 17, 4, 12, 18, 3, 11, 1, 7, 8, 15, 10, 16, 14, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6716aad0>,\n","  [0, 3, 10, 15, 17, 6, 8, 1, 7, 5, 12, 18, 13, 14, 4, 11, 2, 16, 19, 9, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6717db10>,\n","  [0, 6, 9, 7, 13, 8, 10, 4, 19, 17, 15, 2, 12, 11, 3, 5, 1, 16, 18, 14, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67194bd0>,\n","  [0, 18, 8, 7, 3, 16, 5, 4, 11, 9, 19, 13, 12, 10, 1, 6, 14, 2, 15, 17, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6719bc10>,\n","  [0, 4, 19, 13, 3, 12, 9, 1, 16, 14, 11, 8, 7, 6, 10, 5, 17, 15, 2, 18, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b671b2c50>,\n","  [0, 3, 12, 1, 13, 5, 19, 9, 2, 11, 17, 16, 18, 14, 7, 4, 10, 15, 6, 8, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b671c1d10>,\n","  [0, 8, 6, 10, 16, 9, 5, 13, 3, 19, 11, 18, 12, 15, 2, 1, 4, 17, 7, 14, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b671d8d90>,\n","  [0, 7, 10, 17, 12, 6, 16, 2, 9, 11, 15, 14, 8, 4, 19, 3, 13, 1, 5, 18, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b671e3d90>,\n","  [0, 19, 17, 2, 7, 1, 12, 3, 11, 16, 10, 14, 6, 8, 15, 18, 5, 9, 4, 13, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b671f6e50>,\n","  [0, 4, 7, 5, 8, 2, 14, 13, 16, 12, 15, 18, 6, 19, 3, 10, 9, 17, 11, 1, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6700de90>,\n","  [0, 18, 19, 6, 11, 2, 10, 3, 14, 15, 12, 9, 8, 1, 5, 16, 17, 4, 13, 7, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67024f50>,\n","  [0, 13, 18, 7, 16, 4, 19, 5, 2, 15, 8, 9, 6, 12, 17, 14, 1, 3, 11, 10, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6702ff90>,\n","  [0, 4, 18, 9, 1, 13, 7, 19, 12, 14, 5, 11, 6, 3, 15, 2, 17, 8, 16, 10, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67042fd0>,\n","  [0, 14, 19, 12, 13, 6, 8, 11, 4, 2, 1, 10, 18, 7, 16, 3, 5, 17, 15, 9, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67056090>,\n","  [0, 10, 8, 11, 2, 18, 12, 13, 16, 14, 17, 6, 15, 4, 9, 1, 3, 19, 7, 5, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b67065110>,\n","  [0, 10, 14, 8, 13, 12, 19, 18, 17, 16, 9, 4, 1, 11, 5, 2, 15, 6, 3, 7, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6707c150>,\n","  [0, 15, 2, 16, 3, 12, 1, 14, 10, 18, 4, 11, 7, 9, 8, 19, 5, 17, 13, 6, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b670871d0>,\n","  [0, 2, 16, 4, 5, 6, 12, 3, 14, 11, 8, 9, 15, 7, 19, 18, 17, 10, 1, 13, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b6709e250>,\n","  [0, 3, 13, 11, 2, 8, 17, 6, 10, 14, 1, 19, 4, 18, 15, 16, 9, 7, 12, 5, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b670b1290>,\n","  [0, 17, 16, 3, 4, 14, 6, 8, 2, 12, 1, 10, 7, 19, 13, 15, 9, 5, 11, 18, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b670c4350>,\n","  [0, 1, 11, 16, 17, 7, 9, 8, 13, 15, 3, 2, 19, 5, 6, 14, 12, 4, 18, 10, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b670d3350>,\n","  [0, 7, 15, 19, 18, 10, 4, 12, 5, 11, 3, 17, 13, 16, 9, 6, 8, 1, 14, 2, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b670e2410>,\n","  [0, 11, 2, 8, 15, 5, 4, 18, 7, 3, 17, 10, 6, 13, 12, 9, 1, 19, 14, 16, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b670f9490>,\n","  [0, 10, 18, 9, 1, 7, 6, 14, 19, 16, 8, 17, 15, 12, 3, 4, 11, 2, 5, 13, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b66f084d0>,\n","  [0, 7, 16, 17, 12, 3, 10, 2, 1, 6, 5, 19, 4, 15, 13, 11, 9, 18, 14, 8, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b66f1b550>,\n","  [0, 13, 10, 6, 17, 11, 9, 19, 7, 14, 18, 4, 3, 1, 8, 15, 12, 2, 16, 5, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b66f2a5d0>,\n","  [0, 16, 8, 13, 9, 2, 11, 4, 3, 17, 15, 18, 10, 7, 12, 1, 6, 5, 14, 19, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b66f41610>,\n","  [0, 2, 7, 10, 11, 17, 14, 6, 1, 15, 18, 9, 16, 19, 13, 4, 12, 8, 3, 5, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b66f546d0>,\n","  [0, 3, 5, 15, 8, 10, 13, 17, 4, 16, 12, 11, 14, 9, 7, 2, 6, 18, 19, 1, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b66f5f710>,\n","  [0, 15, 17, 2, 12, 14, 11, 1, 5, 6, 4, 8, 18, 13, 10, 3, 9, 7, 19, 16, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b66f76750>,\n","  [0, 14, 16, 12, 4, 18, 10, 11, 9, 15, 7, 1, 6, 13, 19, 8, 17, 3, 5, 2, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b66f89810>,\n","  [0, 12, 5, 11, 1, 4, 13, 9, 2, 15, 8, 16, 18, 17, 6, 3, 7, 14, 10, 19, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b66f9c850>,\n","  [0, 1, 5, 10, 7, 19, 18, 6, 14, 16, 8, 15, 2, 17, 4, 9, 3, 12, 13, 11, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b66fa78d0>,\n","  [0, 3, 9, 10, 6, 12, 5, 17, 7, 13, 2, 1, 18, 15, 11, 4, 14, 8, 16, 19, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b66fbe950>,\n","  [0, 10, 12, 8, 6, 19, 7, 16, 1, 4, 2, 11, 18, 14, 13, 15, 5, 3, 9, 17, 0]),\n"," (<networkx.classes.graph.Graph at 0x7f5b66fcd990>,\n","  [0, 17, 15, 16, 9, 6, 1, 3, 12, 10, 8, 4, 19, 13, 5, 14, 11, 2, 18, 7, 0])]"]},"execution_count":7,"metadata":{},"output_type":"execute_result"}],"source":["dummy"]},{"cell_type":"code","execution_count":8,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["100\n","torch.Size([20, 2]) torch.Size([21])\n","tensor([[0.6049, 