code_2_test.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "ename": "OSError",
     "evalue": "libc10_cuda.so: cannot open shared object file: No such file or directory",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mOSError\u001b[0m                                   Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[50], line 12\u001b[0m\n\u001b[1;32m     10\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtorch_geometric\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mnn\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m Node2Vec\n\u001b[1;32m     11\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mitertools\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m permutations\n\u001b[0;32m---> 12\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtorch_cluster\u001b[39;00m\n\u001b[1;32m     13\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msklearn\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mmanifold\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m TSNE\n",
      "File \u001b[0;32m~/miniconda/envs/weber_env_2/lib/python3.8/site-packages/torch_cluster/__init__.py:18\u001b[0m\n\u001b[1;32m     16\u001b[0m spec \u001b[38;5;241m=\u001b[39m cuda_spec \u001b[38;5;129;01mor\u001b[39;00m cpu_spec\n\u001b[1;32m     17\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m spec \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m---> 18\u001b[0m     \u001b[43mtorch\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mops\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mload_library\u001b[49m\u001b[43m(\u001b[49m\u001b[43mspec\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43morigin\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     19\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:  \u001b[38;5;66;03m# pragma: no cover\u001b[39;00m\n\u001b[1;32m     20\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mImportError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mCould not find module \u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mlibrary\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m_cpu\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m in \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m     21\u001b[0m                       \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mosp\u001b[38;5;241m.\u001b[39mdirname(\u001b[38;5;18m__file__\u001b[39m)\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n",
      "File \u001b[0;32m~/miniconda/envs/weber_env_2/lib/python3.8/site-packages/torch/_ops.py:933\u001b[0m, in \u001b[0;36m_Ops.load_library\u001b[0;34m(self, path)\u001b[0m\n\u001b[1;32m    928\u001b[0m path \u001b[38;5;241m=\u001b[39m _utils_internal\u001b[38;5;241m.\u001b[39mresolve_library_path(path)\n\u001b[1;32m    929\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m dl_open_guard():\n\u001b[1;32m    930\u001b[0m     \u001b[38;5;66;03m# Import the shared library into the process, thus running its\u001b[39;00m\n\u001b[1;32m    931\u001b[0m     \u001b[38;5;66;03m# static (global) initialization code in order to register custom\u001b[39;00m\n\u001b[1;32m    932\u001b[0m     \u001b[38;5;66;03m# operators with the JIT.\u001b[39;00m\n\u001b[0;32m--> 933\u001b[0m     \u001b[43mctypes\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mCDLL\u001b[49m\u001b[43m(\u001b[49m\u001b[43mpath\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    934\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mloaded_libraries\u001b[38;5;241m.\u001b[39madd(path)\n",
      "File \u001b[0;32m~/miniconda/envs/weber_env_2/lib/python3.8/ctypes/__init__.py:373\u001b[0m, in \u001b[0;36mCDLL.__init__\u001b[0;34m(self, name, mode, handle, use_errno, use_last_error, winmode)\u001b[0m\n\u001b[1;32m    370\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_FuncPtr \u001b[38;5;241m=\u001b[39m _FuncPtr\n\u001b[1;32m    372\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m handle \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m--> 373\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_handle \u001b[38;5;241m=\u001b[39m \u001b[43m_dlopen\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_name\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmode\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    374\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m    375\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_handle \u001b[38;5;241m=\u001b[39m handle\n",
      "\u001b[0;31mOSError\u001b[0m: libc10_cuda.so: cannot open shared object file: No such file or directory"
     ]
    }
   ],
   "source": [
    "\n",
    "import torch\n",
    "import torchvision.models as models\n",
    "import torchvision.transforms as transforms\n",
    "from PIL import Image\n",
    "import os\n",
    "import numpy as np\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from torch_geometric.data import Data\n",
    "from tqdm import tqdm\n",
    "from torch_geometric.nn import Node2Vec\n",
    "from itertools import permutations\n",
    "import torch_cluster\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Transformation for the images\n",
    "transform = transforms.Compose(\n",
    "    [\n",
    "        transforms.Resize((224, 224)),\n",
    "        transforms.ToTensor(),\n",
    "        transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),\n",
    "    ]\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/n/home02/rskeweszorrilla/miniconda/envs/weber_env_2/lib/python3.8/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n",
      "  warnings.warn(\n",
      "/n/home02/rskeweszorrilla/miniconda/envs/weber_env_2/lib/python3.8/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=VGG16_Weights.IMAGENET1K_V1`. You can also use `weights=VGG16_Weights.DEFAULT` to get the most up-to-date weights.\n",
      "  warnings.warn(msg)\n"
     ]
    }
   ],
   "source": [
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "vgg16 = models.vgg16(pretrained=True).features.to(device)\n",
