visualize_embedding.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import os\n",
    "import torch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "action_to_desc = {\n",
    "        \"bend and pull full\" : 0,\n",
    "        \"countermovement jump\" : 1,\n",
    "        \"left countermovement jump\" : 2,\n",
    "        \"left lunge and twist\" : 3,\n",
    "        \"left lunge and twist full\" : 4,\n",
    "        \"right countermovement jump\" : 5,\n",
    "        \"right lunge and twist\" : 6,\n",
    "        \"right lunge and twist full\" : 7,\n",
    "        \"right single leg squat\" : 8,\n",
    "        \"squat\" : 9,\n",
    "        \"bend and pull\" : 10,\n",
    "        \"left single leg squat\" : 11,\n",
    "        \"push up\" : 12\n",
    "    }\n",
    "\n",
    "def load_train_embeddings(directory='embeddings'):\n",
    "    # directory = os.path.join(args.out_dir, 'embeddings')\n",
    "        \n",
    "    embedding_dict = {}\n",
    "    \n",
    "    for filename in os.listdir('embeddings'):\n",
    "        if filename.endswith(\".npy\"):\n",
    "            key = filename.split('.')[0]\n",
    "            embedding = np.load('embeddings/'+filename)\n",
    "            if len(embedding)==0:\n",
    "                continue\n",
    "            \n",
    "            embedding_dict[action_to_desc[key]] = embedding\n",
    "    \n",
    "    return embedding_dict"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys([9])\n",
      "Completing loading training embeddings:\n",
      "9 (71, 512, 49)\n"
     ]
    }
   ],
   "source": [
    "embedding_dict = load_train_embeddings()\n",
    "print(embedding_dict.keys())\n",
    "print(\"Completing loading training embeddings:\")\n",
    "for k,v in embedding_dict.items():\n",
    "    print(k,v.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "embedding = embedding_dict[9]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(71, 512, 49)\n",
      "torch.Size([71, 512, 49]) torch.float32\n"
     ]
    }
   ],
   "source": [
    "print(embedding.shape)\n",
    "embedding = torch.tensor(embedding)\n",
    "print(embedding.shape, embedding.dtype)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_1448809/3286874197.py:12: UserWarning: No data for colormapping provided via 'c'. Parameters 'cmap' will be ignored\n",
      "  plt.scatter(embedding_2d[:, 0], embedding_2d[:, 1], cmap='viridis', s=50, label=\"Embeddings\")\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x1000 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "embedding_flattened = embedding.view(71, -1).detach().cpu().numpy()\n",
    "\n",
    "# Apply PCA to reduce dimensions to 2D\n",
    "pca = PCA(n_components=2)\n",
    "embedding_2d = pca.fit_transform(embedding_flattened)\n",
    "\n",
    "# Calculate the mean in the 2D PCA space\n",
    "mean_embedding_2d = embedding_2d.mean(axis=0)\n",
    "\n",
    "# Plot the embeddings in 2D\n",
    "plt.figure(figsize=(10, 10))\n",
    "plt.scatter(embedding_2d[:, 0], embedding_2d[:, 1], cmap='viridis', s=50, label=\"Embeddings\")\n",
    "\n",
    "# Label each point with its index\n",
    "for i in range(embedding_2d.shape[0]):\n",
    "    plt.text(embedding_2d[i, 0], embedding_2d[i, 1], str(i), fontsize=9, ha='right')\n",
    "\n",
    "# Plot the mean point\n",
    "plt.scatter(mean_embedding_2d[0], mean_embedding_2d[1], color='red', s=100, marker='x', label=\"Mean\")\n",
    "\n",
    "# Add title and axis labels\n",
    "plt.title('Embeddings in 2D (PCA)')\n",
    "plt.xlabel('PCA Component 1')\n",
    "plt.ylabel('PCA Component 2')\n",
    "\n",
    "# Add legend\n",
    "plt.legend()\n",
    "\n",
    "# Show plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}