0.5749],\n","        [0.3847, 0.9428],\n","        [0.6102, 0.3708],\n","        [0.5498, 0.5148],\n","        [0.5942, 0.6544],\n","        [0.0619, 0.4332],\n","        [0.7476, 0.3951],\n","        [0.6549, 0.0625],\n","        [0.2211, 0.2142],\n","        [0.6697, 0.4364],\n","        [0.3206, 0.7886],\n","        [0.3776, 0.4964],\n","        [0.8300, 0.3489],\n","        [0.9984, 0.2725],\n","        [0.5306, 0.4147],\n","        [0.2541, 0.8878],\n","        [0.7025, 0.2586],\n","        [0.1113, 0.3599],\n","        [0.3338, 0.1718],\n","        [0.8953, 0.6299]]) tensor([ 0,  3, 14,  2,  9,  6, 19, 13, 12, 16,  7, 18,  8, 17,  5, 11, 10, 15,\n","         1,  4,  0])\n"]}],"source":["from torch.utils.data import Dataset, DataLoader\n","\n","class TSPDataset(Dataset):\n","    def __init__(self, data):\n","        super().__init__()\n","        self.data = data\n","        \n","    def __len__(self):\n","        return len(self.data)\n","    \n","    def __getitem__(self, idx):\n","        G, path = self.data[idx]\n","        node_pos = nx.get_node_attributes(G, 'pos')\n","        pos = [v for k,v in node_pos.items()]\n","        return torch.tensor(pos), torch.tensor(path)\n","\n","ds = TSPDataset(dummy)\n","print(len(ds))\n","pos, tour = ds[0]\n","assert pos.shape == (20, 2)\n","assert tour[0] == 0 and tour[-1] == 0\n","print(pos.shape, tour.shape)\n","print(pos, tour)"]},{"cell_type":"code","execution_count":9,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["torch.Size([4, 20, 2]) torch.Size([4, 21])\n","tensor([[[8.2292e-01, 3.0425e-01],\n","         [4.3008e-01, 8.9945e-01],\n","         [3.8127e-01, 4.6026e-01],\n","         [1.1408e-01, 1.0030e-01],\n","         [8.6969e-01, 2.5279e-03],\n","         [8.9233e-01, 1.2366e-01],\n","         [4.5736e-01, 7.9768e-01],\n","         [2.9592e-02, 9.8846e-01],\n","         [4.6930e-01, 8.3466e-02],\n","         [5.8475e-01, 8.9096e-01],\n","         [4.1078e-01, 6.6984e-01],\n","         [4.1073e-01, 4.0475e-01],\n","         [8.2821e-01, 8.6692e-01],\n","         [3.7448e-01, 1.6267e-01],\n","         [3.9771e-01, 3.1606e-01],\n","         [7.2873e-01, 7.2218e-01],\n","         [6.6316e-01, 3.7021e-01],\n","         [2.0587e-02, 5.4851e-01],\n","         [6.3497e-01, 1.7894e-01],\n","         [5.5469e-01, 8.7134e-01]],\n","\n","        [[3.2790e-04, 2.0287e-01],\n","         [6.1626e-01, 5.7911e-01],\n","         [8.3704e-01, 7.4467e-01],\n","         [9.4347e-01, 8.5304e-01],\n","         [6.4971e-01, 6.6967e-01],\n","         [3.8413e-01, 1.1949e-01],\n","         [8.7256e-01, 5.2821e-01],\n","         [4.2183e-01, 8.0707e-01],\n","         [9.9198e-01, 9.7719e-01],\n","         [9.0114e-01, 2.1798e-01],\n","         [4.4380e-01, 1.2388e-01],\n","         [5.0851e-01, 9.1099e-01],\n","         [4.0889e-02, 4.5401e-01],\n","         [4.0119e-02, 9.0455e-01],\n","         [8.6999e-01, 2.6260e-01],\n","         [7.1178e-01, 3.1264e-01],\n","         [5.1870e-01, 6.6795e-01],\n","         [1.9405e-01, 3.6388e-01],\n","         [2.8712e-02, 8.6977e-01],\n","         [4.6529e-01, 6.9320e-01]],\n","\n","        [[3.8823e-03, 8.8070e-01],\n","         [6.0344e-01, 5.8977e-01],\n","         [5.6831e-01, 5.0574e-01],\n","         [6.1961e-01, 5.4422e-01],\n","         [6.5992e-01, 5.3966e-01],\n","         [6.4312e-02, 4.8416e-01],\n","         [2.1732e-01, 9.5308e-01],\n","         [3.9094e-01, 3.8639e-01],\n","         [8.9557e-01, 6.0599e-01],\n","         [9.9240e-01, 1.9543e-02],\n","         [9.7877e-01, 8.2222e-01],\n","         [4.3404e-01, 9.0780e-01],\n","         [3.1879e-01, 7.6116e-02],\n","         [8.4235e-01, 6.2974e-01],\n","         [8.6191e-01, 9.9339e-01],\n","         [1.5690e-01, 1.2526e-01],\n","         [9.4032e-01, 8.9728e-02],\n","         [6.8744e-01, 8.1421e-01],\n","         [8.5862e-01, 5.3488e-01],\n","         [9.6934e-01, 1.3635e-01]],\n","\n","        [[7.6072e-01, 2.9718e-01],\n","         [6.9526e-01, 1.3849e-01],\n","         [2.2629e-01, 1.6439e-01],\n","         [9.3316e-01, 9.3925e-01],\n","         [2.7506e-01, 9.9285e-01],\n","         [1.9192e-01, 7.7671e-01],\n","         [7.9265e-01, 1.0055e-01],\n","         [4.7861e-01, 2.8525e-02],\n","         [9.8657e-01, 8.2158e-01],\n","         [8.9148e-01, 2.1947e-01],\n","         [3.4004e-01, 6.2431e-01],\n","         [7.0610e-01, 3.2629e-01],\n","         [8.1001e-01, 1.4043e-01],\n","         [1.0817e-01, 9.0770e-02],\n","         [9.0766e-01, 8.1247e-01],\n","         [4.4468e-01, 6.9707e-01],\n","         [4.6825e-01, 1.0058e-01],\n","         [1.0917e-01, 3.2780e-01],\n","         [7.2683e-02, 5.8032e-01],\n","         [8.7685e-02, 2.8376e-01]]]) tensor([[ 0, 16, 14, 11,  2, 10,  6, 15, 12,  9, 19,  1,  7, 17,  3, 13,  8, 18,\n","          4,  5,  0],\n","        [ 0,  5, 10, 15, 14,  9,  6,  3,  8,  2,  4,  1, 16, 19,  7, 11, 13, 18,\n","         12, 17,  0],\n","        [ 0,  6, 11, 17, 14, 10, 13,  8, 18, 19,  9, 16,  4,  3,  1,  2,  7, 12,\n","         15,  5,  0],\n","        [ 0, 11, 14,  8,  3, 15, 10,  4,  5, 18, 17, 19, 13,  2, 16,  7,  1,  6,\n","         12,  9,  0]])\n"]}],"source":["train = pickle.loads(open('data/train_20_DLL_ass4.pkl', 'rb').read())\n","valid = pickle.loads(open('data/valid_20_DLL_ass4.pkl', 'rb').read())\n","test = pickle.loads(open('data/test_20_DLL_ass4.pkl', 'rb').read())\n","\n","train, valid, test = TSPDataset(train), TSPDataset(valid), TSPDataset(test)\n","\n","len(train), len(valid), len(test)\n","\n","train_dl = DataLoader(train, batch_size=4, shuffle=True)\n","valid_dl = DataLoader(valid, batch_size=4, shuffle=True)\n","valid_dl = DataLoader(test, batch_size=4, shuffle=True)\n","\n","## Make sure batching and collation works as intended...