    "#vgg16 = torch.nn.Sequential(*list(vgg16.children())[:-1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# A function to extract features for a given image\n",
    "def extract_features(img_path):\n",
    "    img = Image.open(img_path).convert(\"RGB\")\n",
    "    img_t = transform(img).unsqueeze(0).to(device)\n",
    "    with torch.no_grad():\n",
    "        features = vgg16(img_t)\n",
    "    return features.cpu().numpy().flatten()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "def extract_image_data(dataset_path, extract_features_fn = None, considered_eras = ['00', '01', '02', '03', '04', '05', '06'], early_stop= False):\n",
    "    characters = {}\n",
    "    edges = []\n",
    "    edge_attr = []\n",
    "    # Load and process each image\n",
    "    t = 0\n",
    "    for root, dirs, files in tqdm(os.walk(dataset_path)):\n",
    "        for file in sorted(files):\n",
    "            if file.endswith(\".png\") and file.split(\"_\")[-1][:2] in considered_eras:\n",
    "                char_class, era = os.path.basename(root), file.split(\"_\")[-1][:2]\n",
    "                if char_class not in characters:\n",
    "                    characters[char_class] = []\n",
    "                \n",
    "                img_path = os.path.join(root, file)\n",
    "                if extract_features_fn is not None:\n",
    "                    features = extract_features_fn(img_path)\n",
    "                    res = (t, features, era, img_path)\n",
    "                else:\n",
    "                    res = (t, None, era, img_path)\n",
    "\n",
    "                characters[char_class].append(res)\n",
    "                t+=1\n",
    "        if t > 40 and early_stop:\n",
    "            break\n",
    "    return characters\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "8it [00:11,  1.46s/it]\n"
     ]
    }
   ],
   "source": [
    "dataset_path = \"images_background\"\n",
    "data_images_dict = extract_image_data(dataset_path, extract_features_fn=extract_features, early_stop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "char_class_encoder = LabelEncoder()\n",
    "char_class_id_map = char_class_encoder.fit(list(data_images_dict.keys()))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def generate_graph_data_from_dict(data_images_dict):\n",
    "    edges = []\n",
    "    X = []\n",
    "    y = []\n",
    "    for char_class, data in data_images_dict.items():\n",
    "        node_ids = [data[i][0] for i in range(len(data))]\n",
    "        pairs = list(permutations(node_ids, 2))\n",
    "        edges += pairs # Add pairs to the edges within a character class\n",
    "        features = [data[i][1] for i in range(len(data))]\n",
    "        X += features # Append features\n",
    "        eras = [data[i][2] for i in range(len(data))]\n",
    "        #print(eras)\n",
    "        y += eras\n",
    "        #img_path = [data[i][3] for i in range(len(data))]\n",
    "    E = torch.tensor(edges, dtype=torch.long).t().contiguous()\n",
    "    X = torch.tensor(X, dtype=torch.float)\n",
    "    y = torch.tensor([int(label) for label in labels], dtype=torch.long)  # Convert all labels\n",
    "    # Create PyG Data object\n",
    "    data = Data(x=X, edge_index=E, y=y)\n",
    "    return data\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data is valid\n"
     ]
    }
   ],
   "source": [
    "if data.validate(raise_on_error=True):\n",
    "    print(\"Data is valid\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The object Data is\n",
      "torch.Size([192])\n"
     ]
    }
   ],
   "source": [
    "print(\"The object Data is\")\n",
    "print(data.edge_index[0].shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_graph_from_data(data, device = 'cpu'):\n",
    "    device = 'cuda' if torch.cuda.is_available() else 'cpu'  # check if cuda is available to send the model and tensors to the GPU\n",
    "    model = Node2Vec(data.edge_index, embedding_dim=128, walk_length=20,\n",
    "                    context_size=10, walks_per_node=10,\n",
    "                    num_negative_samples=1, p=1, q=1, sparse=True).to(device)\n",
    "    loader = model.loader(batch_size=128, shuffle=True, num_workers=4)  # data loader to speed the train \n",
    "    optimizer = torch.optim.SparseAdam(list(model.parameters()), lr=0.01)  # initzialize the optimizer \n",
    "    return model, loader, optimizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['02', '03', '04', '05']\n",
      "['01', '02', '03', '04', '05']\n",
      "['01', '02', '03', '04', '05']\n",
      "['01', '02', '03', '04', '05', '06']\n",
      "['01', '02', '03', '04', '05']\n",
      "['01', '02', '03', '04', '05', '06']\n",
      "['01', '02', '03', '04', '05', '06']\n",
      "['01', '02', '03', '04', '05', '06']\n",
      "tensor([1, 2, 3, 4, 5, 6])\n"
     ]
    },
    {
     "ename": "ImportError",
     "evalue": "'Node2Vec' requires either the 'pyg-lib' or 'torch-cluster' package",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mImportError\u001b[0m                               Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[38], line 2\u001b[0m\n\u001b[1;32m      1\u001b[0m data \u001b[38;5;241m=\u001b[39m generate_graph_data_from_dict(data_images_dict)\n\u001b[0;32m----> 2\u001b[0m \u001b[43mload_graph_from_data\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdata\u001b[49m\u001b[43m)\u001b[49m\n",