\n","for sample in train_dl:\n","    break\n","pos, path = sample\n","print(pos.shape, path.shape)\n","print(pos, path)"]},{"cell_type":"markdown","metadata":{},"source":["## Model"]},{"cell_type":"code","execution_count":10,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["torch.Size([1, 4, 2]) torch.Size([1, 4])\n","torch.Size([1, 4, 4])\n"]},{"name":"stderr","output_type":"stream","text":["/opt/conda/lib/python3.11/site-packages/torch/nn/modules/transformer.py:379: UserWarning: enable_nested_tensor is True, but self.use_nested_tensor is False because encoder_layer.self_attn.num_heads is odd\n","  warnings.warn(\n"]}],"source":["class Model(nn.Module):\n","    def __init__(self, n_emb, nlayers, d_model=32, nhead=1 ):\n","        super().__init__()\n","        self.linear = nn.Linear(2, d_model)\n","        self.encoder = nn.TransformerEncoder(\n","            nn.TransformerEncoderLayer(d_model=d_model, nhead=nhead, batch_first=True, activation=F.gelu),\n","            num_layers=nlayers,\n","        )\n","\n","        self.enc2dec = nn.Linear(d_model, d_model) # dims are the same\n","        self.embedding = nn.Embedding(n_emb, d_model)\n","        self.pos_emb = nn.Parameter(torch.randn(1, n_emb, d_model))\n","        self.decoder = nn.TransformerDecoder(\n","            nn.TransformerDecoderLayer(d_model=d_model, nhead=1, batch_first=True, activation=F.gelu),\n","            num_layers=nlayers,\n","        )\n","        self.to_logits = nn.Linear(d_model, n_emb)\n","\n","    def get_mask(self, sz):\n","        mask = (torch.triu(torch.ones(sz, sz)) == 1).transpose(0, 1)\n","        mask = mask.float().masked_fill(mask == 0, float('-inf')).masked_fill(mask == 1, float(0.0))\n","        return mask\n","\n","    def forward(self, x, y):\n","        # x [B, Nodes, 2]\n","        enc_embs = self.linear(x) # \n","        # print('encs', enc_embs.shape)\n","        # enc_embs [B, Nodes, d_model]\n","        enc_out = self.encoder(enc_embs)\n","        # enc_out [N, Nodes, d_model]\n","        dec_qk = self.enc2dec(enc_out)\n","\n","        # y [B, n_emb]\n","        dec_emb = self.embedding(y) \n","        # dec_emb [B, n_emb, d_model]\n","        b, n, d = dec_emb.shape\n","        dec_emb += self.pos_emb[:, :n, :]\n","        # print(\"dec_emb.shape, dec_qk.shape\", dec_emb.shape, dec_qk.shape)\n","        tgt_mask = self.get_mask(dec_emb.size(1)).to(x.device)\n","        dec_out = self.decoder(tgt = dec_emb, memory = dec_qk, tgt_mask=tgt_mask, tgt_is_causal=True)\n","        # print('dec_out.shape', dec_out.shape)\n","        # >>> memory = torch.rand(10, 32, 512)\n","        # >>> tgt = torch.rand(20, 32, 512)\n","        # >>> out = transformer_decoder(tgt, memory)\n","        logits = self.to_logits(dec_out)\n","        return logits\n","\n","# pos, path\n","model = Model(n_emb=4, nlayers=4, d_model=32)\n","# model(torch.)\n","x = torch.tensor(\n","   [[0, 0],\n","    [0, 1],\n","    [1, 1],\n","    [1, 0]],\n",").float().reshape(1, 4, 2)\n","y = torch.tensor([\n","    [0, 1, 2, 3],\n","]).int().reshape(1, 4)\n","\n","print(x.shape, y.shape)\n","x = model(x, y)\n","print(x.shape)"]},{"cell_type":"markdown","metadata":{},"source":["(10 pts) Explain where and why masking is used.\n","\n","Answer: Masking is used in the decoder. \n","\n","Decoder is a essentially a causal model that predicts t+1 token from t tokens.\n","In order to \"fairly\" predict the next token, a causal mask is used in the attention layer\n","of the decoder. Causal mask is a triangular lower matrix (1 means that token can attend)\n","0 means not. \n","To be more specific:\n","given sequence [0,1,2,3], the qkv projection layers encode\n","each token like so [e0, e1, e2, e4]. We have 3 tensors q, k and v. \n","We first do q.T @ k, which essentially does an outer product between all ei embeddings.\n","It is at this point that we mutilply this (q.T @ k) * trilu. This means that q and k embeddings\n","Can only 'talk' with previous tokens. \n","\n","(5 pts) Explain why positional encoding may be omitted in the encoder.\n","\n","\n","Answer: We are passing in a graph node position information. Position information should \n","not depend on ordering, just like the nodes of the graph should not depend on ordering.\n","Omitting PE layer means that encoder doesn't have a way to know positions of tokens.\n","\n","(5 pts) Explain why positional encoding is necessary in the decoder.\n","\n","\n","Decoder is fed a path. Path is an ordered sequence of steps. Just like with words, \n","rearranging words in a sequence might lead to errors. Therefore, the model\n","must know positions of the words. This is why we add PE to the path inputs to the decoder."]