      "Cell \u001b[0;32mIn[37], line 3\u001b[0m, in \u001b[0;36mload_graph_from_data\u001b[0;34m(data, device)\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mload_graph_from_data\u001b[39m(data, device \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mcpu\u001b[39m\u001b[38;5;124m'\u001b[39m):\n\u001b[1;32m      2\u001b[0m     device \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mcuda\u001b[39m\u001b[38;5;124m'\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m torch\u001b[38;5;241m.\u001b[39mcuda\u001b[38;5;241m.\u001b[39mis_available() \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mcpu\u001b[39m\u001b[38;5;124m'\u001b[39m  \u001b[38;5;66;03m# check if cuda is available to send the model and tensors to the GPU\u001b[39;00m\n\u001b[0;32m----> 3\u001b[0m     model \u001b[38;5;241m=\u001b[39m \u001b[43mNode2Vec\u001b[49m\u001b[43m(\u001b[49m\u001b[43mdata\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43medge_index\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43membedding_dim\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m128\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mwalk_length\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m20\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m      4\u001b[0m \u001b[43m                    \u001b[49m\u001b[43mcontext_size\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m10\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mwalks_per_node\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m10\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[1;32m      5\u001b[0m \u001b[43m                    \u001b[49m\u001b[43mnum_negative_samples\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mp\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mq\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43msparse\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m)\u001b[49m\u001b[38;5;241m.\u001b[39mto(device)\n\u001b[1;32m      6\u001b[0m     loader \u001b[38;5;241m=\u001b[39m model\u001b[38;5;241m.\u001b[39mloader(batch_size\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m128\u001b[39m, shuffle\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m, num_workers\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m4\u001b[39m)  \u001b[38;5;66;03m# data loader to speed the train \u001b[39;00m\n\u001b[1;32m      7\u001b[0m     optimizer \u001b[38;5;241m=\u001b[39m torch\u001b[38;5;241m.\u001b[39moptim\u001b[38;5;241m.\u001b[39mSparseAdam(\u001b[38;5;28mlist\u001b[39m(model\u001b[38;5;241m.\u001b[39mparameters()), lr\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m0.01\u001b[39m)  \u001b[38;5;66;03m# initzialize the optimizer \u001b[39;00m\n",
      "File \u001b[0;32m~/miniconda/envs/weber_env_2/lib/python3.8/site-packages/torch_geometric/nn/models/node2vec.py:67\u001b[0m, in \u001b[0;36mNode2Vec.__init__\u001b[0;34m(self, edge_index, embedding_dim, walk_length, context_size, walks_per_node, p, q, num_negative_samples, num_nodes, sparse)\u001b[0m\n\u001b[1;32m     65\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m     66\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m p \u001b[38;5;241m==\u001b[39m \u001b[38;5;241m1.0\u001b[39m \u001b[38;5;129;01mand\u001b[39;00m q \u001b[38;5;241m==\u001b[39m \u001b[38;5;241m1.0\u001b[39m:\n\u001b[0;32m---> 67\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mImportError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__class__\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m     68\u001b[0m                           \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mrequires either the \u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mpyg-lib\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m or \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m     69\u001b[0m                           \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtorch-cluster\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m package\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m     70\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m     71\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mImportError\u001b[39;00m(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__class__\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m     72\u001b[0m                           \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mrequires the \u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtorch-cluster\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m package\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
      "\u001b[0;31mImportError\u001b[0m: 'Node2Vec' requires either the 'pyg-lib' or 'torch-cluster' package"
     ]
    }
   ],
   "source": [
    "data = generate_graph_data_from_dict(data_images_dict)\n",
    "load_graph_from_data(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "43"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.num_nodes\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[t-SNE] Computing 42 nearest neighbors...\n",
      "[t-SNE] Indexed 43 samples in 0.001s...\n",
      "[t-SNE] Computed neighbors for 43 samples in 0.119s...\n",
      "[t-SNE] Computed conditional probabilities for sample 43 / 43\n",
      "[t-SNE] Mean sigma: 8.813629\n",
      "[t-SNE] KL divergence after 250 iterations with early exaggeration: 41.460461\n",
      "[t-SNE] KL divergence after 300 iterations: 0.033091\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1600x1000 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.manifold import TSNE\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Convert string representations of tensors to actual tensors\n",
    "vector_data = np.loadtxt(\"vectors.txt\")\n",
    "label_data = np.loadtxt(\"labels.txt\")\n",
    "\n",
    "# Perform t-SNE dimensionality reduction\n",
    "tsne = TSNE(n_components=2, verbose=1, perplexity=40, n_iter=300)\n",
    "tsne_results = tsne.fit_transform(vector_data)\n",
    "\n",
    "# Plot the results with labels colored\n",
    "plt.figure(figsize=(16,10))\n",
    "scatter = plt.scatter(tsne_results[:,0], tsne_results[:,1], c=label_data, cmap='viridis', alpha=0.5)\n",
    "plt.colorbar(scatter)\n",
    "plt.title('t-SNE projection of the character vectors')\n",
    "plt.xlabel('Component 1')\n",
    "plt.ylabel('Component 2')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "weber_env_2",
   "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": "3.8.19"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}