},{"cell_type":"markdown","metadata":{},"source":["## Training"]},{"cell_type":"markdown","metadata":{},"source":["### Training WITHOUT gradient accumulation"]},{"cell_type":"code","execution_count":12,"metadata":{},"outputs":[{"name":"stderr","output_type":"stream","text":["loss 2.8297: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.41it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.8136\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.7312: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.88it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7237\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.5975: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.49it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.5900\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.4426: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.47it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.4268\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.2759: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:07<00:00,  1.83it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.2678\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.1609: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.54it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.1437\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0933: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.50it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0891\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0544: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:07<00:00,  1.83it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0599\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0400: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.51it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0430\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0331: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.49it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0358\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0312: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:07<00:00,  1.85it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0331\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0242: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.52it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0279\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0192: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.50it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0245\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0144: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:07<00:00,  1.84it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0243\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0130: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.52it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0222\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0170: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.48it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0217\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0139: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.86it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0180\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0132: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.54it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0190\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0117: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.55it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0171\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0139: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:07<00:00,  1.85it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0166\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0113: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.52it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0174\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0102: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.55it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0117\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0105: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.87it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0145\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0044: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.50it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0159\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0039: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.54it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0137\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0017: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.92it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0138\n"]},{"name":"stderr","output_type":"stream","text":["loss 1.9991: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.52it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0114\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0015: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.54it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0080\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.0061: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.87it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0112\n"]},{"name":"stderr","output_type":"stream","text":["loss 1.9989: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.55it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0080\n"]},{"name":"stderr","output_type":"stream","text":["loss 1.9967: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.54it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0102\n"]},{"name":"stderr","output_type":"stream","text":["loss 1.9990: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.88it/s]"]},{"name":"stdout","output_type":"stream","text":["val loss 2.0098\n"]},{"name":"stderr","output_type":"stream","text":["\n"]}],"source":["device = 'cuda' if torch.cuda.is_available() else 'cpu'\n","epochs = 32\n","model = Model(n_emb=20, nlayers=4, d_model=256, nhead=4)\n","model = model.to(device)\n","\n","optim = torch.optim.AdamW(model.parameters(), lr=1e-4)\n","train_dl = DataLoader(train, batch_size=1024 * 4, shuffle=True)\n","valid_dl = DataLoader(valid, batch_size=512, shuffle=False)\n","test_dl = DataLoader(test,  batch_size=512, shuffle=False)\n","\n","val_loss   = []\n","train_loss = []\n","for epoch in range(epochs):\n","    total_loss = torch.tensor(0.)\n","    pbar = tqdm(enumerate(train_dl), colour='green', total=len(train_dl), )\n","    for step, (pos, path) in pbar:\n","        optim.zero_grad()\n","        pos, path = pos.to(device), path.to(device)\n","        path_in =  path[:, :-1] # before last\n","        path_out = path[:, 1:] # after first\n","        logits = model(pos, path_in)\n","        loss = F.cross_entropy(\n","            logits.transpose(1, 2),\n","            path_out,\n","        )\n","        loss.backward()\n","        optim.step()\n","        total_loss += loss.item()\n","        pbar.set_description(f'loss {loss.item():.4f}')\n","    total_loss /= len(train_dl)\n","    train_loss.append(total_loss)\n","\n","    with torch.no_grad():\n","        total_loss = torch.tensor(0.)\n","        for step, (pos, path) in enumerate(valid_dl):\n","            pos, path = pos.to(device), path.to(device)\n","            path_in =  path[:,:-1] # before last\n","            path_out = path[:, 1:] # after first\n","            logits = model(pos, path_in)\n","            total_loss += F.cross_entropy(\n","                logits.transpose(1, 2),\n","                path_out,\n","            ).item()\n","        total_loss /= len(valid_dl)\n","        print(f'val loss {total_loss:.4f}')\n","        val_loss.append(total_loss)"]},{"cell_type":"code","execution_count":13,"metadata":{},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 640x480 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":["plt.figure()\n","plt.title(\"Train/val loss no grad accumulation\")\n","plt.plot(range(epochs), train_loss, label='training')\n","plt.plot(range(epochs), val_loss, label='validation')\n","plt.xlabel('# epoch')\n","plt.ylabel('CrossEntropy loss')\n","plt.legend()\n","plt.tight_layout()\n","plt.savefig(\"loss_nograd.png\", dpi=400)\n","plt.show()"]},{"cell_type":"code","execution_count":14,"metadata":{},"outputs":[],"source":["torch.save(model.state_dict(), open('model_nograd.pt', 'wb'))"]},{"cell_type":"markdown","metadata":{},"source":["(10 pts) Explain the function transformer tsp, which uses the model you\n","trained to predict a tour. You should notice that it’s close to the greedy\n","sampling strategy we used in NLP task. Which is the main difference?"]},{"cell_type":"markdown","metadata":{},"source":["transformer_tsp generates a TSP path using a trained model. It starts by feeding in transformer an initial state\n","and then predicting most likely next step. We sample at most N state transitions greedily (with topk). And then calculate the total cost of \n","the path made from transformer so far. Main difference is that we disallow revisiting the old states. "]},{"cell_type":"markdown","metadata":{},"source":["### Training WITH gradient accumulation"]},{"cell_type":"code","execution_count":16,"metadata":{},"outputs":[{"name":"stderr","output_type":"stream","text":["  0%|\u001b[32m          \u001b[0m| 0/13 [00:00<?, ?it/s]"]},{"name":"stderr","output_type":"stream","text":["loss 3.2551: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.52it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 3.3436\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.9150: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.51it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.8547\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8734: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.91it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.8592\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8260: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.57it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.8466\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8211: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.55it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.8186\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8174: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.96it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.8100\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8125: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.56it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.8050\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8097: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.54it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.8035\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8073: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.89it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7998\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8094: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.54it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.8004\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8065: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.54it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7983\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8056: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.58it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7988\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8066: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.96it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7979\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8053: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.52it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7979\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8051: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.55it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7976\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8047: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.87it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7976\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8045: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.54it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7974\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8047: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.53it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7974\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8046: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.94it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7974\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8037: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.57it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7974\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8041: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.56it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7973\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8025: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.92it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7973\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8021: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.53it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7973\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8031: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.55it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7973\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8034: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.91it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7973\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8043: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.56it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7973\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8030: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.56it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7973\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8027: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.96it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7972\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8008: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.56it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7972\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8008: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.57it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7972\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8020: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:06<00:00,  1.93it/s]\n"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7972\n"]},{"name":"stderr","output_type":"stream","text":["loss 2.8008: 100%|\u001b[32m██████████\u001b[0m| 13/13 [00:05<00:00,  2.57it/s]"]},{"name":"stdout","output_type":"stream","text":["val loss 2.7972\n"]},{"name":"stderr","output_type":"stream","text":["\n"]}],"source":["epochs = 32\n","model = Model(n_emb=20, nlayers=4, d_model=256, nhead=4)\n","model = model.to(device)\n","optim = torch.optim.AdamW(model.parameters(), lr=1e-3) # Larger LR, since we see \n","# less # updates and more effective batches x4\n","train_dl = DataLoader(train, batch_size=1024 * 4, shuffle=True)\n","valid_dl = DataLoader(valid, batch_size=512, shuffle=False)\n","test_dl = DataLoader(test,  batch_size=512, shuffle=False)\n","\n","val_loss = []\n","train_loss = []\n","\n","# Gradient accumulation steps\n","gradient_accumulation_steps = 4\n","\n","for epoch in range(epochs):\n","    model.train()\n","    total_loss = torch.tensor(0.)\n","    pbar = tqdm(enumerate(train_dl), colour='green', total=len(train_dl))\n","\n","    # Reset the optimizer at the start of each epoch\n","    optim.zero_grad()\n","\n","    for step, (pos, path) in pbar:\n","        pos, path = pos.to(device), path.to(device)\n","        path_in = path[:, :-1]  # before last\n","        path_out = path[:, 1:]  # after first\n","        logits = model(pos, path_in)\n","        loss = F.cross_entropy(\n","            logits.transpose(1, 2),\n","            path_out,\n","        )\n","\n","        # Normalize loss by accumulation steps and backpropagate\n","        loss = loss / gradient_accumulation_steps\n","        loss.backward()\n","\n","        # Update weights after every accumulation step\n","        if (step + 1) % gradient_accumulation_steps == 0 or (step + 1) == len(train_dl):\n","            optim.step()\n","            optim.zero_grad()\n","\n","        total_loss += loss.item() * gradient_accumulation_steps  # Undo normalization for logging\n","        pbar.set_description(f'loss {loss.item() * gradient_accumulation_steps:.4f}')\n","\n","    total_loss /= len(train_dl)\n","    train_loss.append(total_loss.item())\n","\n","    model.eval()\n","    with torch.no_grad():\n","        total_val_loss = torch.tensor(0.)\n","        for step, (pos, path) in enumerate(valid_dl):\n","            pos, path = pos.to(device), path.to(device)\n","            path_in = path[:, :-1]  # before last\n","            path_out = path[:, 1:]  # after first\n","            logits = model(pos, path_in)\n","            total_val_loss += F.cross_entropy(\n","                logits.transpose(1, 2),\n","                path_out,\n","            ).item()\n","\n","        total_val_loss /= len(valid_dl)\n","        print(f'val loss {total_val_loss:.4f}')\n","        val_loss.append(total_val_loss.item())"]},{"cell_type":"code","execution_count":25,"metadata":{},"outputs":[{"data":{"text/plain":["11657236"]},"execution_count":25,"metadata":{},"output_type":"execute_result"}],"source":["sum([p.numel() for p in model.parameters() if p.requires_grad])"]},{"cell_type":"code","execution_count":17,"metadata":{},"outputs":[],"source":["torch.save(model.state_dict(), open('model.pt', 'wb'))"]},{"cell_type":"code","execution_count":19,"metadata":{},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 640x480 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":["plt.figure()\n","plt.title(\"Train/val loss grad acc\")\n","plt.plot(range(epochs), train_loss, label='training')\n","plt.plot(range(epochs), val_loss, label='validation')\n","plt.xlabel('# epoch')\n","plt.ylabel('CrossEntropy loss')\n","plt.legend()\n","plt.tight_layout()\n","plt.savefig(\"loss_grad.png\", dpi=400)\n","plt.show()"]},{"cell_type":"markdown","metadata":{},"source":["## Testing"]},{"cell_type":"code","execution_count":20,"metadata":{},"outputs":[{"name":"stderr","output_type":"stream","text":["/tmp/ipykernel_5793/2629809984.py:2: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n","  model_GA.load_state_dict(torch.load(open('model.pt','rb'), map_location=device))\n","/tmp/ipykernel_5793/2629809984.py:6: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.\n","  model.load_state_dict(torch.load(open('model_nograd.pt','rb'), map_location=device))\n"]}],"source":["model_GA = Model(n_emb=20, nlayers=4, d_model=256, nhead=4)\n","model_GA.load_state_dict(torch.load(open('model.pt','rb'), map_location=device))\n","model_GA = model_GA.to(device)\n","\n","model = Model(n_emb=20, nlayers=4, d_model=256, nhead=4)\n","model.load_state_dict(torch.load(open('model_nograd.pt','rb'), map_location=device))\n","model = model.to(device)"]},{"cell_type":"code","execution_count":21,"metadata":{},"outputs":[{"name":"stderr","output_type":"stream","text":["100%|██████████| 20/20 [00:19<00:00,  1.02it/s]\n"]}],"source":["test = pickle.loads(open('data/test_20_DLL_ass4.pkl', 'rb').read())\n","\n","results = []\n","for sample in tqdm(random.choices(test, k=20)):\n","    G, path = sample\n","    results.append(gap(G, model, model_GA=model_GA,  random_seed=42, device=device))\n","df = pd.DataFrame(results)"]},{"cell_type":"code","execution_count":22,"metadata":{},"outputs":[{"data":{"text/html":["<div>\n","<style scoped>\n","    .dataframe tbody tr th:only-of-type {\n","        vertical-align: middle;\n","    }\n","\n","    .dataframe tbody tr th {\n","        vertical-align: top;\n","    }\n","\n","    .dataframe thead th {\n","        text-align: right;\n","    }\n","</style>\n","<table border=\"1\" class=\"dataframe\">\n","  <thead>\n","    <tr style=\"text-align: right;\">\n","      <th></th>\n","      <th>greedy</th>\n","      <th>random</th>\n","      <th>transformer_tsp</th>\n","      <th>transformer_tsp_acc_grad</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>21.915325</td>\n","      <td>152.745933</td>\n","      <td>157.036939</td>\n","      <td>135.066600</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>6.330246</td>\n","      <td>163.914135</td>\n","      <td>214.220333</td>\n","      <td>182.875009</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>18.551933</td>\n","      <td>232.861386</td>\n","      <td>215.653008</td>\n","      <td>187.928569</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>3.229360</td>\n","      <td>211.595870</td>\n","      <td>157.769503</td>\n","      <td>167.299160</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>27.574934</td>\n","      <td>301.523497</td>\n","      <td>246.999309</td>\n","      <td>199.097642</td>\n","    </tr>\n","    <tr>\n","      <th>5</th>\n","      <td>16.661699</td>\n","      <td>86.992984</td>\n","      <td>126.482855</td>\n","      <td>111.847432</td>\n","    </tr>\n","    <tr>\n","      <th>6</th>\n","      <td>10.966365</td>\n","      <td>128.905445</td>\n","      <td>234.120014</td>\n","      <td>228.825281</td>\n","    </tr>\n","    <tr>\n","      <th>7</th>\n","      <td>4.037199</td>\n","      <td>175.594184</td>\n","      <td>231.501946</td>\n","      <td>195.548668</td>\n","    </tr>\n","    <tr>\n","      <th>8</th>\n","      <td>35.961583</td>\n","      <td>153.199773</td>\n","      <td>208.160983</td>\n","      <td>208.771310</td>\n","    </tr>\n","    <tr>\n","      <th>9</th>\n","      <td>13.669226</td>\n","      <td>159.944215</td>\n","      <td>121.647367</td>\n","      <td>203.328338</td>\n","    </tr>\n","    <tr>\n","      <th>10</th>\n","      <td>2.974459</td>\n","      <td>166.911349</td>\n","      <td>161.352725</td>\n","      <td>153.782248</td>\n","    </tr>\n","    <tr>\n","      <th>11</th>\n","      <td>22.001170</td>\n","      <td>157.165433</td>\n","      <td>151.330840</td>\n","      <td>137.472440</td>\n","    </tr>\n","    <tr>\n","      <th>12</th>\n","      <td>0.083218</td>\n","      <td>209.198722</td>\n","      <td>240.475031</td>\n","      <td>199.036424</td>\n","    </tr>\n","    <tr>\n","      <th>13</th>\n","      <td>5.027839</td>\n","      <td>211.155508</td>\n","      <td>142.870202</td>\n","      <td>190.654780</td>\n","    </tr>\n","    <tr>\n","      <th>14</th>\n","      <td>22.597242</td>\n","      <td>151.570243</td>\n","      <td>200.765555</td>\n","      <td>204.435379</td>\n","    </tr>\n","    <tr>\n","      <th>15</th>\n","      <td>9.779923</td>\n","      <td>171.475655</td>\n","      <td>191.014814</td>\n","      <td>150.559510</td>\n","    </tr>\n","    <tr>\n","      <th>16</th>\n","      <td>15.545332</td>\n","      <td>137.981856</td>\n","      <td>153.376619</td>\n","      <td>155.536949</td>\n","    </tr>\n","    <tr>\n","      <th>17</th>\n","      <td>2.568747</td>\n","      <td>159.049363</td>\n","      <td>115.029562</td>\n","      <td>171.753825</td>\n","    </tr>\n","    <tr>\n","      <th>18</th>\n","      <td>16.276164</td>\n","      <td>226.068685</td>\n","      <td>188.320188</td>\n","      <td>168.130590</td>\n","    </tr>\n","    <tr>\n","      <th>19</th>\n","      <td>26.492507</td>\n","      <td>159.643639</td>\n","      <td>131.145517</td>\n","      <td>180.329804</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["       greedy      random  transformer_tsp  transformer_tsp_acc_grad\n","0   21.915325  152.745933       157.036939                135.066600\n","1    6.330246  163.914135       214.220333                182.875009\n","2   18.551933  232.861386       215.653008                187.928569\n","3    3.229360  211.595870       157.769503                167.299160\n","4   27.574934  301.523497       246.999309                199.097642\n","5   16.661699   86.992984       126.482855                111.847432\n","6   10.966365  128.905445       234.120014                228.825281\n","7    4.037199  175.594184       231.501946                195.548668\n","8   35.961583  153.199773       208.160983                208.771310\n","9   13.669226  159.944215       121.647367                203.328338\n","10   2.974459  166.911349       161.352725                153.782248\n","11  22.001170  157.165433       151.330840                137.472440\n","12   0.083218  209.198722       240.475031                199.036424\n","13   5.027839  211.155508       142.870202                190.654780\n","14  22.597242  151.570243       200.765555                204.435379\n","15   9.779923  171.475655       191.014814                150.559510\n","16  15.545332  137.981856       153.376619                155.536949\n","17   2.568747  159.049363       115.029562                171.753825\n","18  16.276164  226.068685       188.320188                168.130590\n","19  26.492507  159.643639       131.145517                180.329804"]},"execution_count":22,"metadata":{},"output_type":"execute_result"}],"source":["df"]},{"cell_type":"code","execution_count":23,"metadata":{},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 640x480 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":["df.boxplot()\n","plt.ylabel('TSP total path cost')\n","plt.tight_layout()\n","plt.savefig('tsp_box.png')\n","plt.show()"]}],"metadata":{"kaggle":{"accelerator":"none","dataSources":[],"dockerImageVersionId":30804,"isGpuEnabled":false,"isInternetEnabled":true,"language":"python","sourceType":"notebook"},"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.11.10"}},"nbformat":4,"nbformat_minor":4}